0:00
[music] I see a large IPO on the horizon.
Search this video
0:09
You're surrounded by journalists. Hold your position.
0:13
[music] Overnight [music] success. The place 3, >> right? There's misinformation.
1:00
According clearing order inbound, >> let's just roll.
1:09
We are surrounded by general. Report your position. Come on. Get up. Trust the experts here. [music] We are experts. Founder VC.
1:42
[music] I see multiple journalists on the horizon. Stand by.
1:51
>> [music] >> UAV online. Blaze. Double blaze. Triple blaze. Double kill. Finally, it's wins. Team deathmatch. We are experts. [music] Triple blades. Let's [music] just roll. Right. Clearing order inbound.
3:13
>> [music] >> You're surrounded by journalists. Hold your position. >> Strike one. >> Strike two.
3:39
Activate golden [music] retriever mode.
3:57
[music] >> Mark clearing order inbound. >> 52.
4:19
I see more and more [music] journalists on the horizon. Stand by. TV.
4:36
Today is Thursday, February 12th. We are live from the TV.
4:38
Ultra, the temple of technology, the fortress of finance, the capital of capital, ramp. com, baby. Time is money. Save both.
4:44
He's used corporate cards, bill pay accounting, and a whole lot more all in one place.
4:48
We have a great show for you today, folks.
4:50
We're taking a little tour of the health world.
4:52
We got Brian Johnson and Andrew Huberman on the show.
4:54
Of course, we also have Matthew Zitelland, June Park, David Richer from FIT from Lyft is coming in.
5:01
Todd McKinnon [clears throat] from Octa.
5:02
Uh we have a number of other folks joining the uh the lightning round. It should be a fun show.
5:08
But first, I have to tell you what linear is.
5:12
Not only do they sponsor the lineup, but they are the system for modern software development.
5:16
70% of enterprise workspaces on linear are using agents.
5:20
[applause] So something big happened yesterday.
5:24
There was a massive viral article and do you know how big the something big is happening? How much that grew? >> It went big.
5:33
I was getting texts about it.
5:35
>> It has over a 100,000 likes now.
5:35
We were looking at it was like 50. How many views?
5:40
>> I think it's over a hundred million now.
5:42
It's uh quickly becoming the most read thing in the world. I don't know. Uh it's it's crazy.
5:49
75 million >> 75 million views >> views. >> Okay.
5:53
And 100 >> They're saying that only 200 people actually read it.
5:56
[laughter] >> But everyone has an opinion. >> Everyone has opinion. I had an opinion.
6:01
I wrote a piece about it yesterday uh comparing saying AI is not like CO.
6:03
I kind of got one shot by the by the co analogy just from a mathematical perspective.
6:10
Spent a lot of time talking about the difference between exponential growth and and sigmoidal curves and logistics curves.
6:15
Um but um I uh I I you know I enjoyed it.
6:19
I enjoyed talking to the author on the show and Will Manitis had a great rebuttal as well talking about uh these are toolshaped what what are they tools? >> Toolshaped objects. >> Tool-shaped objects.
6:30
Um just the idea of like getting your workflow set up.
6:33
It can be a little bit of a nerd snipe sometimes.
6:37
Um there's [snorts] a whole bunch of news.
6:39
Uh but the best reaction for to something big is happening.
6:42
It's got to be John Palmer.
6:45
Something small is happening.
6:47
I was reading this this morning actually laughing out loud.
6:48
So, uh, we'll read through it.
6:51
John Palmer says, >> "Yeah, you were you were reading it to yourself and then you just started belly laughing >> and so then I started filming and then I asked you reading I was filming you.
7:02
Then I sent that to John." >> Oh, I missed that.
7:04
>> To get his reaction, you reacting to his. >> Yeah.
7:07
Before we read through this, let me tell you about Reream.
7:08
One live stream 30 plus destinations.
7:10
If you want to multiream, go to reream. com.
7:11
So, think back to 2014, a little over a decade ago.
7:17
If you wanted a burrito, you had to get in your car and drive to the restaurant.
7:21
You had to stand in line.
7:23
You had to talk to a person.
7:23
If someone told you that one day a stranger would bring you a burrito to your front door because you pressed a button on your phone, you would have said, "That sounds dystopian and also incredible."
7:35
Then, over the course of about three years, it just became normal.
7:41
And this was before Chipotle started skimping out on the portions, too.
7:46
So, it was really, really good.
7:48
I think we're at that same inflection point right now, except of instead of burritos, it's your entire job.
7:56
I've spent the last two weeks reading about AI and watching clips of people building things with it.
8:02
I live in this world now.
8:04
The future is being shaped by a remarkably small number of people, a few hundred researchers at a handful of companies.
8:11
I'm not one of these people, says John Palmer. I work in crypto.
8:15
[laughter] But I am close enough to feel the ground shake.
8:17
I follow several AI researchers on X and I also recently purchased a Mac Mini.
8:22
I haven't set it up yet, but I think you can understand what I'm saying.
8:25
[laughter] I know this is real because it happened to other people first.
8:28
Here's the thing nobody outside of tech understands yet.
8:33
The people sounding the alarm aren't making predictions.
8:35
They're telling you what already happened to them.
8:37
I am relaying this information to you secondhand with conviction for years.
8:40
AI had been improving steadily.
8:44
I absorbed these improvements fine because I wasn't really paying attention to be honest.
8:49
Then apparently in 2025 everything got much faster and then faster again.
8:52
I know this because I watched a three-hour podcast about it last Wednesday.
8:56
The host was visibly shaking.
8:58
I could tell because he wasn't even able to finish a single pint of Guinness [laughter] taking shots.
9:06
Then on February 5th, two major AI labs released new models on the same day and something clicked. Not for me personally.
9:12
I was at a birthday dinner for my wife's coworker.
9:13
But for the people who tried the models that day, it was apparently a huge deal.
9:18
One guy said he tells his AI what to build, walks away for a few hours, and comes back to find the work done.
9:26
This will very likely be what my life is like as well once I set up my [laughter] AI will be doing things while I'm in the other room. That's potentially huge. Huge.
9:38
Let me give you an example so you can understand what this looks like in practice.
9:42
A guy on Twitter built an entire app just by describing it.
9:44
The AI wrote the code, opened the app, tested the buttons, decided it didn't like the layout, fixed it, and only then said, "It's ready. I'm gonna do something.
9:53
[laughter] I'm gonna do something like that with my Mac Mini.
9:56
Some type of cool app or something.
9:59
I'm still figuring it out, but I have all the hardware [laughter] and I'm definitely going to optimize my setup with all types of skills and plugins. I'm not exaggerating.
10:06
That's what my Tuesday could look like potentially.
10:11
[laughter] >> But I tried AI and it wasn't that good. People would say this.
10:15
They say this, I tried AI, it wasn't that good.
10:17
Here's John Palmer's response.
10:18
He says, "I hear this constantly.
10:21
I understand it because I also thought this, but the models today are apparently unrecognizable from what existed six months ago.
10:28
Some of the people who are saying this have hundreds of thousands of followers and make sure you're not on the free plan.
10:35
You have to get on the paid plan."
10:36
[laughter] >> That's key.
10:39
>> How fast is this actually moving?
10:39
Let's answer that question.
10:41
[snorts] In 2022, AI couldn't do basic math.
10:44
By 2023, it could pass the bar exam.
10:47
by 2024 could write working software.
10:50
And by 2025, some of the best engineers in the world had handed over most of their coding work to AI.
10:55
In February 2026, you can make AI send you a morning summary of the top posts on Reddit from your Mac Mini. Here we go.
11:03
What should you actually do?
11:05
I'm not writing this to scare you, says John Palmer.
11:08
I'm writing this because the single biggest advantage you can have right now is being early.
11:13
Early to understand it, early to buy the hardware.
11:14
early to subscribe to the paid tier and ask it to make a meal.
11:18
You can start using it for real work.
11:20
If you're an engineer, give it a >> meal. A meal plan. >> A meal plan. Oh, a meal plan.
11:24
Yeah, [laughter] actually make a meal. That would be cool.
11:28
Uh if you're an engineer, give it a GitHub repo.
11:30
If you're in finance, give it a messy spreadsheet.
11:32
If you're in other industries, definitely figure out what you can give it, then give it, [laughter] >> but give it something for sure.
11:39
I just need to think about it more.
11:41
Mention it to your co-workers.
11:43
Right now there's a brief window where most people are still ignoring this.
11:46
You can be the person ignoring it less which puts you ahead.
11:51
Be the person who walks into a meeting and says, "I used AI to do this analysis in an hour instead of three days."
11:54
It's fine if it actually took you three days.
11:58
The point is people will respect you. Get a Mac Mini.
12:00
This is point number three.
12:02
I think this >> And by the way, you can actually get a Mac Mini on Door Dash.
12:05
>> You can just like a burrito. >> It's important.
12:08
>> Every single account I've seen posting this stuff has a Mac Mini.
12:10
It's the ultimate tool for this stuff.
12:12
You got to get your financial house in order.
12:14
I'm not a financial adviser, says John Palmer.
12:17
That said, I did just spend $599 on a Mac Mini plus Apple Care and a 50 and $50 on Cloud tokens.
12:22
But if it gets too expensive, have a backup plan to switch to an open source Chinese model.
12:29
Fourth, watch the podcasts.
12:29
Actually, don't just watch.
12:31
Study TBPN, Door Cash, etc.
12:34
The good thing is that these shows are several hours long, so you can basically fill up your entire day just watching them >> or making them >> or making them.
12:42
In our case, uh, rethink what you're telling your kids.
12:44
The standard playbook, good grades, good college, stable job, it's over.
12:48
Tell them it's [laughter] tell them it's likely not looking good for them.
12:56
Go to your four-year-old and tell them it's not looking good, buddy.
12:59
>> It's not looking good.
13:00
>> It's not looking good.
13:01
>> What I know I know this isn't a fad.
13:01
the technology works and the richest institutions in history are pouring trillions into it.
13:06
I know this because I've seen it mentioned pretty frequently over the la past few [laughter] months.
13:12
I know I know the people who will come out of this best are the ones who start engaging now, not with fear, but with curiosity, a sense of urgency, and ideally a Mac Mini.
13:20
[laughter] >> The future is already here.
13:22
It just hasn't knocked on your door yet. It's about to.
13:25
And when it does, I will be ready.
13:27
My [laughter] Mac Mini will be unboxed.
13:28
My agent will be configured.
13:28
I will describe what I want in plain English and it will appear.
13:34
[laughter] I just have to optimize everything first and make sure my setup is super legit.
13:39
If this resonated with you, share it with someone in your life who should be thinking about this, too.
13:43
Most people won't want to hear [laughter] until it's too late.
13:45
You can be the reason someone you care about buys a Mac Mini. So funny. >> What a banger. >> Such a banger. So, so good.
13:52
I really I really >> No words. >> No words. I did. Let's see.
13:56
Let's see how uh how Codeex is doing.
13:58
I gave I I went and I described in plain English to codeex 5. 3 Medium. Build me an app. An amazing app. The best app ever.
14:07
One that will win awards and make me a legend in Silicon Valley.
14:11
[laughter] >> An award-winning app. >> An award-winning app.
14:14
And it appears to be working. Let's see.
14:16
We'll we'll we'll we'll check in on this.
14:18
I need to I need to run it. No, it's running it. I don't know. It's describing it.
14:23
doesn't hasn't said what it's thinking of doing, but it says >> you don't need to know that.
14:28
You don't need to know that.
14:31
>> U more seriously, John John Palmer's post feels like a response to Will's essay, Tool-shaped Objects, which is fantastic.
14:38
We won't read through Will's whole essay because I think you should read it yourself.
14:44
>> He might be coming on the show soon.
14:45
>> Yeah, we're going to get him on the show tomorrow. >> Yeah.
14:47
>> Uh which we're excited about.
14:49
>> But we do love reading a Will Manitis essay laced with tons and tons of advertising.
14:53
We know that that's one of his favorite things and you know we'll read one right now for the New York Stock Exchange.
14:59
Want to change the world?
15:00
Raise capital at the New York Stock Exchange. Just do it. [applause] >> Mropic.
15:07
>> Entropic is nearing the completion of a deal to raise more than 20 billion in a funding round led by investors including >> Peter Teal's Founders Fund, Dehaw, and Dragon. Let's dig in.
15:17
Uh, of course being a little bit silly with the pronunciation.
15:23
Sydney over on X said, "British AI researcher be like anthropic."
15:30
[laughter] >> So, I'm going to start uh integrating that into into my sentences as well.
15:32
Um, >> Enthropic is nearing the completion of a deal to raise more than 20 billion in a funding round led by the group I just mentioned.
15:42
The deal is set to value Anthropic at about 350 billion, nearly doubling its prior value, and could be announced as early as this week.
15:48
The funding round includes a range of investors including CO2, Singapore's GIC, Microsoft, Nvidia, and and be has become a who's who of Silicon Valley and Wall Street investors.
16:01
Uh, I think it's worth kind of talking about this just from a PT lens. Yeah.
16:09
>> I don't know if you want to break it down given that uh former boss. >> Yes.
16:14
Uh so Founders Fund is known for the monopoly thesis like there's power law.
16:20
You want to be in the best company in the category.
16:22
You want the best ownership.
16:25
>> Put all your money in that one. >> Yeah. I'll I'll go all in. Famously.
16:26
Um they they they did some other social networking deals but obviously like Facebook was the big one at seed and that was enough to return that fund many times over.
16:38
Um and so the whole thesis of the fund for a long time has been like this concentrated bets concentrated bets.
16:45
Um AI has been an interesting one.
16:47
They're now in three major labs because they have a huge position in SpaceX which now owns XAI and so they're in that lab um sort of coincidentally.
16:59
Peter Teal I believe is listed on Wikipedia as a co-founder of OpenAI or at least he was like an initial donor and got like a co-founder title.
17:05
um and so has been involved in OpenAI for a long time and then they invested I believe in like the $60 billion round right after right right when Chat GPT was launching there was a big funding round that happened and it was an odd deal because the company had not yet converted.
17:23
It was very unclear what Microsoft would get, what the employees would get, what the investors would get, what the nonprofit would get because the nonprofit was going to keep a stake.
17:31
And so it was this like wild bet and I know a lot of investors. It was a leap of faith.
17:37
Um, a lot of investors were scared off because there was, you know, what am I investing in?
17:42
Like, I'm not getting shares, I'm getting units.
17:44
And it was like a very odd structure.
17:46
And I know many investors that were scared off by that.
17:50
They were like, yeah, the product's amazing. >> I'm kind of a unit. >> An absolute unit.
17:56
>> I'm an absolute unit, but I'm used to buying shares. >> Yeah. Exactly.
17:59
So, you had to do a lot of sort of leaps of faith.
18:02
At the time I was sort of trying to I mean I was enjoying Chachi PT.
18:09
I thought it was an important product.
18:11
I didn't really have fully uh full thesis around like the aggregation theory and you know maybe there would be a winner take all in consumer AI any of that.
18:21
consumer AI any of that. I was just looking at the team because at the team at at the time it was Sam Alman who'd founded and sold a company in this like aqua hire then worked at YC been a successful investor and it was just like all green flags for like so if it wasn't
18:39
all the baggage of like open AI and stuff just like >> if if someone with that resume came to you and was like I'm starting a company you'd just be like yeah you check all the boxes great and then you have Greg Brockman who was the CTO of Stripe huge hugely successful company. And so I was
18:51
And so I was like, wait, you got that guy to come be your co-founder? Okay, that's great.
18:57
Then you have Ilia and you had a whole bunch of other folks who were just really really top tier just in terms of like founder quality like worth backing.
19:05
And so you did have to put the valuation out of your mind because it was so not so disconnected from like startup valuation. Yeah.
19:12
But in terms of just is the team high quality, it seemed like it checked all the boxes and that they would figure it out.
19:18
Uh they of course ultimately did.
19:20
Um there was another there was another um sort of uh interesting thing just about you know like the thesis at the time. How big can this be? What is the market?
19:31
What are the opportunities for really really huge categories?
19:37
A lot of founders fund investments like SpaceX, it's like this massive category that's very very hard, but if you crack it, it's going to be an entirely new market.
19:46
Social media was certainly that as well.
19:47
Um and so uh that made a lot of sense.
19:50
Obviously they're in X now they're in anthropic.
19:53
Sequoa is also in all three.
19:53
And the interesting thing is like PT has given a number of talks where he's sort of been anti- AI.
20:01
You've you've seen this whole thing about like crypto is libertarian, AI is communist and that AI is like a consolidating force.
20:07
He's he's never been one to say like oh wow the tools are so amazing.
20:12
It's been a lot of like is this just doing your homework for you?
20:16
He's not putting out threads about using his Mac Mini. >> I don't even know.
20:20
I don't even know if he has a Mac.
20:22
>> You should send [laughter] You should send John Palmer's piece >> to PT. >> Mhm.
20:26
>> Just flip it over there.
20:26
Say, "Hey, I think you should read.
20:29
>> I think you should read this.
20:29
>> I don't you know you kind of understand this.
20:31
You did invest in [laughter] Deep Mind, you know, more than more than a decade ago." >> Yeah.
20:35
The the Deep Mind deal was was was wild.
20:37
So, um, Demis met, uh, PT at a party, I think, and instead of pitching him on Deep Mind, he said, "Let's play chess."
20:47
And so, they they played chess and then Founders Fund invested in DeepMind very early in >> Do we know who won?
20:55
>> I don't know actually. That might be Deep Lore. I'm not sure.
20:57
Uh, but DeepMind the the initial pitch was like this counter force to Google a little bit or like this like renegade group of AI.
21:07
It's it's the same thesis over and over again.
21:09
It's like let let's get the pure play going and then it always gets tied in with a hyperscaler at some point.
21:15
But um Google acquired >> beat them join them >> deep mind in 2014 January for 600 million and uh Founders Fund owned more shares than all than the than the co-founders who had split up and then um and then all the employees because because they had again developed like a very concentrated position. >> Okay.
21:33
Apparently, here's some of the lore.
21:34
Knowing Teal was a chess player, Demis struck up a deep technical conversation about the relative strengths of the bishop versus the knight.
21:43
>> So, they were just talking about chess.
21:44
They didn't actually play. H >> maybe.
21:47
>> Really quickly, let me tell you about Sentry.
21:49
Century shows developers what's broken.
21:50
It helps you helps them fix it fast.
21:52
That's why 150,000 organizations use it to keep their apps working. >> Yeah.
21:56
Apparently, the first investment was 2 and a half mil.
21:58
[sighs and gasps] >> You can't get deals like that anymore.
22:03
The Anything with 2 and a2 millm or a nightmare nightmare check size.
22:08
>> Wasn't the initial Facebook check was it [snorts] $100,000 or was it a million dollars?
22:12
It was like not a lot for 10% of Facebook or something like that. It was crazy.
22:16
Uh yeah, wild wild times.
22:16
But uh so yeah, I mean reading into this, does this say that the founders fund team thinks that there's like a a divergence in the market that there's actually two there are actually two markets going on here?
22:28
One in consumer AI, advertising, knowledge retrieval.
22:32
the other in uh code and um and developer tooling and and codegen and that anthropic has actually carved out a separate space so they're less competitive than people think.
22:42
Maybe they're just more competitive and it just makes sense to to get into both.
22:45
Um we'll have to talk about them once the deal closes and everyone's ready to chat.
22:51
Uh Tyler, do you have something to add about uh >> I was going to say didn't um Scott Woo ran back that playbook, right?
22:56
Because he he played didn't he challenge PT and chess and then Napoleon and poker?
23:01
Wasn't that the whole thing? >> Yeah. Yeah.
23:02
And it was like if I win, I get my terms.
23:04
If you win, you get your terms.
23:06
>> And we have some breaking news.
23:07
>> What's the breaking news?
23:08
>> Anthropic has formally announced the round. >> Let's go.
23:11
>> They said we've raised 30 billion. So they upsized it.
23:15
>> 20 was uh they are now at a $14 billion run rate. >> Wow.
23:23
>> With this figure growing over 10% annually in each of those past 3 years.
23:28
We'll see if they got another 10x in them.
23:30
Dylan Patel on the show last week felt uh fairly confident. >> Yeah. Yeah. Uh I mean, yeah.
23:37
The the the the big shift is that Yeah.
23:42
the big shift is that Yeah. I mean the the the biggest update for me is that the uh the forward deployed engineer thing is like very real and uh and even if the tools can build themselves, there's still so many blockers to actually uh rolling these out that you're going to need a lot of people
24:00
inside these organizations who are actually playing with Mac minis on the weekend and getting obsessed with these tools and then bring them into the enterprise because it's very easy for a lot of people in Fortune 500 companies to just show up and do their job the way they always do it because no one gets fired for switching things up. But if
24:16
But if you roll out some tool and it doesn't work the way you intended or it sucks about bunch of your time even if you're just I have eight hours of work to do in my traditional system where do you get the extra hours to to automate your work uh it takes time and that's where the four deployed engineers come in.
24:32
So great gig for anyone who wants to go implement enterprise agentic workflows because that will be the theme.
24:39
Let me tell you about Phantom Cash.
24:41
Fund your wallet without exchanges or middlemen and spend with the Phantom card.
24:46
>> Let's pull up this chart.
24:46
Andrew Kern over on X pulled it out.
24:50
We can see Anthropics run rate revenue growth.
24:52
See here, >> quite the pop from 0 in 2023 to 100 million in Jan 2024 to a clean 1 billion in January 2025 and now sitting at 14 billion. >> This is acceleration.
25:09
This is accel This is true. Are you feeling it yet? >> True acceleration. >> Feeling it.
25:15
>> Daario sort of underplayed 2027, I guess.
25:18
Or I guess he said that he he was like, I don't think we have another 10x if they go to 100 billion.
25:24
>> Well, I he said like we've been 10xing. Yeah.
25:26
You know, we'll see if we do it again. Yeah. Like wink wink.
25:30
>> But >> if they if they hit 140 billion at run rate by the end of the year, that would be insane. But strange times. Strange times. You love this effect. >> That's all, folks.
25:43
>> I mean, Tyler Tyler's been saying stuff along this lines uh for some time.
25:45
So, >> Octa Octa helps you assign every AI agent a trusted identity.
25:50
So, you get the power of AI without the risk.
25:52
Secure every agent, secure any agent.
25:54
You're the CEO of Octa coming on the show. >> Pumped for that.
26:00
>> Um, back to the timeline.
26:00
There are there are more uh there are more >> more inthropic.
26:04
Nathan Lambert uh was quoting their announcement from yesterday saying, "We're committing to covering electricity price increases from our data centers." This is anthropic. >> Oh, yes.
26:14
>> To ensure rate payers aren't picking up the tab, will pay 100% of grid upgrade costs.
26:19
Work to bring new power online >> and invest in systems to reduce grid strain. >> Makes so much sense. >> Yeah.
26:25
Nathan saying that maybe this should have been their Super Bowl ad. Um yeah. Yeah.
26:28
It felt like when you looked at when you kind of look at clearly how much energy the industry put into the Super Bowl. >> Yeah.
26:38
>> And what the net effect was for the average American.
26:41
I don't think any I don't think the average American came away from that like having more positive feelings towards AI. >> Yeah.
26:51
>> If anything, they were like, [snorts] I'm sick of this stuff. >> Yeah. Yeah. Yeah.
26:53
The meme was like, oh, a beer ad's like refreshing. >> Yeah.
26:56
Um yeah, it would have been good to focus on this.
26:59
Microsoft uh did this what a couple months ago the news was uh on January 14th.
27:04
So just one month ago, Microsoft announced that they were going to do this uh because there was a 267% wholesale re electricity price near data centers and local communities were starting to push back.
27:19
Uh some 64 billion in projects face delays or cancellation because of opposition.
27:23
So, it's pretty easy to say, "Hey, my uh my my electricity bill is going up.
27:29
I don't want a data center in my county, in my city, and you call your city county representative, and they say, "Yeah, I don't like the slop either.
27:38
I I I see I see a bunch of junk that doesn't seem that valuable. Why do we need it here?
27:44
It's not going to create that many jobs.
27:45
It's not going to be that good for taxes, blah blah blah blah blah."
27:46
Um, and uh other companies that do this, too.
27:50
Uh Google had their clean transition tariff and Amazon also pays a surplus above electricity costs already.
27:58
Um but Microsoft was unique because they they timed it and did a whole press cycle around their initiative and the volume was uh was much much higher.
28:06
Um the the other the other interesting fact from that previous report was uh the anti-AI issue is remarkably bipartisan.
28:18
Data Center Watch claims that 55% of Republicans and 45% uh Democrat in affected districts. So 55 45.
28:27
So it's not really like a red a right issue or a left issue.
28:29
Like there's push back on both sides.
28:31
Like when the data center comes in, it's not just oh it's it's it's all green flags in a Republican district.
28:36
No, both are pushing back.
28:39
So um >> uh [clears throat] they're also going in on uh on super PACs, right?
28:44
Uh, so Anthropic is putting $20 million into a super PAC operation that is meant as a direct counter to the Open AI super PAPAC operation.
28:51
The midterms are a battleground between these two rival AI labs, says Teddy Scheffer at the New York Times.
28:57
And so we're going to see some attack ads.
28:58
It's like Sam and Daario are running for election and we're going to see some >> the president of AI. >> Yeah.
29:06
This person can't can't restore the economy.
29:08
What do you think, Tyler?
29:09
>> Yeah, but I mean we just said that like the overwhelmingly on both sides there's like a lot of push back, right?
29:13
So it's like what what are the actual differences between their their approaches?
29:16
Yeah, that is interesting.
29:18
>> Maybe I guess maybe Anthro will be more like about the the China chip issue.
29:23
>> Yeah, it feels like there they should be more in alliance than against each other.
29:27
>> Yeah, it's like crypto there's a crypto super PAC.
29:29
There's like the big one.
29:31
There's no like big AI one right now.
29:34
Yeah, it it it would be weird if there was like a USDC super PAC and then a USDT super PAC and they were like fighting each other and just constantly surfacing bodies that are buried in the closets or skeletons in the closet because every time there's a successful attack on OpenAI. Yes.
29:48
People might be like, "Oh, okay.
29:51
I like Anthropic a little bit more."
29:53
But most most people just come away being like, I don't like AI. >> Yeah.
29:56
>> So, if they're attacking each other, that could just be bad for both, maybe. I don't know. But we'll see. We'll see what they run. We'll see.
30:02
We'll see what they >> Ben over on X says, "If you were falsely accused of a crime and it went to trial, who would you prefer to listen to the arguments and give the verdict?"
30:10
And the options are a jury of your peers or Claude Opus 4. 6.
30:18
>> There's been 2,000 votes and 63% of people prefer Claude.
30:23
>> Yes, might be the audience here. >> Yeah. [laughter] Yeah. A little little biased.
30:28
Ben does work at Orchid Health uh engineering whole genome embryo screening.
30:33
So I wouldn't be surprised if that's his audience. But uh I don't know. I don't know.
30:38
I think the jury of your peers will be around for a very very long time.
30:42
I think that that will be something.
30:44
I mean they still have courtroom synographers and we have >> I think it's going to be around a long time too. >> Yeah. Okay.
30:51
If you're thinking in centuries maybe but uh I mean the courtroom sketch artist.
30:55
We've had cameras for a hundred years.
30:57
Like the jury I'm not saying that I think this will ever happen. >> Yeah. Yeah. >> Okay. But you'd prefer it.
31:01
You you would ideally I mean how many jurors will be sitting there being like, "Okay, Claude, how should I vote?
31:09
I haven't really been paying attention."
31:10
Like I've been playing Candy Crush.
31:12
I've been playing I've been vibe coding my my optimal workflow on Cloudbot.
31:19
>> Uh really quickly, Finn.
31:19
AI the number one AI agent for customer service.
31:21
If you want AI to handle your customer support, go to finn. ai.
31:25
support, go to finn.ai. AI >> uh will open AI or anthropic IPO first as of late last year it was sitting at 75% of people on call she thought that uh it would be open AI now it's flipped almost 70% of people think that anthropic will get out the door >> it is narrowing it is narrowing
31:46
>> but I expect these to kind of fluctuate just with the headlines >> it's all it's all red meat for Michael Grimes I I got so many text messages about about his move back and being like it is the most bullish thing for latestage growth technologies that he's back at Morgan Stanley. Like you just like if
32:03
Like you just like if you had any doubt that there will be major major blockbuster IPOs, it's Michael Grimes being back at Morgan Stanley.
32:11
So >> very excited for that.
32:12
Really quickly, Gusto, the unified platform for payroll benefits and HR built to evolve with modern small and mediumsiz businesses.
32:19
>> Taha over on X says, "I turned on Meta AI's auto reply for Facebook Marketplace.
32:23
I listed an item for $75 and when someone asked if it was available, the AI replied asking if they wanted it for free. >> It's elite.
32:32
It's a It's all It's all just a bait tactic. So, um, hold on. >> Community note.
32:38
>> Yeah, >> we got to dig in.
32:39
It says Meta AI only suggest one reply.
32:41
So, this is a typo in here.
32:44
Only suggests one reply for every message of the listing.
32:46
Then you can say, edit or save it.
32:47
This one is manually written by the user.
32:49
So, >> some fake news, but still funny.
32:51
Yeah, I mean Taha's responding today.
32:53
I learned do not trust community notes. This happened to me.
32:57
No idea why the community note appeared.
32:58
Meta AI only sends one message.
33:00
I cannot edit it, but I can unend it. >> Mars catalog.
33:05
Stop talking trash about it publicly and it might stop giving away your stuff.
33:09
[laughter] That's on you, buddy. That's on you. >> Rockos basilisk.
33:15
>> This is Rockco's Basilisk.
33:17
>> Uh, what's up with Warner Brothers? Where >> console. com first?
33:19
Console builds AI agents that automate 70% of IT, HR, and finance support, giving employees instant resolution for access requests and >> the situation intensified this week as Paramount CEO David Ellison and a vocal investor made new moves to thwart rival Netflix's planned takeover.
33:37
>> The storied Hollywood asset and the home of Batman, Harry Potter, and the White Lotus.
33:41
Uh, Paramount, the long rebuffed suitor for Warner Brothers Discovery, enhanced its 77.
33:46
9 billion allcash offer for the entire company Tuesday in an effort to bring Warner to the negotiating table and ultimately abandoned its deal with Netflix, also an allcash transaction valued at 75 billion.
33:59
Uh, Cleveland investor and Kora Holdings entered the frey by acquiring a small stake in Warner with an eye toward increasing its holdings and pressuring the company to negotiate with Paramount.
34:09
I wonder if they're friends with the Ellison. >> Interesting. >> Paramount's Gamble.
34:13
Paramount is hoping its latest pitch can persuade Warner to ditch its agreement and sell its movie and TV stu studios and HBO Max streaming service to Netflix.
34:21
Unlike Netflix, Paramount also intends to buy Warner Discovery's cable networks, which includes CNN, TBS, and Food Network. Mhm.
34:30
>> But Warner has so far shown little interest in engaging with Paramount, telling investors that Ellison's previous offer was not even comparable to the Netflix agreement.
34:39
>> On Tuesday, Warner said its board will review Paramount's new offer, but it but isn't modifying its recommendation regarding its agreement with Netflix.
34:47
The company advised shareholders to not take any action now regarding Paramount's amended tender offer.
34:50
The bar for Warner to reopen talks with Paramount is very high given its contract with Netflix.
34:57
>> A massive termination fee. said it would pay the 2. 8 billion.
34:59
Let's give it up for multi-billion dollar termination fees. >> Love those.
35:06
>> Uh Warner would owe Netflix should the agreed to deal collapse and pay a ticking fee of 25 cents a share to Warner shareholders for each quarter its deal hasn't closed. >> Whoa. >> We got ticking fees. >> Ticking time bomb.
35:19
>> Some stakeholders are holding out for Paramount to further boost its bid.
35:20
I'm not surprised given that they've repeatedly said with their offers, "This is not our best and final offer." >> Yeah.
35:30
>> Analysts from Raymond James wrote in a note to clients that many Warner Discovery shareholders still expect Paramount and its backers to raise its $30 a share bid by $2 to $3 a share. >> Yeah.
35:40
>> So, everyone is waiting around saying, "Let's get those numbers up."
35:44
>> And remind me of how Paramount is doing right now. They have UFC, correct?
35:48
And the experience watching UFC was a downgrade from pay-per-view but cheaper. What was the takeaway?
36:00
>> Well, I mean I I don't think we can like give an analysis of Paramount as an entire business based on the pro the the experience was not as good as paying for the pay-per-view.
36:11
>> And was that because you had to you had to go through all the you had to jump through all the hoops to sign up for Paramount Plus and you weren't already a member? >> No.
36:19
the the like the viewing experience was different because it's now subscription and ad supported whereas previously pay-per-views had ads but not nearly as much.
36:28
So the thing you're missing now watching is like >> in round commentary walkouts things that like hardcore fans >> really like.
36:38
>> Is there less Joe Rogan commentary?
36:40
>> Basically >> brutal >> brutal.
36:41
This is extremely >> well the markets are reacting.
36:44
Paramount's down 7% today.
36:44
That might be the good day of course not just this analysis of the >> UFC viewing experience. >> Yes.
36:55
So, uh >> but no, the uh the the stock dip was apparently in reaction to the >> breakup fee people being like, "Oh, great.
37:03
You're going to pay even more." >> Yeah.
37:05
Let me tell you about App Love and Profitable Advertising made easy with Axon. ai.
37:07
Get access to over 1 billion daily active users and grow your business today.
37:11
So, for now, Paramount's move Tuesday to sweeten its offer could push Netflix to increase its bid for Warner.
37:17
Uh, Paramount needs to test Netflix's pain threshold without having to bid too high.
37:23
The revised offer does exactly that without further constraining its balance sheet.
37:27
For now, the ball is now with Netflix.
37:29
Warner and Netflix are focusing on getting their deal approved by the Justice Department as well as regulatory authorities in Europe.
37:36
As part of its investigation, the DOJ is looking into whether Netflix has engaged in anti-competitive practices.
37:43
Last week, President Trump reversed course and said in an interview with NBC News that he didn't think he should weigh in on the deal and would instead leave it to the Justice Department.
37:53
I've decided I shouldn't be involved.
37:54
The Justice Department will handle it.
37:56
Trump said in the interview he previously expressed concerns about the size and scale of a Netflix Warner combination, saying the combined company would command significant market share, but he's clearly been watching YouTube and he just says YouTube's too dominant.
38:09
Too many watch hours on YouTube. It's fine. Let it rip.
38:11
He that talking point probably worked on him. We'll see.
38:14
Um Paramount said >> we did um in talking to Ashley, what was it on Tuesday?
38:21
>> Ashley wasn't super excited about having basically one fewer buyer. >> Yeah.
38:25
on the documentary side, but he said it was already so cooked that [laughter] >> he was black billing hard.
38:32
Um, if you want some white pills, our newsletter, tbp.
38:36
com, has a list of AI white pills you may have missed.
38:39
There's a number of them, so go check them out.
38:41
Uh, digging through the archive, finding basically the vibes on the on the timeline have been miserable.
38:46
Everyone's like, you know, fast takeoff, permanent underclass.
38:49
Tyler Cosgrove is not black pilling.
38:52
He thinks that we're in the fast takeoff.
38:53
There's been a lot of white pills. >> Lot of white pills.
38:55
Recursive boling white pill like they they hooked they hooked an AI agent up to a bolab. >> That is a white bell. >> Yes. >> No, I agree. I agree.
39:05
But a lot I understand that a lot of people might see it as a black bell.
39:11
>> Uh Erica over on X posted that Eileen Goo to join uh Benchmark as a senior associate when she returns from the Olympics. She photoshopped this.
39:20
Of course, we did not put this out, but uh it's a good bit.
39:26
Eileen competes for China in uh the Olympics, even though she's a US >> citizen.
39:33
>> Citizen >> and an American student.
39:35
>> Uh and an American student.
39:35
Uh Bill Gurley responded, >> I know a lot of I know a lot of Americans do this.
39:39
I have a buddy who uh was able to go play pingpong in the Olympics because he had dual citizenship or he wasn't even like a dual citizen.
39:46
He just like his parents were from another country so he was able to go be on that team because a lot of the other teams are like way less competitive.
39:51
So it's like you wouldn't make the US team but you want to play anyway so you go with the other country.
39:56
But >> uh I think with Eileen it's because the sponsorship dollars you can make by being a Chinese star.
40:03
There's all these Chinese brands. >> She's a global star.
40:08
>> You can just get paid.
40:09
>> You can get a bunch of it.
40:09
Not not to compete in the Olympics but broader endorsement >> deal. Yeah.
40:13
You're on the wedies box.
40:15
>> Getting getting a gold medal is like UBI because you'll be able to do >> you'll be turned into a Libu potentially. >> Well, yeah.
40:22
In this case in this case. Yeah.
40:23
But Bill Gurley actually responded. >> What did he say?
40:25
>> And said, "I can't believe you actually found out about this.
40:27
We were planning to keep it a secret, but he wrote it in Chinese. >> He's leaning in.
40:31
He's leaning into the meme." Interesting.
40:36
>> Turn post notifications on for when Delhi response.
40:38
>> Girly will be coming on the show. >> Yes. We're very excited.
40:41
>> When his new book launches drops, we've got over there.
40:43
I don't know how embargoed it is, so I haven't been talking about it.
40:47
>> So, we're not going to share it with people.
40:49
>> Uh, but uh, moving on.
40:49
We should talk about >> public.
40:52
We should talk about public.
40:54
Public is investing for those who take it seriously.
40:55
Stocks, options, bonds, crypto, treasuries, and more with great customer service.
40:59
>> Um, amazing ad that Leaf over at uh, Public put together all about uh, uh, teaching kids the power of compound interest.
41:07
A very different positioning from many other platforms.
41:10
Um, and I don't I I you know, it was all AI generated.
41:15
He made it in like two seconds, I guess.
41:16
But, uh, it was like still emotional because the idea hit me because I'm a father.
41:20
But, yeah, >> it's good stuff.
41:22
Anyway, >> we got to talk about Higsfield. >> Yes.
41:25
>> Uh, we've had the founder of Higsfield on the show a couple times.
41:30
>> They've been accused of rage bait basically.
41:32
>> Yes, I think that's what's more than more than rage bait.
41:33
But uh Rashi in Forbes >> uh has the daily cover story. >> Oh wow.
41:43
>> All the racist videos and payment problems.
41:44
The dark side of this AI startup's super fast growth. >> Interesting.
41:47
>> Influencer marketing has helped AI company Higsfield hit 300 million in annual in uh annual revenue run rate in just 11 months.
41:53
But misleading marketing tactics and a social media strategy based on shock has led to backlash among creators.
42:00
In late January, Tim Sorret, a Londonbased video game director, received a message on a social media site from the marketing team at Higsfield, a fast growing AI generation startup.
42:10
This is the biggest moment in Higsfield history, and we want you to be a part of it." It read, "The 1.
42:15
3 billion startup whose tools are used by some 15 million creators and ad agencies to churn out four and a half million video clips every day, was about to launch a new tool called Vibe Motion, which uses AI models to convert text prompts into motion graphics. You like good vibes. You have motion. Use Vibe motion. >> The offer.
42:33
If Sorret shared the startup social media post along with a video clip from preassembled marketing materials, the company would pay him $200.
42:39
But Zoret, who has spent years designing graphics, both manually and with AI tools, could tell something was off.
42:44
Videos Higsfield had shared with him lacked the visual quirks of AI.
42:46
He's like, "This is simply too good." >> Interesting.
42:51
>> And he quickly realized that some clips in the media kit weren't generated with AI at all.
42:54
AI at all. I said this it was a huge alpha in just taking a cinema camera filming yourself and being like this is AI generated and I'm raising it's fraud but alpha [laughter] >> that's not >> don't do it but but there was a moment I mean uh this happens a lot with open AI
43:11
stuff where people will see a video and they'll be like wow they definitely have a new model and it's like they also have a production budget like they can't just hire actors >> well so with with when when Sam Blonde launched Monaco yesterday >> was his video AI cuz it looked it I it was 50/50 for me. >> Yeah. I >> Yeah.
43:28
I >> like there's definitely there's definitely >> I think you would just rip a direct. >> Oh, I know. You obviously could. Yeah.
43:34
>> But clearly the models are good enough where you could >> and it's kind of a funny thing if you're launching an AI company to show like I made this with AI. It's that good. >> Good enough.
43:44
>> Um anyways, so >> they get this guy gets media kit >> says, "Hey, can you share some of these AI outputs?"
43:52
But he found out and said they were video templates that appear to have been lifted from a stock site in VTO in which the startup had pasted its own logos.
44:00
While Soret didn't share the videos, others did circulating the stock video templates on X to promote Higsfield. >> Wow.
44:07
>> All this hype is fake and it's bot.
44:07
>> All this hype is fake and it's bot. S told Forbes Hicksfield's co-founder and chief strategy officer Mahi told Forbes Idi kit was created by an employee in the startup's marketing team for ideiation purposes and was inadvertedly
44:20
shared with creators saying the company's processes went haywire >> uh with its library of 400 presets of camera motions and visual effects Higsfield offers an easy way for creators and advertisers to produce cinematic short videos through textbased prompts. You guys already know this.
44:35
You guys already know this. Mhm.
44:37
>> Uh Mahi says, "We fully admit that we pushed the envelope.
44:39
We learned from what works on platforms like X and very explicitly it's more controversial content that gets attention." So >> there you go.
44:47
>> That's translated into hockey stick revenue growth.
44:48
Last month, the startup claimed it doubled its annualized revenue to 200 million in just two weeks. Yeah.
44:54
>> Driven largely by subscriptions from its 300,000 paying users. >> Wow.
44:59
>> By early February, its annual run rate crossed 300 million.
45:03
>> CEO Alex told Forbes that he hopes to reach 1 billion.
45:06
By the end of the year, after raising 80 million from firms like Excel and Menllo in mid January, the startup is now in talks to raise funding again.
45:13
But that rapid growth appears to be driven by aggressive, shocking, and sometimes misleading marketing tactics.
45:18
Um, anyways, they're moving very quickly.
45:21
I It was It was interesting when we asked Alex about his margins, he kind of he seemed kind of he gave kind of a concerning response.
45:31
He he he immediately was like >> yeah well when you're reselling a model you are you know at risk there at the same time like with a different business model with a different customer acquisition flow.
45:46
It's possible to have good margins on top of models if it's deeply embedded.
45:51
Uh, I I've actually been getting served a whole bunch of Higsfield uh spawn con or like how-to videos that are amazing.
46:02
[laughter] Uh, a lot of them are testaments to the power of the underlying models like Nano Banana.
46:07
I saw there's this cool video where this creator was talking about how hard it is to film yourself.
46:13
Like let's say you want to do uh one of those like cool like I'm locked in videos which I really enjoy.
46:18
Uh, and you want the camera to be uh here getting a profile shot of you and then flip around, go behind you, over your shoulder, then spin around, show your face, then go inside the keyboard.
46:29
That can be a really, really hard shot to get.
46:30
You might need like a automated camera robot.
46:33
Like you need a a cucka robot arm to do that move.
46:36
Uh, MKBHD has one, but they're very, very expensive, hundreds of thousands of dollars.
46:43
Not every creator can afford one.
46:44
And so he was basically showing this workflow where he will film himself on a tripod here, move the tripod, move the tripod, move the tripod, and then put in the start and end frames into Nano Banana Pro, and then it interpolates those and get and does the sweeping move.
46:58
And because it's all in motion, there's motion blur, and it has two reference frames of like Jord's side and then Jord's back.
47:05
It actually looks perfect and it doesn't need to come up with as much imagination.
47:10
So it it so it delivers really well.
47:12
And a lot of those have been sort of those like little like fun uh like tools or like workflows have been uh promoted by influencers and then they'll say like comment and I'll send you the prompts or I'll send you the workflow or I'll break it down or I have a course or something like that.
47:28
Info products on top of these things and uh I haven't really jumped in but I've seen those and been like I would love to make that video of me or my dog or whatever my car.
47:37
It's it's like it's a fun thing and for the right person that's going to actually work as a marketing ad or work as >> content for them.
47:45
Really quickly, let me tell you about 11 Labs.
47:47
Build intelligent real-time conversational agents.
47:50
Reimagine human technology interaction with 11 Labs.
47:52
Also, I got to tell you about >> 11 reader because I used it this morning in the car to read or dictate the Will Manitis essay and I really enjoyed it. So, shout out 11 Reader.
48:04
>> Can it do Will's voice yet? >> No.
48:06
We get Will to license his voice. >> We really should.
48:08
Um, >> well, the author >> I had the option to pick a few different voices and >> you can clone your own voice, but you can't clone someone else's yet because it makes you read like certain text.
48:19
>> The So, uh, the iconic voice collection, it's it serves to you and says, do you want to do Michael Kaine who's a British icon, Bert Reynolds, the masculine iconic storyteller, Richard Fineman, raw genius, Sir Lawrence Olivier, and Dr.
48:35
Maya Angelou and I picked Bert Reynolds, the masculine iconic storyteller. So, he led me well.
48:41
>> Here's maybe where Higsfield uh kind of crossed the line.
48:43
Apparently, they shared they were sharing a Google Drive folder with creators that had uh popular children's characters like Shrek, Moana, and Mickey Mouse saying >> saying things that are on the show, but >> sound a little uh a little racist.
48:58
Uh and then non-consensual deep fakes of public figures like Sydney Sweeney and Zenaia and President Trump.
49:07
So yeah, the the the challenge here is like companies like Higsfield are competing with these Chinese open source models that have no rules.
49:17
They'll generate whatever you want.
49:19
>> And clearly there's an insane amount of demand for video models that have no rules, >> of course.
49:25
>> And so uh this isn't surprising even though it's wrong.
49:28
Yeah, it's I mean it goes back to the the Bit Tor and Napster analogy where like the demand for things that just can't be done legally is really really high and you have to you can't you can't view that as a business opportunity because of course you'll see really really high demand for completely free music if you're not paying anything, right?
49:49
You'll see really high demand for generate a Mickey Mouse character.
49:53
The question is, the business question is, can you actually do the deal, license it, make sure every all the stakeholders and all the legal parties are represented and happy with the deal and everyone's paying.
50:02
Um, if if if you're just throwing out, you know, free IP violation, um, you're going to see huge demand, but >> it's probably unsustainable. >> Yeah.
50:12
So, they did launch a program called Higsfield Earn, >> which was basically like a clipping thing.
50:17
you could make stuff, share it, >> depending on how much engagement it got, you'd earn some money.
50:21
That led to them getting uh banned from X or their account shut down.
50:28
>> I didn't realize that.
50:29
>> Um so yeah, we'll we'll see how this nets out.
50:31
I mean, Alex uh I'm sure will uh I would hope would adjust course. >> Yeah.
50:37
>> Uh he is >> they're clearly very talented at a lot of things, but it sounds like they've been >> blurring the line.
50:43
I mean, for for what it's worth, I still think that I I don't know how fast you would get steamrolled uh by the labs, but I do think that there's an interesting opportunity right now for like little bit of what Cap Cut's doing plus Nano Banana, Sora stuff where you have a lot of video generation, but then you also have the ability to lay over just actual text, actual motion graphics.
51:08
And if you pair up some of the more deterministic tools like basically tool use like how how chat GBT can also do a math problem in Python and be 100% accurate.
51:20
Um you could imagine if you if you want to just draw a rectangle on top of the the video that you've generated or make it black and white.
51:31
That's something that doesn't require a generative AI model.
51:32
you can just call a tool that has the same functionality as Premiere Pro and then it'll just do that.
51:40
Uh, but you have to actually wire all that up, make it all work together and maybe that's where this goes and then I don't really know how defensible that is, but we'll see.
51:50
>> Yeah, I was just going to say like I mean you're you're kind of just describing like a a rapper, right? >> Yeah.
51:54
At least in in in the in the text field like seems like rappers at least of late like have not been >> Yeah.
52:01
You know, people are much less bullish on rappers now than they were like a year and a half ago.
52:06
>> Yeah, especially the the the the huge problem. >> I don't know. SD kid pretty good.
52:12
>> Yeah, [laughter] I like that.
52:13
Let me tell you about Lambda.
52:15
Lambda is the super intelligence cloud building AI supercomputers for training and inference that scale from one GPU to hundreds of thousands.
52:20
The other really hard thing about the video rapper, the video uh like app is that many many video producers uh they start fresh every day.
52:31
Like we when we edit die often our 30 minute cut down of the show um like the project is started from basically a clean slate.
52:40
And so actually shifting from one software to another is really really easy.
52:45
I mean you still have to learn the new tool. It has to be better.
52:47
It has to have have all the features, but it's not like, okay, well, there's all these networks of documents.
52:54
There's a system of record.
52:54
It's different if you're editing a movie over two years.
52:57
You're probably not going to switch tools midm movie.
52:59
But if you're a shorts editor, you can switch from Cap Cut to Instagram edits, like their video edit editing tool, really quickly.
53:06
You can go over to After Effects or or Premiere, and you can just be like, for this for this video, we're using this.
53:14
Like, this is this happens to us all the time.
53:16
we'll we'll have a video editor on the team who's like oh yeah they use Da Vinci Resolve and it's like it doesn't even matter to us because it's like it's like oh yeah they just use that and then they plug in and then they deliver a video file at the end they render it out.
53:29
>> Anyway, >> uh one more thing on Higsfield.
53:30
I would expect that Disney comes after them too.
53:35
They're generating Disney characters right now.
53:37
Disney invested a billion dollars into OpenAI as part of this licensing deal. One-year exclusive.
53:43
There's they I would expect that pressure to come soon.
53:47
>> Let's move over to Bill Aman.
53:47
But first, let me tell you about Shopify.
53:49
Shopify is the commerce platform that grows with your business and lets you sell in seconds online, in store, on mobile, on social, on marketplaces, and now with AI agents.
53:58
>> Bill Aman makes a big bet on Meta.
54:00
Persing Square disclosed a roughly $2 billion position in Meta.
54:06
Uh hedge fund manager Bill Aman likes Mark Zuckerberg's chances in AI.
54:08
at Persing Square revealed the stake at an annual meeting of one of its fund on Wednesday.
54:16
The position amounted to 10% of the firm's capital at the end of 2025 or roughly 2 billion based on past disclosures.
54:23
Meta shares are down about 13% over the past six months.
54:25
A decline that Persing Square attributes >> to investor concerns about the sums the company is spending on AI >> capex.
54:33
Uh Persing's thesis revolves in part around AI boosting meta's content recommendation and personalized ads and potentially unlocking new opportunities in wearables or AI digital assistance for businesses.
54:46
>> Ben Thompson, >> keep reading.
54:48
I'm going to pull up >> Meta's business model is one of the clearest beneficiaries of AI integration.
54:53
Pursing Square said in its presentation, Aman tends to concentrate his stock portfolio in a small number of high conviction bets.
55:01
He only had 13 different positions at the end of 2025, including other big tech companies, Alphabet and Amazon. What did you say?
55:11
>> He just liked me for real. Uh, no.
55:11
I >> You quote tweeted this, right? >> Yeah.
55:14
There's a I mean, I I didn't like really add anything.
55:16
I just posted a Jeremy Gon.
55:18
>> You You created this post and you photoshopped Jeremy's name on there because it was your idea. It came from you. What did Jeremy say?
55:25
>> I just want to give him credit.
55:26
>> He said, "The idea that it's hard to beat the market is mostly trotted out by money managers.
55:30
What they really mean is that it's hard to do when you manage other people's money because it's difficult to get paid for doing the obvious thing.
55:37
I think it's substantially easier than most people think for the person solely running their own cash.
55:42
So, Aman is charging two and 20 to own the Mag 7. You love to see it.
55:48
But a lot of people don't have diamond hands.
55:49
He will diamond hands for you uh by by holding Alphabet, Amazon, and Meta apparently.
55:55
Let me tell you about Railway.
55:57
Railway is the all-in-one intelligent cloud provider.
56:00
Use your favorite agent to deploy web apps, servers, databases, and more.
56:03
While Railway takes automatically takes care of scaling, moderating.
56:06
Speaking of Meta, Rayban maker Luxodica says it more than tripled Meta AI glasses sales in 2025. >> Take off.
56:17
>> The French Italian eyewear brand said it sold over 7 million AI glasses last year, up from 2 million that the company sold in 2023 and 2024 combined.
56:26
>> Wait, so so sorry, say that number again.
56:27
How many did they sell last year?
56:29
>> 7 million devices last year.
56:33
>> I I I think I'm I'm starting to see them.
56:35
I mean, Ashley Vance came on the show and was wearing them for fun, but also for good content, and he's going to actually include that content in the stuff that he makes.
56:42
It's a video makers tool, so it's a proumer tool.
56:44
Uh he of course had a whole bunch of other nice cameras here, but that is another tool that you can't really replace.
56:50
He doesn't want a GoPro strapped to his head.
56:53
Yeah, that doesn't make any sense.
56:55
Then you have the extreme sports.
56:55
I'm going skiing in a couple weeks with a bunch of friends.
56:59
I'm sure I'll see a pair of uh of the vanguards out there on the slopes.
57:02
And then uh and then you just have the funny people on Instagram reels saying, "Computer, give this guy a great day."
57:09
Like that that comedy is great. Did you see that?
57:11
That that guy ran into another creator that was doing this yelling at him. Yeah. Yeah.
57:17
Computer, computer, give this guy a bust down AP. Something like that.
57:19
It It was just like very very funny.
57:22
And every time I see one on a truly viral Instagram reel, I'm like, "Okay, there's there there's something here that's unlocking a different form of of video, not unlike when we saw the first GoPro go out and we started seeing truly crazy firsterson video of surfing, which like was just impossible or like I made I would go surf >> surfing.
57:44
I went surfing with uh Roco Basilis Basiliscoco.
57:49
>> Roco Basilis, >> the uh the Luxodica exec and error who led the project with Meta. >> Yeah.
57:56
>> Uh he took out a new pair of the Oakleys.
57:58
And uh we did lose them briefly, but we recovered them.
58:03
>> They need Meta Croakkes, right?
58:03
Isn't that what they're called? Croakies. >> Didn't have croakies.
58:06
>> Yeah, you need the Meta Croakkes. Let's get on that, Zach.
58:07
Um >> uh but anyways, Bloomberg reported that Meta and Luxodica were discussing doubling production to at least 20 million by the end of this year to meet growing demand.
58:17
So >> I I I think they're going to be >> they're selling they're selling. Give them credit.
58:21
Give Give Zuck some credit.
58:24
And again, I I saw somebody had uh effectively jailbroken their Meta Ray-B bands and they hooked it up to a Mac Mini and they were buying stuff.
58:30
I've talked about this at least once on the show, but >> again, there's um that's it.
58:35
It's I it's funny to watch somebody jailbreak a device to do something that the device will obviously do like in the relative near term. Yep.
58:44
That's like the part of the entire thesis behind the device. But >> yeah.
58:48
Well, I mean I don't know.
58:48
Will it actually tell you when you get an iMessage in the next three years? Is that coming? I don't know. It might not.
58:57
Uh that those wild gardens might hold for a while.
59:00
Do you think the Meta Ray-B bands will outsell sunglasses broadly?
59:07
Or is the TAM sunglass buyers?
59:07
Everyone has one pair of sunglasses, so you sell one pair of these to all the people.
59:12
Or is it And some people just don't like sunglasses, so they don't wear them.
59:16
Or are they like TAM expanding in some way?
59:24
>> They've got to be h That's a good question. >> Yeah.
59:27
and and I will give you assume the next version beyond that is is like truly indistinguishable. >> Yeah.
59:34
So my my framework has been that the immediate market of people over a billion people that just wear glasses daily so that they can read feels extremely >> in reach right because you're wearing this all day long.
59:49
>> Why not add some >> functionality to it? >> Totally.
59:52
Uh certainly there'll be some bloodites that just want the analog magic of just a a pair of plastic with some glasses in it, but um but then yeah, I don't know. I I've never It's funny.
1:00:03
I love the way sunglasses look, but I've never been a sunglasses guy.
1:00:05
I just don't >> really >> I don't I rarely rarely wear them.
1:00:09
I still once or twice a year will buy a pair and then I because they look cool and then I don't actually end up wearing them.
1:00:17
>> I like sunglasses when it's bright out.
1:00:19
I just like the functionality of them.
1:00:21
But I do lose them constantly.
1:00:21
So >> Oh, you're one of those guys.
1:00:25
>> Yeah, I'm I'm I'm a sunglass loser.
1:00:25
Uh but the uh the the question is like uh if you comp to the Apple Watch, like it feels like more people are wearing watches than before the Apple Watch because watches were sort of like a niche.
1:00:40
You had to be into watches to wear one because everyone had the phone in their pocket and they were like, "Why do I need a watch?"
1:00:45
And then once the Apple Watch came out, a lot of people were like, "Well, I I want a superwatch on my wrist.
1:00:50
I want a computer on my wrist.
1:00:52
I want uh the the iPhone >> bus down Apple Watch.
1:00:56
I think we're on to something. They're doing orange.
1:01:01
They're doing Hermes orange.
1:01:01
They're selling very well in China. The bus down. They did the gold. It was $10,000. Do you remember this? >> And it flops, right?
1:01:09
>> The the Apple Watch edition. It was $10,000.
1:01:11
Goldplated, solid gold, something like that.
1:01:14
Uh, I don't know if it flopped.
1:01:16
They did discontinue it, but it it uh >> getting roasted in the chat.
1:01:21
I like sunglasses when it's bright out.
1:01:24
>> Doesn't like sunglasses when it's a hot take.
1:01:27
>> That's a bold statement.
1:01:29
>> Heavyhitting analysis.
1:01:31
>> But, um, I like I like the idea of the of the bust down Apple Watch. Get it at 100K.
1:01:35
Uh the problem with the the the the gold Apple Watch was that you buy this thing for $10,000 and it immediately loses all its value because it doesn't even run the latest software and it just stops working after like 5 years, which is ridiculous.
1:01:47
Like if you buy if you're spending $10,000 on a watch, you want it to be useful in a hundred years.
1:01:53
You want it to be around for a long time.
1:01:54
But >> uh should we read through Jeremy's actual >> Sure. >> response.
1:01:58
>> Oh, I I Oh, did he post again? Because I read his Oh. Oh, okay.
1:02:01
So, he did talk about this really quickly.
1:02:03
Labelbox, RL environments, voice, robotics, evals, and expert human data.
1:02:07
Labelbox is the data factory behind the world's leading AI teams.
1:02:11
And yes, >> Jeremy says, uh, responding to Acman charging 2 and 20 to own the Mag 7.
1:02:15
He owns Google, Amazon, and Meta.
1:02:17
He says, this is good capital allocation, but it requires permanent capital.
1:02:21
People don't actually want the best returns.
1:02:23
They want to feel clever and do some some like weird little thing that makes them feel interesting or different or something.
1:02:30
But sometimes you just got to do the obvious thing and just be really strong on it.
1:02:34
I always admire those managers who have been paid 2 and 20 for like 40 years to just own Bergkshire.
1:02:39
It's awesome because they were right, but they were and they were providing a service.
1:02:43
The service is that the LP does not have the personal confidence to own Berkshire on their own.
1:02:47
So they actually need to pay a huge premium for someone else to give them the confidence to own Bergkshire.
1:02:52
It's just a tax that they pay for not having the conviction to do it themselves.
1:02:56
And I think that that's great and solves a real problem for a lot of people.
1:03:00
>> [laughter] >> I love it. >> Well said.
1:03:02
>> Let me tell you about Crowd Strike. Your business is AI.
1:03:04
Their business is securing it.
1:03:06
Crowd Strike secures AI and stops breaches.
1:03:11
>> Uh we got a one more post uh before we bring in our first guest. Um >> what you got? >> We can pull this up.
1:03:17
People uh are reacting to the men's luge doubles.
1:03:22
>> They're saying it's one of the most baffling things I've ever watched.
1:03:23
and [laughter] national champion speaker says, "How do you find out you're good at this?
1:03:28
What's the conversation look like? I really want to know.
1:03:33
I I didn't I didn't know this >> doubles.
1:03:36
I guess I guess I I may have seen this before.
1:03:39
I I remember as a kid watching singles. >> Yeah. >> But this is baffling.
1:03:43
>> There's also something called skeleton.
1:03:45
>> Like what do you what do you >> I think it's an evolution.
1:03:46
I think it starts with like a toboggan and you have like a bunch of people essentially in like a rowboat, you know, in like a long like if you're doing crew, you have like four or five people and then you're just like, let's do one less. Let's do one less.
1:04:01
Let's optimize and reduce weight and then they're finally down to just two people and they're basically just on this like sled together and uh yeah, I guess you get here. But very very funny.
1:04:12
What else is going on in the Olympics?
1:04:14
personal injury attorney in his 50s is on the cusp of becoming the oldest American winter Olympian in history.
1:04:18
All he needs is for one of his teammates to slip and fall.
1:04:22
[laughter] This guy is elite.
1:04:24
Remember the last Olympics where there was that uh that great meme that came out where it was like the one person with like all the gear and all this other stuff and then the random guy I think from Turkey who just showed up and was just like and just like was uh I think he won the gold or something. I don't know.
1:04:39
That might have been too poetic.
1:04:40
Um, but modest proposal says it's an incredible >> feel like being a personal injury attorney is probably a great gig if you're in one of these kind of fringe Olympic sports.
1:04:49
You know, you want to have some flexibility in your schedule.
1:04:53
You want to be able to make enough to reinvest into the sport.
1:04:56
>> This guy's clearly locked in. >> This guy is amazing.
1:04:58
Okay, so uh Rich Ruonin is a 54 yearear-old personal injury attorney who plays an indispensable role for the US curling team.
1:05:06
He cooks the players omelets before important matches and grills steaks after huge wins.
1:05:10
He handles the early morning grocery shopping so they can sleep in. What a team player.
1:05:15
He chauffeers them around in a rental minivan.
1:05:18
He even pays for some of their flights and hotel rooms.
1:05:20
But those aren't his only jobs.
1:05:22
He is also the most unlikely curler at the Milan Cortina Games where he could become the oldest American athlete in Winter Olympics history.
1:05:31
And just in case people don't believe that someone nearing retirement age could possibly be an Olympic athlete, Ruonin answers the question on a homemade t-shirt.
1:05:40
I'm not the dad and I'm not the coach, he says.
1:05:46
But he is an American curling legend.
1:05:48
Ronin first appeared in the national champion tournament in 1998.
1:05:50
He's been doing it for over 20 years, maybe.
1:05:53
won it a decade later and has been a fixture in the sport since long before any of his teammates were born.
1:06:01
What he had never done was curl under the sport's brightest lights.
1:06:06
Roonan just missed qualifying for the Olympics several times.
1:06:10
His most recent heartbreak came before the 2020 Olympic Games when the US trials mixed in doubles. >> 2022, sorry.
1:06:17
uh when the US trials mixed in in mixed doubles cames down to the last shot where Honen decided to focus on his law practice in the senior circuit.
1:06:25
At 50, he finally gave up on his Olympic aspirations.
1:06:28
I honestly thought it was over four years ago.
1:06:30
He said, "Well, I probably thought it was over eight years before that.
1:06:34
Then he got an unexpected call. >> He's back.
1:06:39
>> The captain for one of the nation's upand cominging teams was battling a rare autoimmune disease and needed a substitute.
1:06:45
On paper, the Gen Xer wasn't the most natural fit.
1:06:48
The other curlers were all in their mid20s, roughly the same age as Ronin's children.
1:06:52
He even has a few decades on Team USA's coach. >> Few decades.
1:06:57
>> But on ice, he blended right in.
1:06:57
So when Skip Danny Casper recovered from his illness and return, Roonan stuck around as the alternate.
1:07:04
Team Casper then qualified for the Olympics and became Team USA.
1:07:12
Rahon says throwing the rock at the Olympics would be the single greatest moment of his entire life.
1:07:15
My kids know it and my wife knows it.
1:07:19
>> I'm gonna start using I'm gonna I'm gonna I'm gonna put on curling. >> Yeah.
1:07:23
>> At home and just start using some of the hardcore terminology. >> Yeah.
1:07:27
>> Yeah. should be like ah [sighs and gasps] I can't I can't >> that you know that happened a couple decades ago because on CNBC which all the hedge funds watch uh after the market closed they would just roll over to NBC coverage when the two fun when
1:07:41
the two companies were together and they would show curling just because of the time change that's what was on and so all these Wall Street guys became obsessed with curling and got really into it started placing bets and stuff it's a fun story. >> Yeah. I wonder how how big is the like >> Yeah.
1:07:51
I wonder how how big is the like gambling and sports betting around the Olympics? >> I don't know.
1:07:56
I I saw some numbers for the for the Super Bowl.
1:07:57
It was really really big.
1:07:59
But um you have to imagine that the Olympics is also huge.
1:08:01
While you look that up, let me tell you about Turbo Puffer Serverless factor in full tech search built from first principles and object storage.
1:08:07
Fast, 10x cheaper, and extremely scalable.
1:08:09
And our next guest is here so we can come back to Olympics coverage after.
1:08:13
But let's bring in Brian Johnson to the TV Ultra Dome.
1:08:15
I will tell you about Vibe.
1:08:18
co while he walks in [music] where DTOC companies, BC, B2B startups, and AI companies advertise on streaming TV.
1:08:24
Pick channels, target audiences, and measure sales just like on Meta.
1:08:27
And you might be doing some advertising soon because there's a huge launch. How are you doing?
1:08:33
>> Hey, I brought you guys a gift. >> Thank you. What'd you bring? >> Are we doing shots? >> We're doing shots. >> Amazing. Here we go. >> Amazing. Yeah. >> Okay.
1:08:40
>> Have you done a shot of olive oil? >> I haven't.
1:08:41
Take out our >> I've never done I've never done a shot of olive oil. What are the benefits? Break it down. Why would I do this?
1:08:45
I like olive oil on a steak. >> Yeah.
1:08:49
I mean, have you Is olive oil a part of your life?
1:08:52
>> Um, only in the cooking context, I Yeah, >> but I should just be drinking it. >> Shots yet? >> No. >> What? What? Yeah. What are the benefits? And what is this? Did you grow the olives? Is it cold press? >> Just this much. Should we not?
1:09:03
Should we not do a little bit more? >> No. No. [laughter] No. No. No. No. This is live TV. 15 ml. This is live TV. >> Yeah.
1:09:11
This is This is the precise deal. 15 ml. 15 ml. >> One tablespoon. >> Okay.
1:09:14
And what's this going to do for >> It's the I think it's the superfood of superfoods. >> Okay. >> It does.
1:09:20
I mean, just like you look down the list of things it does.
1:09:21
It's just good for almost everything. >> Okay.
1:09:25
>> So, >> is it high calorie? How many calories?
1:09:27
>> 120 130 calories in this. Okay.
1:09:27
Calories don't scare you, though.
1:09:30
>> No, I mean it's 15% of my daily caloric intake.
1:09:33
I I consume more olive oil than any food. >> Yeah.
1:09:36
How many How many like shots is that roughly?
1:09:39
>> So, 45 So, three tablespoons a day, 45 ml, three shots a day. >> Okay. Okay. Wow.
1:09:43
>> So, one with every meal. >> One with every meal. Okay. Cheers.
1:09:47
>> Cheers to Brian Johnson blueprint. Cheers. down the hatch. Delicious. >> That is good. >> Smooth. >> Very smooth. No problem. I could do that. >> No bite.
1:10:00
>> And and and and so and so I just drank 15 milliliters of >> of uh of olive oil and I'm going to live forever now, right? >> Basically.
1:10:10
>> So now I can go back to diet and nicotine. Right. >> Yeah.
1:10:14
Do you feel the sting now? >> A little bit. >> I do. I do.
1:10:16
>> Like a little cough bubbling up. >> Yeah. >> Yeah.
1:10:19
Perfect for broadcasting. [laughter] Yeah. Yeah.
1:10:21
That's it's a good sign of a good olive oil.
1:10:23
So we we source this in both hemispheres. It's always fresh. Okay. >> So it comes off. >> Yeah. Yeah. >> Yeah. So >> where Yeah.
1:10:28
Where where more specifically >> this one uh this is from Chile. >> Chile.
1:10:33
>> And and the brand is called Snake Oil. Very funny. Yeah.
1:10:37
>> Um how does this fit into the overall blueprint?
1:10:41
Uh what do you call empire at this point?
1:10:43
point? I mean we the whole thing on blueprint is we are trying to basically say what does the evidence say for what you can do to be healthy and we just went after the best foods the best therapies and what I found is basically most things don't work most things are
1:10:59
>> and so we've tried to narrow in to focus on the most narrow number of set of things to do and olive oil just stacks is like one of the very best things you can do in life >> is part of why this isn't pushed more uh aggressively by the sort of health industry is that it's kind of hard to deliver. A lot of people don't want to
1:11:14
A lot of people don't want to just be consuming oil and that's kind of like the bit is other some people are selling pushing snake oil.
1:11:22
>> Yeah, that [laughter] I mean >> yes I mean people I think have a natural aversion to fat still.
1:11:27
There was that there was a long period of time where fat was bad and you saw marketing 30% less fat, 40% less fat.
1:11:33
So people think olive oil fat therefore bad. Yeah.
1:11:38
>> So there's like hangover. >> Yeah.
1:11:39
I had a buddy that had internalized that so much that he was on a almost entirely fat-ree diet. He got his labs done.
1:11:47
>> He had like low his test was in the low hundreds.
1:11:50
All he did was add like an extra avocado a day.
1:11:52
That was the only thing he changed.
1:11:56
>> His tests like went up like by a pretty meaningful multiple.
1:11:59
>> So, walk me through, you know, average American might spend, I don't know, $10,000 a year on food.
1:12:04
What what does it look like to get into the blueprint ecosystem at that order of magnitude?
1:12:09
Then what can people do at $100,000 a year?
1:12:12
And then I want to hear about the milliondoll a year all-in plan.
1:12:16
>> Well, we announced that today. Okay.
1:12:16
So, it's called Immortals. >> Immortals.
1:12:20
>> A million dollars a year.
1:12:20
And it's my exact protocol.
1:12:24
>> It's your exact >> my doctors, my concierge team, all the testing infrastructure, >> the blood work too, >> everything.
1:12:30
Every test, every therapy.
1:12:32
And so it's I mean this has been really hard to build because it hasn't existed.
1:12:36
So we've just had to scour the evidence, build out the infrastructure globally.
1:12:39
>> And this is basically when someone says >> getting no feedback.
1:12:43
>> Let's turn that uh speaker off. Sorry.
1:12:45
We're we're working on bringing the soundboard to life in the TVP and Ultradump. Sorry. Continue. >> Yeah.
1:12:50
So the I guess like if you look at a few three different frames like one is we measure probably around 250 to 300 things that kill me. >> Mh.
1:13:00
that will kill you or actively >> that actively kill you.
1:13:05
>> Everything in the gas station. >> Yeah. Yes.
1:13:06
Basically, I mean everything in modern the modern world. Yes.
1:13:09
>> And so if you look at it from like don't die. Okay.
1:13:11
So what things actually cause you to die?
1:13:13
And then if you look at it from the positive things of what makes you live, >> that list is probably like 250 things that kill you and 250 things that make you live well.
1:13:22
>> And if you take on that burden as an individual to say, I want to do all of these things that make me avoid the things that make me die and do the things that do make me live well.
1:13:29
That's a like a herculean task.
1:13:33
>> So, this program basically takes all of those things, makes it easy. So, you just show up. It's like, what do I do?
1:13:39
And the system tells you what to do. >> Okay. >> Yeah.
1:13:41
What What is the What does the program look like more specifically like what how how is very it's very So, you're doing three people, right? >> Yeah. Three thoughts. >> Three to start.
1:13:51
What do their weeks, months, year look like?
1:13:56
>> So, we'll do first it depends on who the person is.
1:13:58
So, we're we have an interview process because the person needs to be willing to work. >> Yeah.
1:14:04
>> You can't just like show up and like give me a pill.
1:14:05
You have to put in probably around like >> I was promised immortality with a single shot. >> Yeah. And you tell me.
1:14:11
>> That's why it's called snake oil. >> Yeah. I got you got the best.
1:14:14
>> So, we have an interview process.
1:14:14
We'll make sure the person's willing to put in the work that they've got a pain tolerance.
1:14:19
You know, they can like, >> you know, like they have they have to be willing to do stuff.
1:14:22
You know, sometimes therapies hurt.
1:14:24
Sometimes it hurts to be hungry.
1:14:26
sometimes like so there's some discomfort and then they uh we'll choose the three people >> and then we'll do a comprehensive baseline.
1:14:34
So for me over the past 5 years we we are collect counting yesterday uh we have a few billion data points on me over the past five years. >> That's a lot.
1:14:45
>> And so then then if you say what of those few billion data points are usable probably a few hundred million. Yeah.
1:14:50
>> But still we have a few hundred million data points.
1:14:51
And so what we've done is to date I guess like years ago we that existed in spreadsheets and files.
1:14:55
We've now set up an AI system where we take this um we take all the context >> and now we have this inference engine >> to say like what relationships can you find in all this data. Yeah.
1:15:07
>> And so what we're building is an AI a Brian AI that basically watches after you 247.
1:15:12
So you start you do a whole bunch of comprehensive baselines where you in the world. Yeah.
1:15:15
And then we start doing targeted protocols like how do you isolate cardiovascular health and how do you look at you know so on and so forth.
1:15:22
Then you're going to find issues as well where the person may have a surprise finding how do you address that they'll have some you know anomaly.
1:15:26
So then we just get after it and we just say like how do you basically take the baseline measurements and make all of them better. >> Yeah.
1:15:34
>> Do you want these people to be local? Can they be anywhere? >> Anywhere. Cool.
1:15:38
>> Are they all over the world or US-based? Is there advances?
1:15:42
>> We're open open to everything. Okay. Interesting.
1:15:44
Um, how how are you thinking about AI?
1:15:46
When I think about a product like snake oil, I think uh not a lot of AI disruption risk, not should not be sold off in this SAS apocalypse.
1:15:57
You're not going to vibe code an olive tree.
1:16:00
The the land exists, the tree exists.
1:16:02
You have to select it, press it.
1:16:05
There's probably a lot of people involved that uh there might be some automated ERP system at some point, but uh how are you thinking about the impacts of AI on your business, your life?
1:16:15
I know that that's a big piece of the don't die philosophy is that we are going to go through a change, but how are you processing the most up-to-date advances in AI?
1:16:24
>> Yeah, you're exactly right.
1:16:24
Is like this olive oil we source from farmers all around the world >> that go through extensive screening.
1:16:30
uh we test every single every single molecule that we manufacture, we third party test.
1:16:36
party test. So they're all very manual processes and then on the other side we have AI where you take all the data, you do the inference engine, you try to find new insights that's never been found before, >> but we're this really weird uh mixture
1:16:47
of like leading edge AI >> and like duct taped together >> of all the manual things like we'll go into someone's house, you measure for mold and toxins and water toxins and air toxins and you look at materials in the house. So it's it's still very manual
1:17:00
So it's it's still very manual and automated.
1:17:02
So it's kind of a cool world where we're we're insulated to some extent in terms of software.
1:17:06
You just stand it up, you're very vulnerable to speed.
1:17:10
Where this one you're dealing with physical world constraints, you have to bring together all these different disciplines. >> Yeah.
1:17:15
And there's also this interesting break in that uh if you want to measure the change in someone over a year and you want to do a year-long study like even if you have a million Einsteins in a data center like you have to wait a year to get those results.
1:17:26
I'm sure there's more things that you can do digging up historical data, all the data that you have, but there is just this natural break on progress when you have to wait for results and then feed that back in. >> 100%. >> Makes sense.
1:17:39
>> Uh you talked about testing.
1:17:39
Uh what are you finding in uh organic products?
1:17:42
If something's organic, do you just automatically assume that that uh it's not going to kill you?
1:17:49
>> I assume it's worse than not [laughter] than non-organ. No, I'm serious. Junkyard dog theory. I'm vindicated. >> It's true.
1:17:55
Like honestly, or organic only looks at you know a certain subset of toxins.
1:18:01
It's not all toxins and then it's a very limited screening protocol.
1:18:05
So it's really a marketing tactic.
1:18:05
When we test organic, it typically performs worse than non-organic on many variables.
1:18:12
So no, I think it's worthless.
1:18:13
Uh it really is as a marketing protocol.
1:18:14
And I would say more broadly, I don't trust anyone like literally I don't trust marketing. I don't trust brands. >> Trust no one.
1:18:22
>> I don't trust technicians.
1:18:22
I don't trust practitioners.
1:18:24
practitioners. like when we go out there honestly it just that's why this protocol is we've built it in such a meticulous fashion we don't trust we don't trust ourselves just trust the data do our protocol see what it produces but I when I go out in the world I'm just like this is scary
1:18:38
>> where uh where do you get things like produce things that aren't nec you know you're not going to add apples to the blueprint website right if you want to eat an apple what do you do >> it so there's no safe place [laughter] sadly like >> Brian Johnson no safe place to eat an apple. No apple is [laughter] safe.
1:18:55
No apple is [laughter] safe.
1:18:57
>> Like we, you know, people identify like you like plant-based proteins have high toxins like because they isolate that because you can test it.
1:19:04
But if you eat a carrot, >> it can have the same if not more toxins. Yeah.
1:19:09
>> And so it's just that these this fresh produce is not measured.
1:19:11
So people freak out over what they can measure.
1:19:13
But if you we've been testing fresh foods and packaged foods and like toxins are throughout.
1:19:19
>> And so it's just a very skewed perception on where toxins are at.
1:19:21
But you know why I typically buy we'll farmer farmers market we'll do you know airw do whole foods when we test these fresh foods.
1:19:28
So we do want to start growing our own.
1:19:30
Uh but even then it's hard because you're in >> LA it's a little rough >> and even those systems themselves like someone will try to get an escape route and say like well I've got you know I have a cow in a in a p you know pasture fred irrigated whatever like they can escape the toxins within the environment.
1:19:45
So it's just very very hard to create an isolated non-toxic.
1:19:47
So, there's lots of lots of toxins at risk out in the real world.
1:19:52
Talk about other risks in the real world.
1:19:54
Have you ever been framemoged? [laughter] >> Sure. People have tried. >> You know what? That is so good.
1:20:02
We should put that on the list.
1:20:03
That is that escaped our >> for a million bucks.
1:20:06
You got to protect me from frame.
1:20:08
That's like the first thing that I want from a >> honestly. Yeah.
1:20:12
What a blind spot we had.
1:20:14
We should have picked that one up. >> Yes.
1:20:15
How are you processing the looks maxing virality that's going on?
1:20:17
It feels like very tangential in some ways.
1:20:22
There's some there's some overlap.
1:20:22
I I I've seen, you know, the looks maxers say, "Don't worry about the macros or the micronutrients or the toxins at all.
1:20:32
All that matters is how much protein you're getting because they they have a very uh live fast, die young."
1:20:36
But there's still some overlap and that they're trying to look good.
1:20:40
We've talked about this before. >> Yeah.
1:20:42
In many ways, I feel like you've been you've been looks maxing.
1:20:43
You wouldn't necessarily call it that.
1:20:45
But you can just see every now and then a photo will go viral where it's like you every six months and people are like wait it's working.
1:20:53
So in some ways it's kind of marketing.
1:20:55
There's a lot of work kind of under the surface that's contributing to that.
1:21:00
>> But at the same time it's very clear like people want you to look better every month or they're going to say >> like hey can I really trust this guy? >> Yeah.
1:21:08
I mean there's two things.
1:21:08
One is I'm like you guys your disposition towards advertising.
1:21:12
You know it's like yes and yeah.
1:21:13
Like if someone's out there doing their thing, you know, like more power to them and like whatever their jam is, if they're doing looks macking, like you know, do your do your art and be beautiful. That's great.
1:21:23
>> And so yeah, I'm very much a proponent of like let humans be beautiful in their manifestation. Sure.
1:21:28
>> Um like we're basically we're saying that, you know, as we all realize like something's happening in the world right now. It's big with AI. >> Yeah.
1:21:36
>> In some form or fashion, it's evolving the world.
1:21:38
And our primary hypothesis is as this evolves, the major shift is humans are going to want life. >> Mhm.
1:21:46
>> Like you know if AI takes our jobs or they take what we do and we want to find new identity, there's going to be a major shift towards don't die and it's going to become the probably the most dominant ideology in the world very quickly because we're going to be pointing ourselves towards vibrancy.
1:22:01
That is that that is an interesting encapsulation because it does feel a little bit like the the two paths meme with the dark castle and the and the bright uh you know palace in the sense that a lot of the looks maxers are saying they're blackpilled.
1:22:12
They don't believe that there will be a future.
1:22:14
So health actually takes a back seat to looks because that's what helps you get a get ahead this year this month.
1:22:22
And it's it's very short termism very short-term thinking.
1:22:27
you've always framed it from day one around very long-term thinking.
1:22:31
So even though I think people might say, "Oh, well like he's trying to look good, that guy's trying to look good." They're similar.
1:22:38
There's actually a huge divergence in the philosophy.
1:22:40
At least it feels like that. I don't know.
1:22:42
Jordy, what do you think?
1:22:43
>> Uh I had another question.
1:22:46
>> Please take it down another direction.
1:22:48
>> So you have this you you're starting with the ultra premium product, the million dollars a year, >> and I can see all the justification for that.
1:22:56
I imagine you want to go more and more kind of like down market over time so that somebody if they have a million dollars or $100,000 or $10,000 or a,000 or 100 can can kind of benefit with the program.
1:23:09
Um, it feels like there would be an insane amount of demand for something like right today, a $1,000 program, something that's effectively a PDF that somebody could pay for, download, >> and because they paid for it, we talked about this yesterday because Tai Lopez was in the news because of his whole like Radio Shack debacle.
1:23:30
>> Uh, and I was saying I when I was probably 18 or 19, I paid for a PDF of a workout plan.
1:23:35
And because I paid for it, I followed it really closely and I got a lot more value for my money.
1:23:41
>> Whereas I feel like for you, the second that you launch like a traditional info product, you will be attacked.
1:23:46
Even though Even though I think that at a thousand dollars, you could put something together that would provide somebody potentially tens of thousands, hundreds of thousands, right?
1:23:57
Just because in the health game, if you if you just buy a PDF and then that gets you to sleep an extra hour a night for the rest of your life, like the the economic value creation or or the the money that you're not spending on hospital bills or things like that um >> uh would be tremendous.
1:24:15
Like how how are you thinking about that?
1:24:17
Because I again like I think that there would be value there >> for me paying paying again for that $1,000 product that I could just kind of follow and learn something from.
1:24:30
>> Uh the value would be there.
1:24:30
But how are you thinking about it?
1:24:33
>> Yeah, that that's exactly what we're doing.
1:24:34
And so the the target audience we're after is those who want to say yes to health >> and we say do this. >> Yeah.
1:24:41
So we're not if they if if you if you want to go out and find source your own peptides, inject yourself, you you know and like do your own protocol like do that.
1:24:50
We are going to deliver autonomous health.
1:24:52
And so like there's self-driving cars, they're self-riting software.
1:24:57
>> This is autonomous health.
1:24:58
>> So we're building AI that just says do this.
1:25:00
And so do you remember early on when Facebook >> uh did that study where they said 300 likes on the platform and they could predict you better than your partner could.
1:25:10
This was like the first time social media showed the the power of prediction. >> Okay.
1:25:14
>> And so there's an equivalent question for us is >> well how much data do we need on your body like of of blood of wearables of whatever to predict you >> better than you can >> and how do we stitch this together?
1:25:25
So I think the free version >> is like some minimal viable amount of data. >> Yeah.
1:25:32
>> That goes into our AI inference machine and we then say do this and we can get higher yield than anything they could ever do themselves. Mhm. Yeah.
1:25:40
>> And once that happens, once you take away Mo, you're like, I'm never going to end with a human driver ever again, >> right? It's a very clear.
1:25:45
So this will be the same thing where once you once you deal with the immortals, the product, you be like, >> I'm in.
1:25:51
Like definitely this is a better control system than anything I could produce.
1:25:54
>> How how are you thinking about wearables?
1:25:55
Uh most of the people that come on the show will be winging an Apple Watch or maybe a Whoop or an Aura. It's very very common.
1:26:04
>> Uh at the same time, it's missing visual.
1:26:06
It's not like pulling in information.
1:26:08
It's not like uh the the health decisions that somebody makes every day is not can't always be picked up here.
1:26:15
It'd be great to have be like, okay, what did this person eat?
1:26:17
What time did they actually go to bed?
1:26:18
All all these different things.
1:26:20
What what do you think the future of health wearables is? >> Yeah.
1:26:24
So that's exactly again what we're building with this first three people and immortals.
1:26:28
Like if you say so my life over the past 5 years has generated a few billion data points >> few hundred million or useful >> that has created a a a dimensional representation of a human that is higher fidelity than anything ever produced in in history.
1:26:44
>> And so we've taken that model we're going to replicate with others.
1:26:46
And so we're going to try to basically cast this gigantic net of data capture.
1:26:52
>> We'll bring it all in and say where's the highest signal >> uh inputs and then we'll scale those things.
1:26:57
And so like you're saying like we may find that lighting uh at bedtime >> Yeah.
1:27:04
>> is [clears throat] an incredibly important factor and like you know right now people don't really think about that or the lighting environment in the room and you can't do like specific photon counts. Yeah.
1:27:11
>> And so we're going to try to tease out like where are the little teeny tiny things that have this gigantic yield.
1:27:17
But it's all about data collection.
1:27:17
It's about the inference engine and it's about people who willing to say yes I'll follow the protocol because it's inherently better than what I would do myself or just guessing >> and then we'll have everything.
1:27:26
Do you ever think about a program where an actual live human just follows around the person in the program and is effectively there 24/7 kind of like effectively a nanny for the human because it's like hey like you say you're doing everything right but like right before bed you were [laughter] you had a phone just blasting light in your face.
1:27:46
Uh it it feels like that's kind of what you can get to with wearables over time and you can probably get there to get today.
1:27:52
But I think a lot of people even if they do a program will still be kind of sneaking little things in throughout the day that are that are working against them >> and they even forget you know they're like I do everything you say like I you know I finished my final meal of the day four hours before bed and blah blah blah and I can't sleep.
1:28:07
>> You are hiding you are hiding bright lights before bedtime. [laughter] >> Yeah. Exactly. I can't sleep.
1:28:10
Like what do you do before bed?
1:28:12
Well I'm in my bed like you know scrolling Tik Tok and >> Okay.
1:28:15
Well, that that could be the source.
1:28:18
>> Uh, what's the current thinking on peptides?
1:28:19
There's a variety of them.
1:28:19
Uh, simaglutide all the way down to rea.
1:28:23
Uh, what's the most upto-date that like evidence or benefits cost that you've seen?
1:28:31
>> I mean, generally peptides are fantastic.
1:28:32
They are drugs, drug equivalent, drug like effect sizes and some peptides have done the clinical work like Deptide and >> but then there's a whole bunch of the class of peptides that we have no human data on >> really.
1:28:44
And yes, so the the things >> that people are taking a lot of, right? >> A lot of Yeah.
1:28:48
>> So we're going [clears throat] to get some human data.
1:28:50
>> And is that [laughter] is that currently Retta?
1:28:51
Is that the one that's like not >> I don't know about I don't know about Retta, but >> or there's like different ones like that.
1:28:56
I'm not even thinking of.
1:28:58
>> There's a lot of Yeah. >> Okay.
1:28:58
There's a lot there's like a long tale of like all sorts of slightly different slight one molecule change or something.
1:29:03
>> And it's currently I it's all the rage. Yeah.
1:29:05
Uh the >> the Chinese peptide thing is a whole meme in San Francisco. >> It's a whole thing. Yeah.
1:29:10
And so my take on this is is if you if you do it like you know be cautious because >> we do not have them characterized like when when you hear about a drug that does blood pressure control >> you hear all the benefits and then you hear like in the commercial you hear like this may cause death or kidney damage or like you know like you hear all the side effects because it's been well characterized.
1:29:31
So it's been it's gone through clinical trials you understand the pros and the cons and you make that trade-off.
1:29:37
>> With peptides you don't know. >> Yeah.
1:29:38
It's like your your bro buddy's like it does this amazing thing and it fain statement after the fact of like hey by the way this could do this terrible thing. >> Yeah.
1:29:49
So if you're doing it proceed with caution because they're not characterized. We don't know.
1:29:52
It's not to say they couldn't have benefits.
1:29:53
It's that you don't know.
1:29:55
And that's the >> the worst thing in health is the blind spots.
1:29:59
People want to chase the positives but rarely adhere to like you know there could be some downsides. So keep costly. >> Yeah.
1:30:05
>> Wasn't that the name of Marty Moore's book? Blind spots.
1:30:07
the new FDA commissioner.
1:30:09
I'm pretty sure that's what he called it.
1:30:10
Um, do you do you have confidence in the FDA right now?
1:30:12
Do you have confidence in the FDA's ability to actually collect all the data and make good recommendations?
1:30:17
Because there's some things where I I totally agree with you.
1:30:21
I'd much rather take the one that's gone through FDA trials, been approved.
1:30:24
It tells me the downsides, but at the same time, people are like, "But I also got all this sugar in my cereal for years."
1:30:30
Like, what were they asleep at the wheel back then? Like, what's going on? >> Yeah.
1:30:33
I'm I'm guessing that we like the FDA has a very valuable purpose in society.
1:30:39
They do like they do good work.
1:30:41
>> Um I'm guessing >> they stopped the literal snake oil famously. >> Exactly. Yes, they did. Yeah.
1:30:44
Like they've they've had a positive role.
1:30:45
I think that we will um exceed them in value in many core areas. >> Okay.
1:30:52
>> Because they you they can take you know well structured wellunded clinical trials but then other things like peptides that are very promising >> they're not going to have the clinical work. So what do you do?
1:31:01
So if we could get, you know, a certain number of people on the platform who are doing peptides, great.
1:31:05
And if they could start adding dimensionality to their their bioction, we could have the most valuable data set.
1:31:10
So it's it's immortals is the product.
1:31:12
That's the high-end right now, but then also the free version we're building.
1:31:16
>> And our goal is to create the world's largest data set. >> Yeah.
1:31:19
>> Uh across all these actors and then you basically can create an ecosystem where the person could be paid for their data.
1:31:25
So they collect the data, they're starting to experimentation, >> companies come in and buy the data, they give out.
1:31:30
So, we're trying to create this >> financially incentivized ecosystem where like in the world right now, >> you can't go anywhere without >> somebody or something trying to take advantage of you.
1:31:41
>> Like, it's it's just everywhere.
1:31:41
Like, try to think of a company that genuinely acts in your best interest.
1:31:46
Like, unequivocally always shows up to act in your best interest. >> Diet Coke.
1:31:52
[laughter] >> It's always there when John needs it.
1:31:56
Except Except when the flight attendant says, "How about Pepsi?" Yeah, it's true.
1:32:00
>> So, we're we're going to try to stand in that role like where like we're also jaded because everyone's trying to take advantage of us.
1:32:05
So, we're going to try to stand in that role that you can unequivocally trust us.
1:32:07
We will always act in your best interest.
1:32:09
And if we do that, then like I'm willing to share my data.
1:32:12
I'm willing to engage in the platform and do the exploration. >> It's a high bar.
1:32:15
Uh I want to run through kind of a lightning round of questions.
1:32:19
What have you learned from Ray Pete?
1:32:22
>> Uh that there's a lot of Ray Peters on X. >> Yeah.
1:32:24
[laughter] And they really want um >> you to follow their >> Yeah, they do.
1:32:30
>> They want you to pee out. >> They do. Yeah.
1:32:31
So they they're definitely evangelist for So Ray Pete Ray Pete has a lot of evangelists.
1:32:36
>> Carrots underrated, overrated. >> Yeah.
1:32:38
Uh carrots are a great food, you know.
1:32:40
They they're they're not benign like no food is benign. It has Yeah. pros and cons. >> Thanks.
1:32:45
>> Uh >> what else you got, Jordy?
1:32:46
Uh, if where would you live if you weren't a public figure >> and you could just optimize for health?
1:32:55
>> Yeah, you know, Kate's here. Hey, Kate. >> Hey, Kate.
1:32:59
>> Yeah, we were just having this conversation. She's from Australia.
1:33:01
Uh, we were just having this conversation trying to figure that out.
1:33:04
It's it's really an interesting question. Um, >> yeah. >> Yeah.
1:33:09
>> LA is certainly not the place if you're purely optimizing for health. >> Yeah. Yeah. It that's very true.
1:33:12
So like it has been top of mind for us and we are trying to figure out like what if we did something uh next level where would we go and how we do it?
1:33:20
I wouldn't have an answer yet. >> Yeah.
1:33:22
Uh yeah I think it's an important question.
1:33:25
There's data on golf courses.
1:33:27
I don't know if you've seen this.
1:33:27
If you live near a golf course it like reduces your life expectancy because they use such heavy pesticides >> on the golf course because golf course.
1:33:37
>> There's one in my town.
1:33:39
>> I'm like >> that's probably not great.
1:33:39
These studies were done on like people that live in like a golf.
1:33:42
So like you're living on the golf course and they're they're basically, you know, golfers just want it to look pretty.
1:33:48
They're not asking like, "Hey, what did you do to make it look so pretty?" >> How are you golfing?
1:33:54
>> Uh what are your health long-term kind of concerns around the health of us being in this studio [laughter] 11 to 2 >> rubber smell? >> Yeah, you smell it. >> Yes.
1:34:04
So that is >> I think it's these things.
1:34:06
I love that question so much.
1:34:08
I would love [snorts] >> and I' and I've thought about it a lot and I I would love to work on this because like we have an LED light right here.
1:34:14
So, we're laughing in our head.
1:34:14
I don't know about these lights.
1:34:17
>> Maybe you can consult on our next on our next we're moving soon.
1:34:19
So, >> I would love to come in here and do a comprehensive analysis. >> Oh my god.
1:34:24
>> No, it would be so good.
1:34:25
>> No, the air the air quality.
1:34:25
I mean, we spend a lot of time in here and we plan to spend a lot >> going to come back and say you guys have 10.
1:34:31
We're going to make 10 years.
1:34:31
You make nine, eight, seven.
1:34:32
[laughter] >> I mean, first of all, your CO2 levels for sure are too high. Sure.
1:34:39
>> I can feel the CO2, right?
1:34:39
Do you Kate, do you feel the CO2?
1:34:42
>> Yeah, CO2 meter for sure. >> Yeah.
1:34:44
So, the the space is >> small.
1:34:46
I have those in my Yeah, I have that uh meter in my house.
1:34:47
I'll move around to the different rooms to kind of track it.
1:34:51
And I've got a whole like I I've done a bunch of work to to the HVAC system.
1:34:56
>> And see, over over a thousand is going to impair your thinking. >> Totally. Yes. That's obvious.
1:35:00
You watch it like you're like these guys takes like >> I like sunglasses right [laughter] out.
1:35:08
Um, uh, what do you think, uh, how are you thinking about parasites?
1:35:12
There's tons of health influencers that will make parasites their whole thing and that like every the solution to every problem is like got to fix a parasite.
1:35:21
>> Uh, how how are you thinking about them?
1:35:23
>> We I've not done a deep dive into that. I don't know.
1:35:26
>> So, these are people who have parasites and they're trying to rid of them or >> there the the argument is everyone has them. >> Yeah.
1:35:31
some amount are probably helpful because they'll like eat heavy metals and then you'll eventually kind of Yeah.
1:35:39
uh release them, things like that.
1:35:39
But you should do a deep dive on that.
1:35:43
>> Um do you think uh how do you cycle supplements?
1:35:45
Right now I'm taking atheenine, glycine, and a and a number of other things before bed.
1:35:49
It's been super effective at getting REM and deep sleep up for me.
1:35:54
But uh I'm curious when you find a supplement that is working and it's like measured in my case, how should you cycle it?
1:36:03
How do you think about cycling it to make sure that you um continue to get the benefit?
1:36:09
>> Yeah, this is one of the questions we're asking the inference engine is because we've done a whole bunch of these different cycles >> and the it's difficult because uh it's never truly a um a non-confounded experiment, right?
1:36:22
always some compound >> and so when you're trying to tease out little tiny signal you do need to be robust about the analysis.
1:36:27
So we don't know like I you know we have like some short-term data on that but it's not not strong enough that I think I'd express an opinion.
1:36:34
So we're hoping hopeful that the bigger data sets will give us insight. >> Yeah.
1:36:38
Uh same kind of question on fasting.
1:36:41
I've started I inverted the fasting schedule because we do the show.
1:36:46
We'll have like a big breakfast at 8 am.
1:36:48
We'll have I'll have a second meal at around 2:30 and then I won't eat until the next day at 8 am.
1:36:52
So, typically somebody might start eating at 12, have another meal around dinner and then wait until the next day. So, I flipped it.
1:36:59
Uh any any uh learnings from that?
1:37:05
Personally, >> the the marker I'd look for that probably is >> and and to be clear, part part of the reason is I actually there's a functional reason which is that the way that we do the show, I need to be like >> fueled up to then podcast for 3 hours. >> Yeah.
1:37:18
>> Uh but then I also don't like being having a big dinner and going to bed. >> Yeah.
1:37:24
You guys work out in the morning together, is that right? >> Yeah. >> 6:30. >> So, yeah. At 6:30, huge meal.
1:37:30
>> Uh huge meal at [clears throat] 8. >> 8. >> Yeah. 8 to 9. >> Yeah.
1:37:33
Anyway, uh thank you so much for coming on the show. This is fantastic.
1:37:38
>> Lots more to talk about.
1:37:39
>> Yeah, we could go all night all day.
1:37:40
>> Let me tell you about MongoDB choose this flexibility and scale with best-in-class embedding models and rerankers.
1:37:47
MongoDB has what you need to build what's next.
1:37:49
And without further ado, we have Matthew Zaitelland, the correspondent from Heat Map News in the Reream waiting room.
1:37:57
>> Let's give it up for correspondence.
1:37:58
>> I'll tell you about Figma while we wait for him.
1:38:01
Figma make isn't your average vibe coding tool.
1:38:03
It lives [music] in Figma so outputs look good, feel real, and stay connected to how teams build.
1:38:08
Create codeback prototypes fast.
1:38:08
Tyler Tyler tested the CO2. We're at 790.
1:38:14
>> We're at 810 right now. 810 and rising.
1:38:17
>> You caught us on a low. It's going up. >> He said, thousand.
1:38:19
He said I was It was It was good.
1:38:22
>> To see what it was at the end of the end of the show. >> Okay. Okay.
1:38:24
Cuz we will add more CO2 as the show goes on.
1:38:28
>> Edge in the chat says, "Jordy has three days in one.
1:38:30
He's figured out how to manipulate time.
1:38:32
[laughter] >> You are manipulating time.
1:38:33
Uh anyway, we have our next guest in the reream waiting room.
1:38:37
Let's bring him into the TVP and Ultradome. How are you doing?
1:38:40
Sorry to keep you waiting. >> What's happening?
1:38:42
>> I'm doing great, guys.
1:38:42
Any any opportunity to follow up Ryan Johnson is is one I will take.
1:38:47
>> It is a fun It is a fun story. He was your opener.
1:38:50
>> I mean, you've clearly been doing a lot of looks maxing.
1:38:51
Uh what's working for you?
1:38:53
You know, I I do have friends actually John and I uh play a poker game with a Ray Pete Ray Pete acolyte. So, >> okay.
1:39:02
>> I've been at least hearing about Peting, you know, forever >> early.
1:39:06
>> I've been I I immediately resonated with the work of Ray Pete. >> Yeah.
1:39:12
>> Have followed a lot of it myself.
1:39:12
I'm I I think it would absolutely >> astonish and scare Brian [laughter] Johnson how much sugar I have.
1:39:19
that that's kind of one one of the main >> takeaways from Ray Pete's work is that sugar is is uh not nearly as bad as many people would say and and may may be actually great >> in a lot of ways.
1:39:31
So, I' I've been running that experiment. >> We'll find out.
1:39:34
>> But I have a >> I have a 2-year-old at home, so uh I can't eat cookies around him, you know.
1:39:40
I can't >> It's a It's you know, then he starts asking for them and it's getting in the way of my peing.
1:39:44
Actually, >> I had a friend who had had a bunch of and wound up just obsessed with dark chocolate because it was the only thing that was at all sweet that he could eat that the kids wouldn't immediately go for and consume all of >> the the explaining to my four-year-old why I can have cookies before bed, but he [laughter] can't is not good. >> That's a tough one. That's a tough one. >> Yeah.
1:40:04
I'm still waiting for that level of cognitive development from uh from my son.
1:40:08
We're not really at the explaining stage yet. >> Yeah. Yeah.
1:40:11
>> Well, uh >> anyways, it's great.
1:40:12
It's great to have you on the show.
1:40:14
Uh I've we've we've covered a bunch of your posts over the last uh 6 or 12 months and always always enjoyed them.
1:40:20
Why don't why don't you give a quick intro on yourself and then you know we can talk about all the stuff that you talk about. >> Yeah. >> Yeah. My name's uh Matt Zelin.
1:40:27
I'm a reporter at Heat Map.
1:40:29
We're a climate and energy focused newsroom.
1:40:31
We report on climate and energy every day and uh I mostly write about you know uh energy policy generation a lot of data centers.
1:40:40
You know we've kind of really reoriented what we're doing around data centers.
1:40:44
That's where all the demand is coming from. >> Yeah.
1:40:45
When did that reorientation actually kick off?
1:40:47
Because it felt like it became a story sort of mid last year.
1:40:54
But have you been tracking it longer?
1:40:55
Like when was the inflection point?
1:40:58
>> So I think around like uh summer of 2023.
1:41:02
>> Um I wrote a story about how just load growth was coming >> that we'd essentially had flat electricity demand uh for since in the 21st century. Yeah.
1:41:10
And when I wrote the story, it was more about electrification of cars and like home heating, you know, heat pumps, stuff like that, stuff that climate people are really interested in.
1:41:20
And then >> and data centers to some extent also factories were something I was really looking into as a big electricity demand.
1:41:27
>> And then yeah, it's I mean it was since the growth of G it was since the launch of GBT3, it became pretty clear that data centers was was where it was at as far as the load growth story was. >> Yeah.
1:41:35
And then how uh like how widespread is the focus of AI data center energy consumption?
1:41:44
Because it feels like if it's going up next to your house, you're going to be calling your representative.
1:41:49
But for a lot of people, it's abstract.
1:41:51
Uh they maybe see an Instagram reel, some slop, and they're like, I don't like it and I don't want my rates to go up.
1:41:59
>> We've been kind of wrestling with this because uh in general, I think AI is very good.
1:42:04
there's a lot to be excited about.
1:42:06
I think that uh more intelligence having, you know, people globally being able to access intelligence, even if they're accessing a little bit of slop is like genuinely good.
1:42:14
But at the same time, I understand the people that it's like, what's the what's the benefit of having a data center in your backyard?
1:42:22
It's like actually probably close to zero because or negative because it doesn't I don't need the data center in my backyard. It's not exactly pretty.
1:42:31
We're not making them look like cathedrals yet.
1:42:32
uh maybe we should uh and I can just if it if it's out somewhere that I've never even been to I can still get all the benefit of it and so um so yeah I think these are things that the whole industry is wrestling with politicians are wrestling with and and for good reason. Yeah.
1:42:47
I mean, so the way we kind of I like to look at electricity prices.
1:42:51
That's kind of a nice solid thing to focus on.
1:42:52
And obviously they've been up in a lot of parts of the country.
1:42:56
You know, you guys are in California. I grew up there.
1:42:58
Um >> I hope you don't have a pool at home.
1:43:02
You know, >> anything around heating or something like that, a pool or something like that, a disaster as far as prices go.
1:43:09
>> Um but no, where we've seen electricity, >> you're saying we got to replace the pools with data centers. [laughter] >> Exactly. Exactly.
1:43:14
You need you need a >> you need a Blackwell rack, you know, [clears throat] >> or just a stack of Mac minis.
1:43:22
>> Mac John can send you over a Mac Mini. >> Yeah.
1:43:25
>> But no, um, where we've seen kind of electricity price increases, California has been a big one and the Northeast have been big.
1:43:30
Those are not areas with lots of data center development.
1:43:32
I think in part because the prices are so high.
1:43:37
>> Um, where we have seen both data center development and price increases has been kind of in the Mid-Atlantic uh, Midwest areas. This is Virginia. This is Ohio. This is Indiana.
1:43:46
This is areas um where if they're all in one electricity market called the PJM interconnection. >> Okay.
1:43:52
>> And the way this market works is that um these uh utilities have to procure capacity in advance.
1:43:59
>> And that market, that capacity market and and the ability to get new generation online there has been completely messed up.
1:44:05
And so we're seeing these auctions generate billions of dollars of revenue uh for the existing generators.
1:44:12
And that is directing that is translating pretty directly into higher prices.
1:44:15
New Jersey has probably seen their prices residential prices increase you know say 20%.
1:44:21
That's because they're in a system that is not incorporating the data center demand that well.
1:44:24
But there are other parts of the country where electricity prices you know haven't gone up.
1:44:30
>> How have you as much how have you been reacting to some of the big AI companies some of the hyperscalers uh start to put out statements saying that they will pay above market rates subsidize energy.
1:44:38
Do you think that that's going to be effective?
1:44:43
Can you just do that and then keep the prices stable?
1:44:46
Also, a little bit of grid history here would be helpful because it feels very intuitive.
1:44:53
But I believe like we sort of have a common carrier rule where even if if I'm doing, you know, research and you're watching Netflix, like we still pay the same price, we don't put like a moral weight that I'm aware of, but any of that would be helpful. >> Yeah.
1:45:07
So typically, you know, residential users pay the same kind of kil per kilowatt hour rate.
1:45:10
Sometimes there's, you know, time of use differences, especially in a state like California.
1:45:16
>> Um, but to to get to this point about and then industrial users often pay >> as well.
1:45:20
Um, those would be the data centers.
1:45:22
Uh, what what they mean when data center say they're going to pay for their own electricity, >> it's not just like the literal electricity, the electrons they use, but when you bring on a gigawatt size data center, that's like putting a whole city. Yeah.
1:45:34
you know on on you know a gigawatt can power 800,000 homes.
1:45:39
>> So that's like putting a whole city on the grid.
1:45:40
So this requires a lot of uh upgrades to the whole system a lot of new equipment you have to buy and those typically system costs are spread out to everyone on the system and this is how electricity utility regulation and pricing has worked for like over about a hundred years.
1:45:58
hundred years. uh this guy Samuel Insil who worked for Thomas Edison in in Chicago in I think the 20s basically came up with a version of this idea and um so in theory having new sources of demand over time should lower prices for
1:46:13
everyone because you have a big electricity buyer that can deal with those kind of distributed system costs but in the short run you know they're incurring lots of new costs and new infrastructure development to the system as a whole and so I think what the
1:46:28
Microsoft oft has made a commitment to this in Anthropic uh yesterday did you know it's you can't uh you know do a utility rate case in a press release but what they are committing to I think is that in their electricity rate that they
1:46:42
pay they want to get all of those extra system costs that they're incurring to just be on to them >> what about the other uh the other the other stories and and push back to data center buildout uh how did you process the the whole water risk hype cycle. It
1:46:57
It feels like there was a lot of fear and then there was some more statements from Google, some studies.
1:47:05
How has uh your community uh and you personally like processed that uh narrative?
1:47:13
>> Yeah, so like water this water issue is I think kind of a surprise for a journalist covering this space.
1:47:19
>> It is definitely something that uh that people especially people online are like really interested in.
1:47:23
You know, it's actually pretty easy to track how data centers can affect electricity prices. Yeah.
1:47:28
How data centers can affect water usage in any kind of systemat systemic sense.
1:47:32
It doesn't really show up that much.
1:47:34
The water usage isn't that great, but is definitely something.
1:47:36
>> But the numbers sound great. The numbers sound great.
1:47:39
It's like we're using 100 million gallons.
1:47:44
The typical typical golf course. >> Yeah.
1:47:47
Well, golf courses are extreme, but you know, household usage of water is huge, you know, and if people individual households usage of water over a year sounds >> really, really big. Yeah.
1:47:58
>> Um, I think there's some issues around construction and water usage, but those are obviously more temporary. Sure.
1:48:03
>> Um, >> yeah, but it's definitely something that people are concerned about.
1:48:05
And so, I think these hyperscalers are really going out of their way to kind of talk about being water positive.
1:48:09
they're really kind of leaning in on the water aspect, but I think the electricity aspect is much more where the rubber meets the road and they have to make some commit.
1:48:18
You know, a 1 gawatt data center people think cost about $50 billion$50 billionish dollars.
1:48:24
Obviously, the vast vast majority of that is the chips, but you know, 10 to 15% of the lifetime cost of that is electricity and you're paying all these extra grid costs.
1:48:32
That means with like a one of these really really big data centers that all these companies are rolling out now and they're volunteering to incur what could be hundreds of millions if not billions of quote unquote extra costs but that just might be the price they have to pay to get approval from the state governments to build these facilities. >> Yeah.
1:48:51
>> Who who have been uh some of the craziest beneficiaries of the general energy uh the boom in in sort of energy demand uh due to AI?
1:49:00
any anybody that was just kind of like sitting around a few years ago uh and then you know just kind of like tripped into making like a [laughter] hundred million dollars or a billion dollars due to >> Yeah.
1:49:13
I mean there's two great examples of this that are kind of on the opposite ends of of kind of technological sophistication.
1:49:19
Uh one is Caterpillar um which we obviously associate with uh mining and construction.
1:49:24
You know anyone I think with boys at home probably has some Caterpillar toys.
1:49:27
Oh yeah, they probably identify all all the equipment they use.
1:49:31
But they they also have a business called solar which builds these kind of small turbines, small gas turbines, you know, of a big gas turbine and a power plant is, you know, hundreds of hundreds of megawatts of capacity.
1:49:43
These are a few dozen maybe at the biggest and they're kind of used for like remote power start often in the oil and gas industry or like you know compress on a turbine uh pipeline or something.
1:49:52
This business has gone huge and they're now a huge supplier to data centers because they're so desperate for power.
1:50:00
I mean Elon Musk XAI is a big innovator here of realizing that grid connection studies can last for years but if you just kind of pile up these small inefficient turbines, you can get the power going really really quickly.
1:50:14
Um Caterpillar is a big supplier to Meta in Ohio.
1:50:16
The Socrates project I think is a few hundred megawws and Williams which is actually a pipeline company is kind of assembling all these kind of uh turbines and reciprocating engines which essentially like giant car engines um to power these data centers and so that business has been doing incredibly well.
1:50:35
>> Car engines they run on gasoline, diesel >> uh natural gas. Yes.
1:50:40
Uh what uh what kind of environmental studies have been done?
1:50:43
If a data center sets up, you know, in your backyard and they've got a bunch of these gas turbines and they're running, is that like, you know, if you can, you move away or does it dissipate quickly?
1:50:57
Do has has there been any kind of significant there?
1:51:01
These turbines are less efficient and dirtier than, you know, a very large scale turbine system you would buy from a GE Verna or a Seammens.
1:51:09
I know with with XAI, they were kind of really monkeying around with these rules about how you could only have it up for 364 days there.
1:51:18
By the time the EPA certified it, they were already moving to Mississippi.
1:51:22
I mean these systems were not really designed to kind of be to function as power plants in the same way that like a power plant is but people are so desperate for power that >> they are going to be using them. >> Yeah.
1:51:35
I mean it feels like there's going to be a variety of backlashes, a variety of pendulums swinging.
1:51:41
The hyperscalers were very focused on net zero clean energy.
1:51:47
Now it feels like they're trying to go as fast as possible.
1:51:49
So there's a lot of natural gas, but then they're also still firing up nuclear projects, more wind, more solar.
1:51:54
How do you see the mix like shifting over the next couple of years?
1:51:58
Do you think there's any sort of white >> bill?
1:52:00
The biggest hyperscalers, your Google's and your Microsoft especially and and Meta and Amazon do have these sustainability commitments, these 100% renewable power commitments.
1:52:08
Um they still are procuring lots of clean energy.
1:52:12
Um Google bought a clean energy developer um pattern uh >> for $4.
1:52:19
5 billion at the end of last year.
1:52:21
And what they do is that you know in some cases they get the natural gas turbines and they also buy say solar projects in a similar area.
1:52:27
That's what Meta is right doing right now in Ohio.
1:52:32
But yeah, I mean they do these emiss they do these reports and say Microsoft's emissions have gone up a lot because they're just consuming um a lot more power and they've gotten way more adventurous in the stuff they're willing to invest in because the scale of their power needs have really changed.
1:52:45
So I think all four of the biggest hyperscalers have nuclear projects going on.
1:52:51
Uh Google's a big investor in enhanced geothermal.
1:52:53
Um >> we just talked to Jeff Lawson yesterday. He raised what?
1:52:57
$450 million to do uh not nuclear fision, traditional nuclear, but but fusion, which is >> Yeah, he's doing the he's doing the lasers.
1:53:06
>> Yeah, lasers, which is, you know, he said there's no science risk, and I hope that there's not, but it feels like, you know, it hasn't really produced energy on the grid just yet.
1:53:13
So, it is, you know, very forward-looking.
1:53:15
But Google Ventures is one of the investors because they're and they also have a deal with Commonwealth Fusion Systems and bunch of nuclear projects as well.
1:53:21
Uh, yeah, that's fascinating.
1:53:23
Um what can you tell me about the nature of the push back politically against data centers?
1:53:29
I was looking at this data center watch uh stat uh 55% Republican, 45% Democrats in affected districts.
1:53:39
So uh it it feels like I don't want a data center in my backyard is not a left or rightwing issue, but >> No, no, it's a fusion.
1:53:48
It's one of these kind of fusion issues.
1:53:50
We've done a lot of great reporting at this.
1:53:52
We have a pro service called Heatmap Pro that really focuses on this stuff. >> Okay.
1:53:57
>> And what we found is that uh yeah, it's like it's it's anyone who's kind of distrustful of institutions in general uh will tend to really not like a data center in their community.
1:54:08
Also, people who um it also brings together Democrats and Republicans because people who oppose renewable energy projects, which at heat mapap is what we were kind of born to start tracking was local opposition to renewables. >> Interesting.
1:54:21
um the same people are opposing the data centers and they're using the same techniques.
1:54:25
They're opposing say reszoning agricultural land into industrial land.
1:54:29
Something really common with solar panels.
1:54:31
Same thing's going on with data centers.
1:54:33
Same type of people are are opposing it.
1:54:35
And it's interesting because at the at the highest levels of politics, you know, uh the the administration right now is very pro the data center AI buildout.
1:54:44
You know, they have, you know, they have Sam Alman at the White House announcing, you know, multi-billion dollar investments in this stuff. Yeah.
1:54:53
>> And yet, >> a lot of data centers are built in Republican areas.
1:54:56
That's typically where there's enough space.
1:54:57
They tend to kind of be in exurban areas.
1:55:01
>> Um, and yeah, they get a lot of local opposition.
1:55:02
There's been a lot of local opposition in Indiana, a lot of local opposition in Kentucky. >> Yep. >> Virginia.
1:55:07
This is really kind of transforming the politics around there as well.
1:55:11
>> So, OpenAI has a pack now.
1:55:11
Philanthropic has a pack now.
1:55:15
I think uh for the drama heads in the chat, we hope that they're just going to fight with each other endlessly and make content for us.
1:55:22
But assuming that they want to bring about support, broad support for their data center buildout efforts, what would you recommend?
1:55:32
What's the playbook for an AI lab that wants to build a gigawatt data center somewhere in America? Where are they going?
1:55:39
What promises are they making?
1:55:41
What promises can they keep?
1:55:44
What would you recommend?
1:55:46
>> Yeah, I mean we see this in the kind of renewable and nuclear space a lot is essentially you do a lot of kind of expensive extensive long-lasting local engagement >> with the community and like look if you're a data center developer >> you're going to end up you know building a lot of high school gyms.
1:56:01
a lot of high school gyms. um you're going to really have to get the local people there on board with your project because typically at the state level you actually governors like data centers you know >> they provide tax revenue um unions like data centers the IBW national international brotherhood electrical workers very pro data center and they're
1:56:22
pro data center force in the Democratic party but often times these decisions are made at these kind of community board meetings zoning meetings >> and for them you really have to convince that you're providing something on the Because what was saying at the very very beginning, there's no benefit in terms of what the data center produces to having it near you. >> Yeah. There's no local AI movement. I >> Yeah.
1:56:40
There's no local AI movement.
1:56:40
I want my >> for yourself.
1:56:43
I want low latency inference.
1:56:45
I want it in my backyard. I'm a Yimi.
1:56:47
Give me a massive data center.
1:56:47
I was hearing bad stuff about living next to a golf course.
1:56:51
Tee up a one gigawatt data center right there. I'm good to go.
1:56:55
>> My locally I like the idea of drone inference.
1:56:57
I like the idea of showing up to the local uh te-ball league and being like, "Why is the team called the Core Weavers or something like that?"
1:57:06
>> So, nuclear power plants famously do this. >> Yeah.
1:57:09
I mean, the Simpsons parodyied it for, you know, you have the isotopes or whatever.
1:57:13
>> And you do see this in communities of nuclear power plants.
1:57:15
Um, >> but the thing about data centers, they don't provide very many jobs after they're constructed.
1:57:21
>> So, that that part is very difficult to get over.
1:57:24
>> I have Wait, I have the solution. Sorry to interrupt.
1:57:25
The solution is you get they have the data center and then there's just a big room with a bunch of pelatons and you can just go in there and just generate energy for the data center and you get paid that way.
1:57:38
So we get a gym jobs and a data center in one space >> be good.
1:57:45
>> I mean I I think uh I think Brian Johnson for one might support that.
1:57:48
Although you know >> as we know with the 49ers they don't like being by an electrical substation when they practice.
1:57:53
They're worried about the injuries from the EMF.
1:57:55
So, that might be that might be a tough cell. >> Yikes.
1:57:59
Demystify New York for me.
1:57:59
What's it like living there? >> Oh, I love it.
1:58:03
I, you know, I've been here for over for over a decade. You know, it's back.
1:58:07
Uh, it's too cold right now.
1:58:10
There's too much snow, which is quite weird.
1:58:11
Uh, I I cover climate change.
1:58:14
Um, where is the so-called global warming right now? >> Who knows?
1:58:18
Um, but no, it's it's uh it's great.
1:58:21
People are on the subway again.
1:58:23
Um really nothing to complain about.
1:58:23
You know, apparently uh bonuses on Wall Street were great in last year.
1:58:27
Our projected budget deficit almost got cut in half magically in a few weeks.
1:58:31
So >> what could you complain about? >> There you go.
1:58:35
What was your review of Davos from an energy perspective?
1:58:36
From a tech perspective, it felt like oh like Davos is like a real place that needs to be taken seriously.
1:58:44
Like every company should be figuring out their Davos strategy for next year.
1:58:48
Whereas like previous years were sort of quiet. >> Yeah.
1:58:51
I mean, Davos used to be the place where um they would you would you would hear the most bogus sustainability commitments, you know, and uh for better or worse, they kind of stopped doing that.
1:59:02
I think uh as it's gotten more serious, uh >> they've they've stopped, you know, you know, you're not hearing that you're going to own nothing or be and be happy anymore.
1:59:13
So, >> it's just a more it just seems more serious.
1:59:16
And uh you know as a as obviously as a newsroom and a reporter that's focused on climate change like I have a point of view on this but also like I just like seeing people deal in reality >> and I think Davos is more of a reality based place now which I as a reporter sense is great. >> That's good. Yeah.
1:59:32
Jordan, you have something?
1:59:34
How uh you said something which is funny because you cover you cover climate change like professionally, but you said there's snow out which is like we're supposed to have climate change.
1:59:45
Doesn't feel like we have climate change.
1:59:47
Um but like how how is kind of the broader clean energy industry even processing like how are they processing that?
1:59:59
Because that that is kind of like more like the average viewpoint of Americans.
2:00:04
like it's cold this winter, no climate change here.
2:00:06
And that's obviously like, you know, you can have, you know, different spikes and and your kind of like lived experience is not necessarily entirely representative of kind of >> but it might inform who you vote for, what you vote for. >> Yeah.
2:00:18
I mean, one of the great ironies of climate policy in the United States is that some of the strictest uh decarbonization standards and rules and restrictions uh are in the Northeast.
2:00:29
>> Um Maine, Massachusetts, New Hampshire, Vermont. >> Okay.
2:00:32
uh they're really into decarbonization.
2:00:33
They're really into preventing climate change.
2:00:36
>> Uh and then in Florida, they don't really seem to care that much at all.
2:00:39
And obviously that's the place that's most at risk in the United States from high temperatures and rising sea levels.
2:00:43
>> So yeah, I mean just an issue where uh I mean this is something that people in the industry complain about a lot.
2:00:46
Well, now they're complaining about it.
2:00:50
>> This stuff is super politicized.
2:00:52
>> And you know, I think a lot of people in the industry are when a Republican is president really wanted to say like, "Oh, this is not about politics.
2:00:56
This is not about climate change.
2:00:59
just is about clean, fast, cheap energy.
2:01:01
You know, solar is the thing we can get on the grid the quickest.
2:01:05
Solar and storage we can get on the grid the quickest.
2:01:09
>> But yeah, I mean it's funny.
2:01:09
The biggest um the most successful green tech entrepreneur and the biggest solar maxi in human history >> uh is Elon Musk.
2:01:16
He's also the second or third biggest Republican donor of all time.
2:01:21
the Republican party has tried to dismantle a lot of tax benefits and programs that benefit Tesla and he doesn't seem to care that much.
2:01:30
>> So yeah, I mean people's politics are really kind of often override >> everything else and that's that's not obviously that's not something unique to climate change but it's a huge challenge for these businesses. >> Yeah.
2:01:40
Uh what are you tracking on the solar panel manufacturing supply chain?
2:01:46
Uh we we we talked to the CEO of T1 Energy.
2:01:49
Energy. seems like there's some green shoots but even his optimistic case was oh maybe we get to like 1% production and relative to China >> I mean yeah China has uh solar and batteries >> uh the amount of kind of time energy and money that has been devoted to that has
2:02:07
been immense and every time it's a it's a big balloon squeezing effort too I'm sure T1 and first solar is the other big American solar manufacturer can talk about this we try to you know put tariffs on Chinese-made solar These factories start popping up everywhere. They pop up Malaysia,
2:02:21
They pop up Malaysia, Cambodia, Thailand, Indonesia.
2:02:25
>> And then we start putting tariffs on those factories because they're owned by China. >> But it's tough.
2:02:28
I mean, they're the cheapest.
2:02:29
They're the cheapest panels.
2:02:31
They get done the fastest.
2:02:31
And yeah, these domestic solar manufacturers are really >> similar to the car companies, XTesla, they're very dependent on government policy.
2:02:41
You know, the developers, if they could, would just buy Chinese solar panels. >> Yeah.
2:02:45
But um you know there's a lot of political reasons why they're being encouraged and now because of these tariffs. >> Yeah. >> To buy American ones.
2:02:54
>> We have a friend Casey Hammer who says like hey look if they're going to be subsidizing them we need to buy by by buy every solar panel we can.
2:02:59
And like that's I I talked to Casey for stories too and he he is a mind expanding person >> to talk to.
2:03:09
I mean, yeah, but it's it's true cuz it's just tough, you know, uh if you're trying to get everything onto solar and batteries, you don't want to then disemploy an American supply chain, especially because with energy right now, American energy is largely domestically produced, which is a huge change >> from, you know, 10 to from, you know, 15 or so years ago.
2:03:30
And so if you're trying to have a a climate policy that encourages solar, you you don't want to then also disemploy, you know, lots of people.
2:03:41
>> Uh zoom out a little bit for me and and and tell me your prediction or or feeling around like long-term energy production capacity in America.
2:03:49
It's been flat and growing very low rates.
2:03:54
Nowhere near the roll out of phones or electric cars or compute like everything else is growing exponentially.
2:04:01
It feels like you're you're going to watch a you know a sharply uh exponential line crash into a linear line.
2:04:08
What is your feeling around this idea that we might actually start seeing 10% growth rates in energy production nationally?
2:04:19
I mean, just look at how hard it is to build a data center, which is just a warehouse with computers in it.
2:04:23
And then think about building power plants on that same scale.
2:04:28
I mean, people have a lot of trouble building solar power, which is zero emissions and it's fine. It just looks weird.
2:04:35
>> Imagine trying to do that with gas with gas power plants.
2:04:37
So, >> what if I billion AI lawyers in my college [laughter] just suing everyone left and right?
2:04:42
I show up into town, I sue everyone, [laughter] make friends.
2:04:48
>> Well, I mean, this is a serious concern.
2:04:50
You know, litigation risk is a huge issue with energy projects, >> and one can only imagine what Harvey could do with filing with, you know, gumming up the NEPO process. Oh, yeah. For infrastructure. >> Yeah. Well, >> yeah.
2:05:02
So, I think it's more like >> we're kind of growing at half a percent a year. >> Yeah.
2:05:08
>> Maybe two, maybe 5% some years. 10 would be tough. >> Okay.
2:05:12
Well, that's the next bottleneck after the chip bottle.
2:05:15
>> Well, I have something.
2:05:15
So, your friend John Palmer is also our friend.
2:05:18
We read his fantastic piece.
2:05:21
To me, it was more impactful than something big is coming, something little is coming.
2:05:27
>> It hit me like a ton of bricks.
2:05:29
>> It really hit it really hit us hard.
2:05:29
But he just announced that Stripe is acquiring >> his company. Oh, no way. >> Party Dowser.
2:05:34
I wanted to ask, can we hit the gong with you for our friend John? >> Be an honor. Oh. Congratulations.
2:05:44
>> Congratulations, John.
2:05:44
Uh, this means you have to start buying in bigger, more frequently.
2:05:50
Uh, [laughter] >> yeah, >> you have to call my You have to call my worst bets. No more excuses. >> Poker game. You Texas Holde.
2:05:59
>> Uh, no limit Texas hold. Yeah, >> no limit Texas.
2:06:02
>> We're actually thinking that maybe the guys we could get together and start a podcast.
2:06:05
>> I think we call it Allin.
2:06:05
I think it's probably kind of like a poker themed, a little talk about a little business, a little politics, a little tech, energy, geopolitics.
2:06:14
I think I think you guys would have something there. >> Yeah, I know.
2:06:18
We'd wear expensive sweaters and Yeah, it' be great.
2:06:21
>> I I I would be the first one to subscribe.
2:06:24
>> Uh well, it's so so great to finally have you on.
2:06:26
Come back on when you when you're going to publish your next story.
2:06:30
>> We'll talk to you soon. >> Thanks so much. >> Have a good one. Goodbye.
2:06:34
>> Let me tell you about Plaid.
2:06:34
Plaid powers the apps use to spend, save, borrow, and invest securely connecting bank accounts to move money, fight fraud, and improve lending with now with AI.
2:06:43
Uh, is there any other breaking news, Jordy?
2:06:45
Anything we need to talk about before?
2:06:48
>> Super I'm super excited for John Palmer and the Party Dow team.
2:06:50
Uh, it's funny, John and I, uh, didn't meet until last year, >> but we used to kind of have this like tension because >> I had a company called Party Round. He had Party Dow.
2:07:02
We're both in that 2021 era.
2:07:05
I was doing >> sort of a team of rival situation.
2:07:09
>> Crypto kind of stunts with party round.
2:07:12
>> He was he was getting quite a lot of attention.
2:07:14
We both raised from Andrees. >> Oh, okay.
2:07:16
>> Uh but uh now we're now we're now we're buddies and very happy for the whole team >> turned business and very bullish for Stripe Crypto >> for sure.
2:07:25
And that's why he's probably on Cloud9 writing long form jokes which is great.
2:07:29
You're like, [laughter] >> "Oh yeah, >> shocking the internet." shocking me.
2:07:34
>> Uh well >> well you know what time it is?
2:07:34
It's time for the lightning round.
2:07:36
The lambda lightning round. That's right.
2:07:38
Bring down the second gong mallet.
2:07:40
Fire up the lambda lightning round.
2:07:45
>> Let's bring in >> I'm going to tell you about Vanta automate compliance and security.
2:07:47
Vanta is the leading AI trust management [music] platform.
2:07:50
And we will bring in June Park, the co-founder and CEO of Similey.
2:07:55
Welcome to the show June. How are you doing? >> What up? Hi everyone. Great to be here. >> Welcome to the show.
2:08:00
Dude, you know you're you mean serious business because >> behind you. I love the whiteboard.
2:08:05
Uh is incredibly detailed.
2:08:08
Please uh first time in the show. Introduce yourself. Introduce the company. >> Yeah.
2:08:12
So really excited to be here. Uh I'm Jun. I'm the CEO of Simile.
2:08:17
>> Simile is a company that is building foundational model of human behavior >> that can simulate our society from ground up at the level of individuals and populations and predict its outcomes. >> Interesting.
2:08:27
We use that technology to create products that enable people to talk to millions of simileies we call agents. Yeah.
2:08:33
>> Uh that represent real people. >> Yeah.
2:08:36
>> Uh I'm super excited about this.
2:08:36
Uh when uh when we were working to get you on the show, I was thinking about it.
2:08:42
There's been so many times in my career where I I was doing something like launching a product or launching a campaign or talking with a founder that was doing one of those things.
2:08:52
And like often times you have a really strong sense of how something will do after you've built up some intuition.
2:08:57
But I've always thought it was crazy that for some, you know, think of like a physical product company, like you have to spend a year and all this time and money uh and then just like launch it and be like, well, I hope I hope people like it.
2:09:11
And it feels like with enough uh data and maybe a model like you're building it similly, you could get to the point where like you can get a much more accurate read on how society or a niche will respond to something without having to actually make the thing. >> Yeah.
2:09:27
What was the first thing you built for the company?
2:09:31
>> So for the company, we work with vendors uh that actually connect us with real human subjects.
2:09:36
So we actually partner with people and get their data.
2:09:38
And what we found was if we have rich data about individuals in our community, we can actually create really high fidelity simulation of these people.
2:09:46
This was on the basis of the research that we've done at Stanford where we led agents and simulations based research.
2:09:51
And what we are now finding is that this technology can actually create insights for our decision makers.
2:09:57
So our core product basically is one that allows our Fortune 500 customers to actually interact with and talk to these agents that represent these real people that we collected >> and basically provide insights.
2:10:08
So that's the platform that we're creating.
2:10:12
>> Uh what's your thought on game engines as a piece of simulation?
2:10:15
I've seen some economic research that maybe used Unreal Engine, but is there a role for uh simulated worlds or deterministic worlds that are then puppeteered by agents uh in any of this or is that just like a different fork of research entirely?
2:10:33
>> No, I think it's all connected.
2:10:33
In fact, we actually started our research when we published the generative agents paper which introduced the idea of creating simulations and agents.
2:10:39
We actually demoed this in a game town.
2:10:41
One because we thought it was so visceral and so fun, >> but also at the same time, if you want to create a simulation of our entire world and how our society might get shaped in the coming decades, then you really ought to be modeling not just people, but also the environment and the model as a whole.
2:11:00
>> So there's absolutely a place uh for these kind of game simulation like characteristics to this field.
2:11:04
Basically the idea here is down the line can we actually create simulations of the entire market. entire nation.
2:11:09
That's where we're headed. >> So, yeah.
2:11:13
What what what are some of the case studies that you've seen with uh large companies?
2:11:16
Is it something as granular as should they charge $2 or $ 250 for a particular product or is it more broad?
2:11:24
Like if we do a Super Bowl ad, will it will it convert or you know or just >> Yeah, Super Super Bowl ads are such a good example where you're spending uh maybe a million uh 8 million on the buy >> uh you know half a million on creative all this stuff and then >> and then uh it'll either get people to hate you or love you.
2:11:43
[laughter] >> But we should be able to predict these things better.
2:11:48
things better. And obviously historically there's focus groups you could try to get a reaction but we're still at this point where sometimes you just need to do something to see >> indeed in fact uh as somebody who's running this company I actually
2:12:03
simulated how today's conversation might go >> really because a few of you are so well known online that we could actually get in there >> to actually see how conversations like this one actually might pan out so I could be better prepared to talk about some of the topics that we wanted to cover. So core use cases today actually
2:12:17
cover. So core use cases today actually does cover things like so we work with fortune 500 companies that have earnings call many of them have come to us and said can you actually create simulations of let's say earnings call that's going to happen and there basically the idea here is can we simulate the each analyst who might actually ask questions during these earnings call the way to how CEO
2:12:40
and CFO might be able to respond to these questions >> that's a common ask so these are the kind of simulations that we create uh for that particular use case and what we find is we can actually predict eight out of 10 questions on average for these earnings score like the actual questions that will be asked and we can actually simulate how the conversation might flow from that point on. So you can test out
2:12:58
So you can test out different strategies. >> Yeah.
2:13:01
>> But beyond that >> one of our sort of customers right now is CVS.
2:13:05
>> CVS of course a large retailer. >> Yeah.
2:13:07
>> And there CVS has been working with Simul to create simulations of hundreds of thousands of their customers. Mhm.
2:13:13
>> What they want to know is let's say CVS has new product, they have new store layout, they have new concepts they want to test.
2:13:19
These are the kind of things that CVS today might go to a large human panels but instead they can now work with Simil to get insights that's much more granular and that's much faster to gain.
2:13:30
>> Now another example here Gallup sort of a longstanding polling company has been working with Simile to create digital panels of their human panel members.
2:13:36
So if they want to ask questions to that panel but for whatever reason they are not reachable then you can actually come to similate to create simulations of that panel. >> Mhm. What's the moat?
2:13:47
How do you think about um like in the earnings call example?
2:13:54
A lot of earnings health transcripts are already in pre-training corpuses and you could dump them in and sort of do a quick fine-tune or even just throw it in the context window.
2:14:02
And if you're a public company CEO, you could probably just ask any of the frontier models.
2:14:08
What do you think the analysts are going to ask?
2:14:10
And is it that that strategy only yields two accurate questions and you're at a higher benchmark rate, or do you think that there's a different piece of your business that the big labs won't go after?
2:14:23
>> So fundamentally, the model that we're trying to create is different than what big labs might be interested in today. Okay?
2:14:28
that we are less interested in creating intelligence that's superhuman, that's super rational, >> but instead what we're interested in creating are models that actually model people's irrational half of their brain.
2:14:40
These are people's values, preferences, taste.
2:14:44
And the way we go about doing this is a true combination of modeling and product frontier.
2:14:49
So, similar is unique in that we as a team actually is composed of both of those talents.
2:14:54
So myself and my co-founders Michael Bernston and President Leang are all researchers who have been making contributions at the frontier of AI ranging from generative agents to generative AI and foundation models >> and there the core mode here is what are the best kind of data that we can collect >> and then using the data to create the most highfidelity agents that represent the people we have interviewed collected data from. >> Mhm.
2:15:21
And then we overlay that on top of amazing product.
2:15:26
>> So here one of the amazing part about the models that actually map people's values and preferences is the better model immediately means better insights for customers like CVS and Gallup.
2:15:37
>> So that there's an amazing alignment between the modeling frontier and product frontier which is not something that we see every day.
2:15:43
>> So it's a true combination of those two things coming together that makes sim special. >> Yeah.
2:15:48
what uh what what do you see progress scale with more?
2:15:51
Are you compute constrained or is it uh much more about data constraints?
2:15:58
>> So we are trotting the frontier on both front.
2:16:01
So down the line so today we model hundreds of thousands of people >> who signed up to participate in these studies >> and then our customers of course are they have appetite to down the line simulate millions.
2:16:13
Our goal eventually is to ask ourselves what would it mean if we can create simulations of eight billion people the entire earth.
2:16:23
>> That's the future that we're headed. >> Yeah.
2:16:25
>> So there certainly there's a question around what is the data? >> Yeah.
2:16:29
>> But also how do we compute these kind of simulations efficiently? >> Yeah. >> Right.
2:16:32
>> Are you worried about getting too accurate that life just be kind of becomes boring because >> is free will is free will real? >> Do we have free will?
2:16:41
>> Are we living in a simulation?
2:16:41
It's like you've read you you screw you went to the end of the book and you read the last page and you're just like ah like [laughter] [snorts] >> yeah there's something fun about actually this discussion.
2:16:52
Um so I don't know if you have watched the Matrix or I don't know how many of the audience actually have watched the Matrix. >> I love the Matrix. >> Amazing. >> I have seen that one.
2:17:02
>> You have seen that one.
2:17:02
Jord hasn't seen a lot of movies but he's seen the Matrix.
2:17:06
>> There's actually a quote that I actually like quite a bit that I think it's actually quite profound.
2:17:09
It's something that the oracle says which basically is you're not necessarily here to make a decision.
2:17:15
You've already made that decision.
2:17:16
You're here to understand the decision and why you made that >> and that sort of captures essence of simulation. Right?
2:17:24
simulation. Right? that the models that we're creating is isn't merely trying to predict the future on the we are good at that and that is one of the core capability but what to me is more important is understanding why we made the decision that we created and the
2:17:38
creating these kind of bottomup simulation where we can actually go back and actually trace through the audit logs of how society might unfold >> is actually an amazing way to gain that interpret interpretability layer of our reality >> so that is to me the best way to shape the future Right? If you want to shape
2:17:55
If you want to shape new policies, new product that would actually serve the people that is in our society, the best way is actually to understand those people >> and then actually communicating with them at a scalable way >> to basically create those policies.
2:18:09
So that's how we view things. How much did you chat?
2:18:13
Uh before before that, the chat wants to know, do you think uh do you think if you worked with uh Clvicular the live streamer, you would be able to predict whether or not he would get frame mogged if he went to different you know locations or or maybe universities like maybe you could have said don't go to ASU just don't just don't.
2:18:31
[laughter] >> How much did you raise?
2:18:35
>> So we raised $100 million uh from index and other participants Uh >> we're also really excited to invite angels uh such as Andre McCarthy. >> They're very excited.
2:18:51
>> Okay, it's great big dogs.
2:18:51
But ultimately what we're excited about is that there's a vision that I think is extremely compelling that this is a vision that this particular team have been working towards for the past five years and that we can create with the technology that we have today.
2:19:04
Really high fidelity simulation >> that can actually provide insights to create better society.
2:19:12
>> Well, thank you so much for taking the time to come chat with us.
2:19:15
Congratulations on the massive round and all the progress and >> great great to meet you. Very exciting >> stuff.
2:19:20
Keep tearing through the Fortune 500.
2:19:23
>> Don't Don't be afraid to send us some alpha.
2:19:24
If you're predicting anything about us, let us know.
2:19:27
>> There's any I like this.
2:19:27
I think you're starting a new meta.
2:19:29
Somebody in the chat said it's kind of the Jensen Steve Jobs combo effect with the outfit. >> I like it. Looks good. >> It's It's great.
2:19:38
>> Well, have a great rest of your day. We'll talk to you soon. >> The team. >> Goodbye. >> Cheers.
2:19:42
>> Let me tell you about Gemini 3 Pro.
2:19:42
It's Google's most intelligent model yet.
2:19:46
state-of-the-art reasoning, next level vibe coding, and deep multimodal understanding.
2:19:50
Up next, we have David Richer.
2:19:52
He's the CEO of Lyft, and we're going to ask him about the Ferrari Luche.
2:19:58
>> Ferrari Luche, yes or no.
2:19:58
Is it the ideal lift mobile? Yes.
2:20:02
You like the Johnny Drive, the Johnny IV inspired interior? You're down? >> I am down.
2:20:08
I mean, I look, I like innovation. I like cool new stuff.
2:20:10
So, why not give it a try?
2:20:12
I've not been in one, though, to be clear.
2:20:13
Yeah, I don't think anyone has.
2:20:15
I mean, they did a they did a PR uh you know uh a day.
2:20:16
They didn't show the exterior of the car, but they did let some journalists come and twist the knobs and >> hear the click pl I'm with you.
2:20:25
I like that there's some fresh ideas in there, and I hope the other manufacturers >> Yeah. >> Uh adapt as well.
2:20:34
>> What's the freshest idea you have for 2026 for Lyft? What's happening?
2:20:36
What's happening in your world? >> Oh, dude, a lot.
2:20:39
Well, so we just announced earnings, which is always kind of an interesting time.
2:20:45
>> So, record bookings, which is great, and accelerating.
2:20:47
Record profits, which is nice as a business.
2:20:52
>> Over a billion dollars in cash generated, which again is nice as a business.
2:20:55
And the only reason I start there is because to me it says like this is customer obsession driving profitable growth, right?
2:21:02
Like I took this job maybe almost three years ago now.
2:21:05
Actually, fun fact, I was sort of convinced to do it on Valentine's Day three years ago.
2:21:09
So, it's coming up right on that magic moment.
2:21:11
My wife and I had such a romantic evening talking about Lyft that night.
2:21:15
But anyway, so the point is like >> that's amazing.
2:21:19
>> It was [laughter] it was it was a thing of beauty.
2:21:21
But anyway, so um yeah, and so here we are super well positioned for 2026. Why?
2:21:26
Because international expansion, that's a fresh thing.
2:21:28
That's a big deal for us.
2:21:31
>> Uh autonomous vehicles, self-driving vehicles, that's a super super um big deal for us.
2:21:35
So we've got a couple of big what we call uh growth vectors in the business. Yeah.
2:21:39
What's the biggest uh what what's the market internationally that's still ripe for ride sharing to come? >> Yeah.
2:21:47
So, it's really interesting.
2:21:47
You know, Europe uh is is an established ride share market of course, but there's still a huge huge taxi market in Europe.
2:21:55
And it's really interesting like if you look at the total mobility like outside of people driving their own cars in Europe, still a ton of people every day pick up the phone and super old school, you know, call the phone number and get a taxi to their house and stuff like that.
2:22:08
So that's actually a big transition thing that's happening that we're really uh kind of helping power is the kind of moving from offline to online.
2:22:13
And then just like all over the world, I mean self-driving cars are going to be a big deal for all of us.
2:22:18
And you know, I look yeah, you know, 3, five, 10 years, you know, it'll be 5% and then 10% and then 20% and then 50%.
2:22:25
Then that's going to be huge.
2:22:25
Um Oh, and bikes are always big in Europe, too. Even bigger than the US.
2:22:30
>> Uh how are your how are your conversations uh like with other public company CEOs?
2:22:35
Right now we're in this period of insane volatility.
2:22:37
Do and and it seems like no matter what uh management team says on an earnings call, you're going to have uh you know uh some some massive volatility, but um yeah, what what are those conversations like? I'm curious.
2:22:51
Do you guys have a little group chat where you're like, I'm going [laughter] in. I'm going in. >> I'm going in. I'm going. Wish me luck.
2:22:58
What's going to happen here soon? >> Uh you know what?
2:23:00
We don't actually have those conversations that much.
2:23:04
just because they're kind of boring, you know?
2:23:07
I mean, it's sort of like everyone has the same level of frustration, of course, where it's sort of like, oh my god, you know, everything seems so volatile and it's so weird.
2:23:13
I was actually just thinking about this yesterday.
2:23:16
yesterday. I mean what happens right now I don't know if you guys know this but what happens is you come out with earnings and then like for example again we had so record um uh you know bookings record rides record profits you know
2:23:27
good go good go good go good go good go good go good go good go good go good go good go good go good go good go good go good go good go good go good go good go good good okay so for like the first 30 seconds headlines say you know whatever
2:23:32
lift beats on profits or you know lift you know highest rides ever something like this and then there's some market reaction now the market reaction might have something to do with some totally different thing you know some weird or cruel thing that some bot remember
2:23:44
happens after hours so some bot picks up some weird cruel thing that doesn't fit in the model and then all of a sudden you know starts selling shares and then you know the sentiment turns and this is all after hours stuff super thinly traded. Okay. So then the headlines get Okay.
2:23:54
So then the headlines get rewritten in real time by other bots who are not reacting to the news of the earnings.
2:24:01
They're reacting to the reaction of the news and that tends to >> and then they're creating so they they see the stock price move and then they're creating a written narrative of why it's happening that then gets turned into a a an article that then other people are reading.
2:24:14
people are reading. 100% and remember a lot of people don't read much beyond the first paragraph and so they'll just see this headline which used to say you know lift beats on profit or lift beats on whatever and now you know lift shares
2:24:25
down you know 10% or whatever it is and so it's such an interesting thing anyway it's a little bit of a meta point but like right now as a society it's like we're almost we're almost reacting to the reaction more than to the thing itself which is which is a little bit mind-bending. >> Yeah that's
2:24:38
>> Yeah that's uh yeah that's super fascinating. We're reacting.
2:24:42
We're we're we're letting the bots react and then deciding how we feel based on based on that.
2:24:48
>> That's that's exactly right.
2:24:48
And the funny thing is the the bots are reacting for their own algorithmic reasons that might not particularly be well plugged into any reality.
2:24:56
Anyway, you can go on this for hours, but it's a little like at the sort of super like species level, we're kind of liking we're kind of like letting the machines take over a little bit.
2:25:06
We're not forming our own opinions.
2:25:08
we're letting them tell us what our opinion should be and then reinforcing that which is a little bit a little counter but anyway like of course some version of this going on for a long time it's just sort of exaggerated now I remember this is something I I worked for as you guys know Jeff Bezos for a
2:25:22
long time and you know Amazon for example back in the 2000s was you know $6 stock or something like this and people would ask him like how do you stay um you know sort of uh sort of focused he's like the good news is like I know the data and this is how I feel too like I was in a meeting yesterday about our self-driving car initiative. It's amazing. And I was in another It's amazing.
2:25:39
And I was in another meeting yesterday about our Super Bowl performance. It was amazing.
2:25:42
We were up 13% year-on-year.
2:25:44
We picked people up faster and we took we were down 20% in surge pricing.
2:25:49
In other words, our prices were were 20% better than they would have been if we'd had our search pricing algorithm from last year, which is great because it means that we're able to do this at a lower cost to to customers.
2:26:00
So, like I see the data and I see how well positioned we are right now.
2:26:03
So, my last thought on this is this is now my my new favorite CEO tool. Back to your point. You know what these are?
2:26:09
>> Noise cancelling headphones.
2:26:09
[laughter] >> Yeah, noise cancelling. >> That's great.
2:26:13
Can you unpack a little bit more of the Whimo partnership?
2:26:15
Talk about how that's progressing um and just broad roll out your plans where you how how fast do you see that becoming meaningful and growing? >> For sure.
2:26:26
So, I'm going to zoom out one click and you just to sort of remind everyone.
2:26:30
So, I'm super bullish on >> uh AV's autonomous vehicles entering the ride share space.
2:26:36
And the reason I'm so bullish is because they're they're um market expanding.
2:26:41
You know, when when and we see this in market aftermarket when AVs come in, people take more rides overall because it's cool.
2:26:46
It's a it's a cool new product. It's kind of fun. It's sort of futuristic. It's reliable.
2:26:50
So, so it expands beyond the current, you know, population of people who take rides here today.
2:26:55
And then the cost over time will become lower.
2:26:57
you know, for example, insurance costs are lower.
2:27:00
That'll be pretty obvious.
2:27:00
So, it's nice when the market expands and when your cost of doing business goes lower. >> Okay.
2:27:06
So, then you ask about Whimo.
2:27:07
Whimo, of course, is the leader here in the United States. No question.
2:27:09
And we have a very um specific partnership with them announced in Nashville.
2:27:14
>> Um that was literally one of the meetings I was in yesterday where we're talking about all these different ways where we're integrating across the two platforms.
2:27:20
Everything from fleet management to demand generation.
2:27:22
And you know if we get a ride are we going to are we going to you know ask Whimo to take it or is a human driver going to give it like all these different you know sort of details that you have to work out when when you're doing a deep partnership which we expect to go live um later this year.
2:27:38
>> The expectation of that partnership certainly that we have and I believe that Whimo has as well is that this will scale you know beyond just Nashville but Nashville is very much our focus right now.
2:27:47
>> I'll tell you one last thing.
2:27:47
>> I'll tell you one last thing. um uh Teara who's the CEO of Whimo was just on a a show yesterday and she was asked about this partnership too and she was it was nice to see her also being very enthusiastic about it and saying you know the thing that I think should be obvious to many people which is even if Whimo wants to go it alone in some
2:28:04
markets like they're doing here in San Francisco it's really helpful for them to have a demandgen fleet management partner like us in a place like Nashville and hopefully elsewhere as well because it allows them to really focus on what they want to do well which is you know create the world's safest driver that's the they call it and not have to worry about all the other details of riding of a ride share network. So anyway, it's, you know, stay
2:28:22
So anyway, it's, you know, stay tuned.
2:28:24
We've actually, you know, broken ground on the big depot.
2:28:26
I mean, we're doing all this sort of work.
2:28:27
And in the next couple of months, we'll start to roll uh rides out in um in in Nashville.
2:28:32
>> What is uh talk about I'm I'm super curious to actually understand like what does the depot look like?
2:28:36
Like what is what is the process?
2:28:38
What are all the things that goes into it?
2:28:40
What are what are the jobs? That kind of thing. >> Yeah.
2:28:43
So, first thing is it's big.
2:28:45
[laughter] you know, it t it takes some space uh because you you have to think I mean you're talking about hundreds is something that has to be able to clean, maintain, charge, um uh you know, tune up all all these different things, repair in some cases, you know, couples, hundreds of cars, hundreds of hundreds of cars.
2:29:04
And you know, like at at 2 o'clock in the morning, you know, these cars are not necessarily going to be out on the road.
2:29:10
So you got to have a place for you know hundreds of cars to sort of go through uh this cycle when the when the demand is low.
2:29:15
Um there'll be real jobs there and actually we're hope and we have a we have something called flex drive there which is our own fleet management subsidiary and that kind of is the basis of this thing that we're kind of building it around let's say and so some of the folks working at flex drive you know we'll transition into doing this AB stuff.
2:29:29
We'll also frankly end up hiring some drivers in the community who kind of want extra money and know a lot about cars and so forth.
2:29:35
But yeah, I mean, think of it as sort of a a big very kind of modern uh you know, kind of um let's say char there are also some pit stops around the around the town that just do charging, but think of it as sort of charging and maintenance, but done for kind of you know, 2026 and beyond instead of >> Yeah.
2:29:50
So, so yeah, that was going to be my next question is is uh do you have uh for a city like Nashville, you have one major depot, you have some charging stations.
2:29:59
Would you ever break out the depot?
2:30:01
Basically having like multiple spaces for storage or is like one for a Nashvilleized city is like one facility going to be enough uh to to to kind of likely operate.
2:30:14
>> I think you know one big central and then a couple of satellites is probably right.
2:30:18
It's it's sort of tricky to site these things because if you put them too far outside of town the good news is it's cheap but the bad news is they've got what we call deadhead miles.
2:30:26
So just miles going back and forth with probably no rider in them.
2:30:29
So you know so you have to find a place that's kind of the right you know ours is kind of near the airport all this sort of stuff.
2:30:36
>> Uh and then and then you want a couple and it's got to have a lot of power too literally electricity like we work with a utility to figure out what's the place that can support a fairly big power load and then so anyway that becomes your central thing and then you have a couple of satellites around that.
2:30:48
How do you think about the the the scaling and roll out of self-driving vehicles just broadly in America?
2:30:55
I think a lot of people lean back on what the AI systems can do, incidents per mile, and there's a lot of smooth curves there, but we in a lot of AI world, we're running into the real world.
2:31:10
Is there enough energy for this data center?
2:31:12
Is there enough are there enough chips for this uh new model or this new new training run? at a certain point.
2:31:17
Um, how are you feeling about the ability of the OEM market, the car production, supply chain, uh, even things like regulation, like how how do you see all that playing together?
2:31:32
What's the most important thing to focus on?
2:31:34
>> I mean, you're you're you're literally laying it out.
2:31:37
Like, think of this whole thing.
2:31:38
You've got, first of all, you got to have great tech that works at scale >> without a driver in it that can go, you know, for a long time with with no instance.
2:31:45
So that's a very very important thing.
2:31:46
Then you have to have an OEM, right?
2:31:48
And that OEM has to be willing to make an investment, right?
2:31:52
Because it's not just like cool, let's just take the same lines and just, you know, throw some AV stuff on at the end.
2:31:57
It's like you want to design production lines that build that A and think about what you know, you can see this with Whimo, right?
2:32:02
Like Whimo literally is receiving and I'm talking about the Jaguar platform that they used mostly so far.
2:32:08
You know, they're literally like receiving these kind of half-built cars and then kind of disassembling them and reassembling them. It's very complex.
2:32:12
So you you want to be able to build that in a very costefficient way.
2:32:15
So OEMs have to have to be willing to make a big investment there.
2:32:18
Then you've got to have the regulations line up and that's a city by city, state-by-state thing, right?
2:32:23
So we just signed an agreement, this is outside the US, we just signed an agreement with Hamburg, Germany.
2:32:27
That'll be the first city in Germany and one of the first in Europe to say, "Yeah, we welcome ABS onto our um uh roads.
2:32:34
Here's the kind of framework we have to use."
2:32:36
Anyway, so that's got to line up.
2:32:38
Then someone's got to buy these things and finance these things.
2:32:39
So there's got to be financing.
2:32:40
By the way, these things might last 3 or 400,000 mi, but unlike a normal used car, they'll have zero value at the end, right? Zero value.
2:32:47
No one's going to buy a used AV that's been on the road for 400,000 mi and uses old tech.
2:32:50
So, okay, that's a whole financing model that people have to figure out.
2:32:54
Then there's got to be all the stuff that we were just talking about, all the fleet management stuff, and that's physical world stuff, right?
2:33:00
That's like who's going to charge these things, who's going to reboot them when they need rebooting, who's going to clean them when someone, you know, throws up in the car, all these things.
2:33:07
And then, uh, who's going to maintain them?
2:33:09
And then you have to have all the demand, you know, management and so forth that we've we're pretty good at.
2:33:13
So I think your analogy is a good one, which is that and it's why you see so many tiny pilots which are all kind of handbuilt and and artisally done.
2:33:19
But when you think about scale, it really is going to be years and years and years until all of these things line up in market after market after market.
2:33:28
Um yeah, I don't know if that exact track do you track the EV AV roll out in China like what what's the timeline for manufacturers to just be you know a BYD that can just plug into a a platform a Chinese equivalent platform like Lyft and just like start earning uh autonomously.
2:33:51
>> I mean I will say this is an area where it's just flat out China is ahead. Flat out China is ahead.
2:33:56
out China is ahead. I mean this maybe slightly different from the question you're asking but when I talk to AV so if you're if you're in the business of of producing AV technology uh you know it could be a mobile eye it could be a a wave it could be you know any number of people who are producing this technology independent of the OEM when you talk to them their biggest frustration I'm
2:34:18
actually going to quote one of them right now is saying we talked to quote western manufacturers and and it really doesn't matter whether you're talking about VW or Stalantis or board or whatever just western manufacturers and and then you talk to Chinese manufacturers and and in China they're talking 1 to two years and then that same conversation with a western manufacturer is a 5-year conversation. You know it's a it's a 2032 conversation
2:34:40
You know it's a it's a 2032 conversation instead of like a 2027 2028 conversation.
2:34:45
So the short answer is, you know, there's good AV tech from a lot of places, but frankly, the OEMs are in very very different positions in terms of their ability to to to spin up quickly and and um uh yeah, and get and get a platform out that actually works. >> Well, makes sense.
2:35:02
>> Congratulations on all the progress.
2:35:04
Thank you so much for taking the time to come chat with us.
2:35:06
>> And fantastic shirt, by the way. >> We love the shirt. >> Coming in strong.
2:35:09
>> Well, thank you, gentlemen.
2:35:09
It means a lot coming from two fashionable guys. >> Yeah.
2:35:13
In the plain white shirt.
2:35:13
I'm really taking a risk today.
2:35:15
[laughter] >> I mean, you're bold.
2:35:17
It's part It's It's on brand for you. You're You're bold. >> Yes.
2:35:21
In a world of uh you know, plain black t-shirts of tech, the white collar shirt white collar shirt does kind of stand out. So, thank you. >> I appreciate it.
2:35:28
Hey, congrats to you guys on all your progress and your Super Bowl ad. That was amazing. >> Thank you. That was a lot of fun. It was a good time.
2:35:35
>> Slightly smaller ad than yours, but we're working our way up. >> We're getting there. >> No, dude.
2:35:39
You got to start, you know, step by step by step by step.
2:35:40
I will tell you, I did zoom in on all the logos.
2:35:43
I was looking for the lift logo.
2:35:44
It must have been hidden. I don't think I saw it.
2:35:46
But any it was >> somebody's getting fired.
2:35:47
[laughter] >> Tyler is looking into it right now.
2:35:51
We're going to send you evidence if it's in there.
2:35:52
If we But we might have made a mistake. It's possible.
2:35:54
There were a few that were behind certain letters and so they got >> And that was it. I looked very closely.
2:36:00
Pretty sure it was hidden. It's no problem at all.
2:36:01
I'm not holding it against. >> We'll figure it out.
2:36:03
>> No, we take we're taking >> Good thing the Super Bowl happens every year. >> Yeah. >> There we go. 61.
2:36:06
It's just around the corner. >> There you go. Great to see you, David.
2:36:10
>> Have a great rest of your day. We'll talk to you soon. >> Cheers. Thank you.
2:36:12
Let me tell you about Cognition.
2:36:14
They're the makers of Devon, the AI software engineer.
2:36:16
Crush your backlog with your personal AI engineering team.
2:36:19
And without further ado, Todd McKinnon from Octa is in the reream waiting room. Welcome to the TV room. Todd, how are you doing? >> What's happening? >> I am excellent today.
2:36:28
Thank you so much for having me on.
2:36:31
>> Thanks for thanks for joining. First time on the show.
2:36:32
Please kick us off with an introduction. >> I'm Todd McKinnon.
2:36:36
I'm the founder and CEO of Octa. >> Yeah.
2:36:39
And uh yeah, I'm excited to be here coming from the office in San Francisco and I feel like yeah, >> my hair is not nearly good enough to be on the show. >> It looks fantastic.
2:36:47
What are you talking about?
2:36:48
Uh I mean, since it's the first time, would you mind going back and telling us the the a little bit of the history, the story, the I I love these like founder journeys.
2:36:57
So, you can give it you can take as long as you want or as short as you want, but whatever whatever you haven't gotten sick of talking to people about.
2:37:05
>> Yeah, it's interesting.
2:37:05
It's a lot there's a lot of echoes in what's going on today with AI, but 17 years ago, I started Octa the same year Steph >> the same year Steph Curry was drafted. So, >> no way.
2:37:17
>> I don't know if that makes it seem like a long time ago or not that much time ago, but 17 years ago.
2:37:21
>> Octa really is the Steph Curry of the Enterprise. >> It is. I've always said that.
2:37:24
Yeah, I've always said that. Lots of people. >> Yeah.
2:37:27
Needs better supporting cast around him maybe, but I'll take it.
2:37:31
>> But yeah, I mean, how did you even come up with the idea?
2:37:32
I think a lot of people uh start a company, they think like I'm going to build something for myself.
2:37:37
I'll go consumer because that's hot and trendy.
2:37:39
Like how what was the inciting element the like the problem that you identified?
2:37:45
>> Well, I was running engineering at Salesforce and I was working with all these customers that were adopting Salesforce and then I looked around.
2:37:50
I said, "Hey, AWS is launching, Google Apps for Domains, which is what it was called back at the time, is launching."
2:37:57
And I said, "Hey, you know, there's really going to be a cloud version of everything in it, and it's probably a great time to go out and start something new and build something that would be needed in that world."
2:38:04
And we settled on this platform for identity management, which seems maybe uh maybe pedestrian or incremental, but at the time, people were had these applications like Box and Salesforce and didn't really work well with their Windows login and Windows PC.
2:38:20
And that was our wedge to get started to what now is a $3 billion a year company, 6,000 people >> and a really important part of the world.
2:38:28
>> Was this in the sweet spot of venture capitalist during that time?
2:38:30
I mean Salesforce pedigree, it feels obvious in hindsight, but what was the pitching process like?
2:38:37
>> Well, it was very different world because we were in the, you know, aftermath of the financial crash. >> Yeah.
2:38:44
>> Yeah. and venture capitalists were were you know where's are our capital calls going to be returned are we gonna have any money like are we gonna be so people were very reticent we actually were the first company to raise money from Andre and Horwitz
2:38:56
>> no way you were the very first company >> very first yeah >> wow [laughter] and on the first investment it's octa that's insane I mean congrats to you but also a lot of lot of people are like yeah the first investment I made was not great public company I mean I'm person. The first fund has Slack and
2:39:14
The first fund has Slack and >> Yeah, it's a great fun.
2:39:16
Yeah, >> nice Sarah's in there and uh >> and then Skype, wasn't Skype in the first one?
2:39:22
>> Yeah, but Skype's a weird one cuz it was like a take private and then a IPO or something or sell.
2:39:27
>> Yeah, I remember like talking to my friends and they're like, "Oh yeah, you raised Andrea Horses. That sounds great."
2:39:31
They the company did that weird Skype thing where they bought it eBay and took it and worked out okay. >> Yeah. >> Yeah.
2:39:37
But at the time, you know, it was it was um people thought, oh, it's not maybe it's not big enough.
2:39:43
>> They said, oh, you're providing a single sign on for cloud apps to your, you know, Windows domain. Is it big enough?
2:39:50
And it's like for the last 17 years, it's like, you know, is it a big enough market?
2:39:54
market? is and every four or five years more cloud and then mobile and then what happened with the really focus on security and how important identity management and passwords are to security and now what's happening with AI co remote work identity just becomes more
2:40:08
and more and more important and now with AI we're on this precipice of the and a huge catalyst for identity that's why we're so pumped up and excited about what's going on >> yeah so let let's get right into it you octa powers identity for biggest companies in the world and uh right
2:40:24
We're at this inflection point right now where people are adding effectively adding uh thing adding adding uh agents to their workforce and oftentimes those agents are are uh playing the role of of humans in some way functioning with different applications working across the company like how are all those
2:40:43
conversations going uh how are how are organizations adapting and and how are you guys um fitting into it all >> well on a personal level I've never seen anything like it the level of interest the uniformity of the interest from customers in every vertical industry in every size company. They want to do
2:40:59
They want to do something with AI and they want to invest in agents and it's it's quite a tidal wave of interest.
2:41:04
I think now what really gets exciting is is how we turn that continue to turn that interest into actual revenue and actual usage of the new products.
2:41:13
But the precondition is there.
2:41:15
People are definitely interested.
2:41:16
I think what's happening is it's not a huge shock.
2:41:18
Every, you know, boards of directors and CEOs, they all read X. Yeah.
2:41:23
They know how excited everyone is about this and they're like, "Oh, give me some of that." Right. Come on. Come.
2:41:28
>> I'll take one million agents, please. >> Exactly. Yeah.
2:41:30
And then and then the poor the poor people put together these prototypes and these demos and and were kind of like hacked together and they had access to every piece of data in the whole company and they're like, "That's awesome. Put it in production."
2:41:41
And the developers are like, "Wait a minute. >> I don't care. Put it in production.
2:41:45
I told my board I was going to have an agent and this is an agent. Go, go, go."
2:41:50
And so what we're trying to do is trying to help them manage that and help them, hey, track these things, make sure whatever these agents are connecting to, you know, how it's connecting and and lock it down, make it secure so when they go into production, it'll actually work. >> Yeah.
2:42:02
>> What kind of general horror stories have you heard at at companies? Is it Yeah.
2:42:07
Is it an agent that can effectively go and pull data somewhere that they shouldn't be able to >> prompt injection like give me a refund?
2:42:15
There's so many there's so many different risk vectors that are that are you know most of the time when when this happens at a company hopefully it's just internal and like you can correct it quickly >> but it doesn't feel like we're that far
2:42:27
off from you know a company not being ahead on this kind of stuff and you know having an agent that can just query their own internal data and and having some type of bigger leak but obviously you're working to prevent that. >> It's very it's very simple. It's it's
2:42:40
>> It's very it's very simple.
2:42:40
It's it's the the agent was hooked up to multiple data sources and to get access to more data than you wanted, you just need to ask the agent.
2:42:50
I mean, pro prompt inj prompt injection sounds, you know, it's something cyber people say to convince software engineers it's a, you know, maybe a hard thing to prevent, but we're talking about simple cases where the agent has too much access and by a couple couple prompts you can get it to tell you the customer information of a different customer.
2:43:09
So it's we're and luckily I think if you zoom sometimes in Silicon Valley we're so obsessed with the latest and greatest we we think that everyone has fully agentic customer support and fully agentic everything.
2:43:22
But the reality is mo most companies are very early.
2:43:24
So it's a good time to layer in some maybe controls and some visibility into what these agents can connect to and the data they can access so we can avoid some of the really bad mess ups going forward.
2:43:34
Do you think the Claudebot now openclaw sort of Mac Mini agent personal open- source story is useful in concretizing what is going to happen in the enterprise over the next 12 months for you?
2:43:54
Has has it been a reference point in discussions with leaders that you work with?
2:43:59
work with? I think it's it is it's another catalyzing event because I think in some cases people are a little bit behind on the power and the business value that could be delivered from these agentic systems and I think when they can I was just at my kids soccer game the other day and they were talking
2:44:19
about clawbot they were talking about you know just these weren't tech people they're on the sideline and I think they were talking about how powerful it was and how they're going to get restaurant reservations and I think that catalyzes is the the side of the equation which is go go go. And I think unfortunately
2:44:31
And I think unfortunately you're also going to see uh more security issues and more connections that were exposed and Mac Mini with personal data.
2:44:39
It wasn't a new Mac Mini.
2:44:41
It was like your family Mac Mini and now your family Mac Mini is is you know the details are all over the web.
2:44:45
You're going to see some of that >> and uh so we have to get both right.
2:44:48
We have to get the business value and and get people headed in the right direction and then also put the right controls in place.
2:44:54
How do you think uh business models evolve in a world of you know potentially millions of agents?
2:44:58
Uh companies aren't really used to running payroll for millions of people, identity for millions of people.
2:45:06
Some people are, but how do you think all of that changes going forward?
2:45:10
>> Well, I I have a bunch of super strong thoughts on this.
2:45:13
Some some of them which are maybe counter uh intuitive.
2:45:15
I think first of all, I think in five years there's going to be way more software engineers than there are today. >> Wow.
2:45:23
And and it's a little bit of a trick answer because I I mainly because I think there's going to be way more software. >> Okay.
2:45:29
>> And you know they'll be more productive but you you'll have more software engineers.
2:45:32
I guarantee you all these companies >> Yeah.
2:45:34
>> Meta, Amazon, Google, uh Anthropic, they'll all have more software engineers in five years than they do.
2:45:39
And double clicking on that, do you think it's that we're going to see more computer science degrees uh become software engineers in the traditional sense or that designer that you have, that PM, that business leader, maybe even your CFO all of a sudden is a software engineer or is creating software? >> I think so.
2:45:59
I think what a software engineer does, but I'm not trying to cop out and say all all people are going to be vibe coding and that's my software engineer.
2:46:07
I mean people building and architecting software and knowing internals of how it works.
2:46:11
They're not going to be writing as much code clearly.
2:46:13
But I'll give you a secret.
2:46:15
>> The last 30 years we've had integrated development environments that are helping us write code more efficiently.
2:46:21
This is a whole another level.
2:46:21
But it's not a new thing to get better tools to help us make software.
2:46:24
So there's going to be more software engineers, but there's going to be even more than the increase in software engineers.
2:46:29
There's going to be way more software, which is super exciting and and super super daunting.
2:46:35
Now back now if you layer on uh you know are people going to build all their own software or buy software for v vendors I I think that is unequivocally they're going to buy more software from vendors for sure.
2:46:45
Now the trick >> yeah you kind of people have to people that are just bearish on enterprise software have to remember that there's been open-source equivalents of like almost every major [laughter] platform forever.
2:46:58
It's like you could just pick it off the shelf and run it and maintain it and >> uh >> it's easy, >> right? >> Yeah.
2:47:05
And I think kind of like there's going to be this kind of goes back with my first thing I said about more software engineers.
2:47:11
>> There's going to be more software.
2:47:11
So there's going to be more package software.
2:47:14
There's going to be more customuilt software.
2:47:15
Like for example, there'll be a lot of I think people use companies using these software development tools to build the applications that have to span the multiple silos and the multiple systems.
2:47:27
Right now you these companies have a very hard time of building something that goes from SAP to content management to Salesforce and an end toend process and maybe there'll be more some more customization and more vibe coding and
2:47:40
more you know people doing cloud coding to build that but the core vendors and all of the core data sources and the core business logic and processes that's going to get bigger and bigger and I'm 100% convinced of that which is dangerous. You should never be 100%
2:47:54
You should never be 100% convinced of anything, but [laughter] I'm trying to make a point here dramatically. >> I love it.
2:47:59
>> La last year, last year and continuing into this year was all about coding agents.
2:48:03
Do you have based on the conversations you have with with, you know, Fortune 500 companies and and and things like that that are maybe out slightly outside of of just like San Francisco tech?
2:48:14
Like what other forms of agents are you most kind of bullish on uh for this year?
2:48:21
>> Orchestrators or orchestration?
2:48:21
I I think I think it's so back to my what I was talking about before.
2:48:26
Now, by the way, I I don't I don't think that the vendor lineup of established software companies is not going to change.
2:48:32
Just because I think there's going to be more and more vended software sold, it doesn't mean that the people that are leading all the categories right now are going to be leading.
2:48:41
I think one of the most exciting things about what's I mean this is an amazing time in our industry and personally it's super motivating for me and for our team but the most amazing thing is every leadership category is kind of up for grabs.
2:48:54
>> You have I mean how crazy is this?
2:48:56
>> 5 10 years ago thinking oh AWS might not be the number one infrastructure provider. >> Yeah. >> I mean that's crazy.
2:49:01
You thought that was locked in but you see Google and Oracle and these companies that's up for grab.
2:49:06
You see and and I think the same thing is true for all of the categories.
2:49:10
So it's not like there's no potential for disruption.
2:49:12
I mean in our world it's very possible that if you look in cyber that in five or 10 years the biggest category in cyber is identity. >> Yeah.
2:49:21
>> I mean that's pretty exciting right now it's you know firewalls and sock and >> endpoint but it could be identity and that and everyone sees this.
2:49:27
You pile network sees this and service now sees this.
2:49:31
They're doing acquisitions and that's exciting.
2:49:32
But anyways, back to my point was um is that it's exciting things could change and the real exciting agentic or the new type of software I think is the type of software that crosses all these silos.
2:49:44
If you think about it, that's what people were uniquely good at doing.
2:49:48
They were uniquely good at doing.
2:49:50
Okay, I'm going to look in this system and I know about this little corner case and this other content management system and I can apply some intelligence on that and go back to update.
2:49:58
back to update. That was very hard to do for packaged application vendors because they're inherently in their own silo and they have to make things comply to their own business rules and database and it's very hard I mean I worked in enterprise software for the all the years before I
2:50:13
started octa it's very hard for an HR company to think about end toend or a Salesforce automation company to think about end to end so I think this agentic systems that can cross these silos is a very very exciting opportunity >> last question for me is 2026 the best year here to move to San Francisco in history. >> Oh, now you've got me you're getting me
2:50:32
>> Oh, now you've got me you're getting me really worked up now. My strong conviction. I I love San Francisco.
2:50:36
And I think all of these people that said San Francisco is dead and it's a mess was totally overblown.
2:50:43
San Francisco is an amazing, beautiful place.
2:50:46
And no matter, you know, is the government perfect? No.
2:50:49
Is any government perfect? No.
2:50:51
But it's an amazing place.
2:50:51
And all the people still want to come here. >> Yeah.
2:50:55
>> You look at I mean, how many people moved to Miami are now coming back?
2:50:56
I mean, where are the great new companies being started?
2:50:59
So, I love San Francisco.
2:51:02
I have a sign behind me that says San Francisco. Yeah. Yeah. You got a bridge. >> You got a bridge.
2:51:07
>> Hey, we think somebody we think somebody should we let you know that the Golden Gate is not really gold.
2:51:11
It's red, but >> but your gate is gold.
2:51:15
>> We think you should you should sponsor you should pitch the city to sponsor the Golden Gate Bridge. It's wide open. Let's put an octa.
2:51:22
>> It is it's a good metaphor >> hanging under it >> for octa connects >> your identity as you go into the city to connect Marin to to Fort Point. Yeah. >> Yes. This is integration. This is fantastic.
2:51:34
Uh well, thank you so much for taking the time to come hang out. We'll talk to you soon. Come back. >> Thanks. Love it.
2:51:41
>> Let me tell you about Cisco critical infrastructure for the AI era.
2:51:43
Unlock seamless realtime experiences and new value with Cisco.
2:51:47
And without further ado, we have Dr.
2:51:51
Alex Zendowski from the Biological Computing Company finishing out our Lambda Lightning round. >> What's going on?
2:52:01
>> How you doing, doctor? >> What's going on? >> It's good. Good to see you.
2:52:03
Uh, >> lot of whiteboards today.
2:52:06
>> Yes, I I I'll ask you to explain that shortly, but first give us an introduction on yourself since it's the first time on the show. >> Yeah, absolutely.
2:52:15
So, uh, my background, I'm a neurosurgeon, a neuroscientist.
2:52:19
>> I've been growing neurons on electrodes for about 20 years. >> Okay.
2:52:23
>> Uh, I spent the bulk of my career studying how neurons and brains process information. >> Okay.
2:52:28
>> Uh, first in humans by implanting electrodes in the laboratory by growing neurons, as I mentioned, on electrodes. >> Yeah. >> And Yeah.
2:52:35
Now, >> really quickly, when you say when you say growing neurons on electrodes, does that look like a petri dish? Is that organic?
2:52:42
Is that meat space or is this a digital representation of the neuron?
2:52:49
>> So, there's no neurons on this.
2:52:49
Uh, but this is the dish that we grow them on. >> Okay.
2:52:53
>> Um, you can see it here. >> Yeah.
2:52:55
>> Uh, that little the little box in the middle has about 5,000 electrodes. >> Wow.
2:53:00
>> Um, and yeah, and we grow them on there. They're alive.
2:53:01
And we use them to process information as computers.
2:53:04
>> They're they're biological.
2:53:04
These are cells that are all connected to create the neuron. Fantastic. What can you do with it?
2:53:10
I feel like a lot of people are pretty happy with their NVIDIA GPUs.
2:53:14
What do I need a biological computer for? >> Yeah, absolutely.
2:53:17
So, first uh first of all, um neurons and brains are extremely more energy efficient in silicon. It's it's well known.
2:53:27
>> Um and our first application is to help solve this energy this looming energy crisis that's happening.
2:53:32
>> Um you know, second of all, neurons can change the the way they're connected to each other.
2:53:36
So, as you all know, current hardware is rigid. It doesn't change. Yeah.
2:53:48
>> So, we're all for compute. >> Mhm.
2:53:52
How much of this is a science project?
2:53:55
How long will you be in R&D? What is your timeline?
2:53:58
Uh I mean, we're No, we're we're not afraid of science projects here.
2:54:02
We talk to folks all the time that are give give us timelines like we're going to be live in 2035 and uh we love that but it feels like you're making some pretty advanced progress.
2:54:10
So talk to us about how the business shapes up when you see this going into production what the milestones are. >> Yeah.
2:54:20
So uh I have like a big note here, sticky note on my monitor.
2:54:22
Uh this is not science fiction. This is not research.
2:54:27
You know if if I was going to get anything across that would be the thing, right? We're doing this now. We we're deploying it. >> Yeah.
2:54:34
>> Um >> so far our products are twofold.
2:54:35
One, we're using the biological network to create a software layer that plugs directly into ANN's artificial neural networks now to make them better better fast better faster cheaper. >> Okay.
2:54:48
>> Um so that's currently happening.
2:54:50
>> In addition, and you asked about what this thing is, this is what we call our algorithm discovery platform.
2:54:53
So we're using real brain cells to identify what's coming after the transformers. Yeah.
2:55:01
>> So the way we do that is we build state-of-the-art models in house.
2:55:05
>> Um we parameterize them.
2:55:05
We ultimately use the biological network to understand neuroscience principles and define the neuroscience principles that can plugged into these state-of-the-art models.
2:55:14
And we do this in a loop.
2:55:17
>> Um and in the end um these plugins improve them and make them better, faster, cheaper.
2:55:22
>> Do do you have any reflections on the shortcomings of current AI models? Why they differ?
2:55:29
they differ? There's that sort of famous quote of like you don't need like a like a teenager can just learn to drive a car in like a month of you know a couple hours in the seat and yes we're able to train Whimos but on an energy and time
2:55:47
perspective like on an apples to apples basis you're effectively you know doing like a million hours of training and you don't have to give a human a million hours of training to drive a car so what is different about the current structure of AI systems versus the way humans actually learn. >> Yeah, absolutely. So, that's a bit of a >> Yeah, absolutely.
2:56:04
So, that's a bit of a history lesson.
2:56:05
Um, so, you know, going back to the 1950s where we had the first perceptron makitz neuron, that was actually based on how we understood at least at the time how neurons function. >> Mhm.
2:56:18
>> Fast forward to the 1980s, uh, back propagation won the Nobel Prize a couple years ago.
2:56:23
>> Uh, that's when a divergence happened at that point as back propagation is not a real neuroscience principle.
2:56:26
Um and that's when this divergence happened and ultimately since then uh software and hardware became very unbiological in the sense that again it's rigid uh in the way it processes information.
2:56:40
>> At the same time though interestingly around the 1980s neuroscience flourished >> uh we started to be able to grow brain brain cells in a dish.
2:56:47
We started to understand what the neural signals coming from brain actually means.
2:56:52
This is the basis for a lot of the brain computer interface companies that are currently out there.
2:56:58
Um and so three years ago when we started the company we said okay well you know now we're going to take a 2026 approach and understanding of how uh neurons process information and we're going to apply it towards novel AI systems.
2:57:13
>> How do you uh keep the neurons alive? What do you feed them? >> Yeah.
2:57:19
So um they live in a media that's got proteins. Um it's got sugar in it.
2:57:23
It's it's actually relatively simple. >> It's Pete. It's Ray Pete. It's on the Pete diet. It's got sugar.
2:57:27
Jordy, you're you're going to prove this. This is great. >> Sugar superfood. [clears throat] >> Yeah. Yeah.
2:57:34
These these these techniques have been kind of perfected over the years.
2:57:38
Again, this is one of the things that allows us to do this.
2:57:38
We can bl for over a year. >> Yeah. >> Um Yeah.
2:57:42
>> Sugar, protein, any any olive oil, any fats.
2:57:45
We got We got some Brian Johnson snake oil right here.
2:57:48
You could put that in there.
2:57:50
>> Definitely some olive oil. We'll try it out.
2:57:52
[laughter] >> Uh what uh yeah, coming like >> Yeah.
2:57:55
Fast forward to the end of this year, >> what what are you what have you done that uh allows you to feel comfortable taking like you know Christmas off or or New Year's Eve, something like that?
2:58:11
>> Yeah, that's a great question.
2:58:11
Um so customers, collaborations, design partnerships, you know, right now um the adapter product that we have plugs into video generation models.
2:58:20
video generation models. Um and so you know we are um open to working with any foundational model lab that's compute constrained and so um we can make our models more efficient >> which is every lab [laughter] >> that's right I mean you know similarly
2:58:36
as I mentioned earlier with our algorithm discovery platform we're looking for partnerships with the hyperscalers I mean you know they have research arms that are interested in what's coming after transformers right and so >> um we'll take the holidays off uh if you know once we're working with them as well. >> Mhm. Well, uh you raised some money. How >> Mhm.
2:58:53
Well, uh you raised some money. How much did you raise?
2:58:59
>> I'm very happy to say.
2:58:59
So, we raised a $25 million seed. >> Seed. >> There we go. There we go.
2:59:06
That's But what's after a mango seed?
2:59:08
What's bigger than a mango seed?
2:59:09
Cuz I think of a mango seed as like eight.
2:59:11
This is like a watermelon. >> Yeah. >> Watermelon.
2:59:15
>> No, watermelon has small seeds.
2:59:16
Watermelon, big fruit, small seeds.
2:59:17
Doesn't >> mango is the biggest seed.
2:59:18
It's just it's the end of the road.
2:59:20
After that, you just go just go giant AI seed round.
2:59:25
Well, congratulations and thank you so much for coming on the show and breaking it down for us.
2:59:30
>> Yeah, come back uh like a as you guys have breakthroughs and things like that. Come back on.
2:59:34
You don't need to have fundraising news to come on and talk about biological computing.
2:59:39
>> You guys, you guys come to the lab.
2:59:40
>> Yeah, San Francisco and Mission Bay.
2:59:40
You guys are always welcome to come. >> Amazing. >> See some brains. >> Thank you so much. We'll see you soon.
2:59:47
Have a good rest of your day.
2:59:47
Let me tell you about Graphite.
2:59:49
It's code review for the age of AI.
2:59:51
Graphite helps teams on GitHub ship higher quality software faster.
2:59:55
And without further ado, Andrew Huberman is the founder of Huberman Lab.
2:59:59
He's here in the TVP and Ultradom. Welcome to the show. >> Here we are. >> Cheers, guys. >> Cast in a can. >> Can you hear me?
3:00:10
>> Yes, we can hear you loud and clear.
3:00:10
Uh g give us the latest on Matina Yerba Mate.
3:00:15
>> Yeah, zero sugar cold brew yerba mate.
3:00:17
I'm half Argentine, so I've been drinking mate since I was a kid. >> Okay. >> It's got caffeine.
3:00:21
Um, and it's a super smooth arc of caffeine. That's why I love it. Yeah.
3:00:26
>> But, uh, the zero sugar cold brew. Yeah.
3:00:28
We made five different flavors. It's organic.
3:00:30
BPA, BPS, be fast free cans because a lot of people worry about that stuff. >> Yeah.
3:00:36
>> Nobody wants the credit cards worth of microplastics in their brains, testicle, and ovaries.
3:00:40
And, um, it's free of all that. And it's awesome.
3:00:43
I'm I have very high caffeine tolerance.
3:00:45
So I drink, no joke, like four to six of them a day, but most people do just fine with one or two. >> I'm right with you. It's crazy.
3:00:53
So So we have like such a regimented schedule that my caffeine intake during the week is I wake up, >> I don't follow your protocol around waiting, I get right into the Yerba mate.
3:01:04
I actually have AB tested it.
3:01:04
I notice I personally notice no difference whether I get started.
3:01:09
>> So you have one in the car before the gym?
3:01:11
>> I have one in the car on the way to the gym.
3:01:12
Caffeine before the gym is such a hack and I I don't do it and I need to get back into it. I don't know. >> Sometimes it's tough.
3:01:18
>> How how underrated is caffeine as a performance-enhancing drug in the gym?
3:01:20
I feel like that's it's good.
3:01:24
>> Well, look, if you have anxiety issues, you should be careful with caffeine.
3:01:26
But if you don't, um caffeine's great.
3:01:27
It's definitely a performance-enhancer.
3:01:30
So much so that you know um the people that regulate this stuff in sport set an upper limit on the milligrams of caffeine that you can ingest.
3:01:36
But um what I cut off the caffeine intake at about 2 pm.
3:01:42
But I listen I also drink caffeine first thing in the morning if I'm going to work out in the first two hours of the day. >> Sure.
3:01:49
>> And that's about 3 4 days a week I do that.
3:01:51
Other times I >> you know I wait a little while.
3:01:53
Um so it's really an option. But I love yerba. I also drink coffee.
3:01:57
I'll have a couple double espresso and um I love it.
3:01:59
I mean, but like you guys, I'm get up, hydrate, electrolytes, caffeine, family, work, work, work, work out, work, work, work.
3:02:10
And then by 8:00 or 9:00, if you cut the caffeine out earlier, you're going to crash on a steep cliff.
3:02:15
And it feels great because you fall asleep easily. >> Yeah. >> Yeah.
3:02:19
>> I always I always stop at uh I always stop basically when the show ends.
3:02:23
That's when I'm having my last caffeine for the day.
3:02:25
I sleep amazing and then get it started again. Yeah.
3:02:27
the the challenge is on the weekend because in the week our our schedule is so regimented and then on the weekend it's a little more flexible and I'll start feeling if I haven't had my fourth yerba by >> my fridge at home is filled with ma because if I don't I'm like not a good dad like sleeping.
3:02:44
>> I mean in South America you'll see people drinking yerba after meals because it helps with digestion.
3:02:47
It's a mild appetite suppressant.
3:02:48
So if you like I like to work out fasted.
3:02:50
It's not going to kill your appetite which is good because you know we still need to eat.
3:02:55
Everyone's like trying to eat less these days it seems.
3:02:57
still need to get the the nutrients and then it's high in antioxidants.
3:03:01
It's got a lot of positive benefits and um it's not necessarily an alternative to c to coffee. You could do both.
3:03:08
Um but it's not an energy drink quote unquote in the sense it it doesn't have alpha GPC torine all those things that a lot of energy drinks have.
3:03:16
Not not to knock on those things but you know I think yerba coffee it's >> the challenge is is u making traditional energy drinks a habit.
3:03:26
You're like supplementing a bunch of things that you don't necessarily like you're you you have to understand you're signing yourself up to be like constantly dosing supplements that a lot of people just aren't fully aware of because they they just look at it.
3:03:40
They're looking how much caffeine does this have and not not really paying attention to the other stuff.
3:03:45
>> How is the business Oh, sorry. >> No, go ahead.
3:03:48
>> I was just curious about uh the the the state of the business distribution.
3:03:50
uh the the I mean you have a massive audience so the logical place to start is direct to consumer but the final boss of almost every consumer product is retail.
3:04:00
How are you thinking about the roll out of the product?
3:04:03
>> Yeah, it's going great.
3:04:03
I mean we partnered with Matina.
3:04:05
So I'm a partial owner with SICOM.
3:04:06
You guys know Rob and rest of guys at Sycom andrew Wilkinson from Tiny. Oh yeah.
3:04:10
And um >> we are available online um through Matina through Amazon.
3:04:15
We launched in Whole Foods, in Costco, Canada, and in Sprouts Markets just recently, and um pretty soon it will likely be everywhere you look.
3:04:26
So, um and people love it.
3:04:26
You know, I'm very happy when people say they love it. It's delicious.
3:04:30
You know, 90% of the world's adults drink caffeine every single day.
3:04:34
It's the most popular drug in the world, >> and most people need it to function pretty pretty well.
3:04:41
recent study showed they looked at coffee but a couple of cups of caffeinated coffee per day seems like it can offset some of the dementia risk.
3:04:49
risk. Now there I mean with with a study like that there there are a bunch of you know issues that's not necessarily causal right people drinking caffeine probably working more focusing more more cognitively engaged etc etc but um as long as it doesn't send you into a um a spiral of anxiety you know caffeine is um is quite good for you that's very clear >> uh sort of random but uh do you have any insight into what uh what uh different
3:05:16
groups of Olympic athletes are are doing Uh there's when when I I've seen some I've heard some stories like there there's such a range with these athletes where you might have like a a an Olympic snowboarder who's like you know hard
3:05:30
hardly sleeping and just like having some energy drinks and heading out there and then you have like the guy that's doing like curling is like the you know the most regimented or whatever. Obviously everybody's different but uh
3:05:40
Obviously everybody's different but uh what do you think the what do you think the best athletes in the world are are doing this Olympics?
3:05:46
like what's what's cutting edge that maybe normal people aren't thinking about. >> Yeah.
3:05:52
So, I mean this also showed up in the NFL a little while ago, but um people are taking vasoddilators um you know Viagra and Tadalapil commonly goes by Seialis. Yeah.
3:06:01
You know, Toddalapil um is a vasod diilator.
3:06:05
It lowers blood pressure and you know people know of it as seialis for erectile dysfunction right but it was originally developed as a drug to improve prostate health because um when you vasoddilate when you you know dilate the vasculature you get more profusion of of the prostate and you need profusion of the prostate avoid infections but other things as well.
3:06:23
So it it the basic takeaway is that most every male 40 and older should probably be taking somewhere between 2.
3:06:32
5 and 5 milligrams of tadalapil not necessarily for erectile function although it will augment that as well but to lower blood pressure and to improve vasoddilation for the brain for the prostate and I'm not saying this as a biohacker um or a podcaster we had our head of male sexual health from uh Stanford his name is Mike Eisenberg he's an MDP PhD.
3:06:53
He is um best-in-class in terms of male sexual health, endocrinology, etc.
3:06:58
And uh that's his recommendation and it's one that most every male maybe 35 but at least 40 40 and older should take just as a preventative um you know strokes.
3:07:10
>> So you don't think that'll people are taking it, athletes are taking it as effectively some type of performance >> to lower to lower anxiety pregame to um because it lowers your blood pressure a little bit.
3:07:22
um they think it might be a performance enhancer by delivering more blood to the muscles.
3:07:26
You know, you have to look at what's on the band list and there's obviously lots of things on the band list.
3:07:31
And then they're going to be the things that people are going to use because they're not on the band list and they'll be banned the next Olympics.
3:07:37
That's often how it goes.
3:07:37
So I pay a lot of attention to this over the years.
3:07:39
If you looked in the '9s, you saw things like bromeocryptine, apomorphine, things that um people who know those names will augment acetylcholine, dopamine [clears throat] in order to improve fast transmission of neurons and and alertness and focus for getting quickest out the blocks in sprinting for instance.
3:07:57
Those drugs are now banned. Okay?
3:07:59
You'll see in sports where staying calm is an advantage.
3:08:02
Um any shooting sport for instance, >> most of the banned drugs are things that lower heart rate.
3:08:09
Imagine skiing up to a target, picking up your gun and shooting.
3:08:12
The the the less you're shaking, the better.
3:08:14
So, there's a bunch of banned drugs and then there are all the things that people would take to lower their blood pressure.
3:08:18
And some of those are not banned.
3:08:20
And this is kind of how the cycle goes.
3:08:21
We heard about the skiers injecting their penises in order to um expand their um bulge size, I guess it, in order to get a little more air resistance, in order to perhaps capture a little more air time.
3:08:32
That was hit the press mostly because it's kind of it's kind of humorous when we think about it. Um, >> that was real.
3:08:39
If >> you look at any sport, you're going to see this.
3:08:42
You're going to see this.
3:08:43
You're going to see peptides like BPC57, I believe, is banned.
3:08:47
>> Um, but that's a naturally occurring gut peptide.
3:08:49
Very hard to test for because you have tons of it circulating.
3:08:52
And the synthetic versions may or may not differ from that version as much so much that you can detect it. >> Yeah. >> Interesting.
3:09:01
>> Do you have a question about peptides broadly?
3:09:03
Yeah, I would love I would love like an update.
3:09:05
Since we last talked, probably tens of millions of Americans have jumped into the fray or >> taken over San Francisco.
3:09:13
Chinese peptides is like a whole meme in tech.
3:09:17
And I think everyone's wondering Fouian bargain at what level should they be experimenting with these things?
3:09:22
What type of research they should be doing?
3:09:25
Um what is the what is the current thinking around peptides broadly?
3:09:28
Okay, I'll hit the the key bullet points here for the noviceses and for the afficionados.
3:09:34
So, a peptide is a short chain of amino acids.
3:09:36
Um, you know, so insulin is a peptide.
3:09:39
So, when people say peptides, they're generally referring to things that people take by injection or orally in order to augment some effect, healing, growth hormone release, uh, etc.
3:09:48
Um, just as when we hear the word steroids, I mean estrogen is a steroid hormone, but when we hear about steroids, we're typically thinking about androgenic steroids, right? testosterone etc.
3:09:59
Okay, so when it comes to peptides, some are FDA approved, right?
3:10:05
Some are made by drug companies have passed through their patent.
3:10:07
Um, many of the so-called growth hormone secrets, tessamellin, these are things that cause the pituitary to release more growth hormone.
3:10:16
These have been tested FDA approved.
3:10:19
Most people are not getting them from uh drug companies.
3:10:21
They're getting them from either compounding pharmacies which are our lowerc cost alternative that there isn't as much um stringency in terms of purity.
3:10:30
So that varies by compounding pharmacy or from what we call gray or black market sources.
3:10:36
So when you say Chinese peptides, what you're talking about is you can go online and buy peptides that are listed or labeled as for research purposes only.
3:10:43
And that's what a lot of people are injecting. And that is a concern.
3:10:46
And I, you know, I'm not saying this because of my relationship to Stanford or anything like that, but I'll just say I do not think anyone should inject peptides that are labeled as for research purposes only because even if it says 99% purity, the 1% is likely to be something called LPS, lip polysaccharide, which um can cause inflammation.
3:11:07
It's not a good thing to be injecting.
3:11:09
Maybe one injection doesn't do anything, but when you're injecting every day for five, you know, 5 days a week and for a month that you can start running into some issues, autoimmune effects and so and so on.
3:11:19
That said, let's just acknowledge what people are doing.
3:11:21
And there's some peptides that have very impressive results for which there's essentially no human good human data, lots of animal data, and lots of anecd which is a synthetic version of a gut peptide can accelerate their wound healing. What's the problem?
3:11:41
We don't have a control experiment. It acts systemically.
3:11:45
So I can't inject it in one elbow, not the other.
3:11:47
if I have two elbow injuries and say, "Okay, it worked better for this one, not the other."
3:11:51
People claim it could be placebo, but people claim that it helps them.
3:11:54
>> Study where they break both of your arms, >> but it but it but but the issue is going through your whole body. >> Yeah.
3:12:02
You know, and it increases vascular growth, it increases nerve growth, it increases cartilage regrowth.
3:12:07
From the animal studies, we know that.
3:12:08
So, the logical backbone is there.
3:12:10
People are using it like crazy.
3:12:10
The lethal dose is very, very high.
3:12:12
I don't even know if anyone's achieved a lethal dose.
3:12:15
The one study that was done in humans looked at, believe it or not, BPC enemas at very very high doses for a a gut like a colitis um issue. I believe it was.
3:12:25
Listen, I I think you have to just decide what your margin for risk is.
3:12:29
The other one that's very interesting that not a lot of people are talking about is pineal, which um I've tried it.
3:12:34
I don't take it any longer.
3:12:34
Um I tried it for a little while and it doubled my REM sleep.
3:12:38
I was getting close to 3 hours of REM sleep per night.
3:12:42
It's thought to increase regeneration or the health of pinealytes.
3:12:44
and the pineal which make melatonin and so on and so forth.
3:12:48
But the big one the the the peptides that's going to change everything.
3:12:52
>> So I I think I I tried that the the one you just mentioned last week.
3:12:55
I was hanging out with uh with uh David Senra and a friend of ours had it and uh it was in a I think it was in a capsule form.
3:13:05
I expected it not to work very well.
3:13:07
Friend said it works fine. I tried it. No effect on my data.
3:13:11
Is that is that uh how how do you think about uh different form factors?
3:13:19
>> Injectable only and on an empty stomach not having meaning not having eaten in the previous 2 hours 30 minutes before bed. >> That's an issue.
3:13:25
That's an issue because it's a gray market right now.
3:13:26
People just be like, "Yeah, I'll sell you a peptide in this capsule."
3:13:29
Technically, it's it's what I'm saying it is, but I'm giving it to a form factor that doesn't actually work.
3:13:35
And so, it's all placebo.
3:13:37
But I just looked at the data and I was like, I tried this two nights in a row. It had zero effect.
3:13:40
And I don't think a lot of people >> I doubt it was pineal and it might have and and sometimes they'll throw some melatonin in there so you feel quote unquote an effective for some people.
3:13:49
Um you got to get it from a compounding pharmacy or a reliable source.
3:13:53
I'm not suggesting people do it.
3:13:55
I actually stopped taking it because and this is not a promotional but I take AGZ which is their sleep formula. I help design it.
3:14:02
My sleep on AGZ is amazing. Amazing.
3:14:05
I'm getting two and a half hours of REM sleep per night.
3:14:07
I get tons of deep sleep. I feel great.
3:14:09
So, I don't need pinealin.
3:14:12
>> But the peptide that's going to change everything and um this is more up your guys' alley because this gets right into the business sphere >> is Lily has a patent.
3:14:18
Eli Lily has a patent on a peptide called retatride.
3:14:25
>> Retatrutide is sometimes referred to as GLP3.
3:14:27
It's a triple agonist of the GLP1, which we're all familiar with, glucagon, and something called GIP.
3:14:33
I believe if I'm wrong, someone will correct me, of course.
3:14:37
The the interesting thing here is that in a phase three CL clinical trial in humans, it caused up to um onethird loss of body weight.
3:14:44
So a loss of one-third of body weight in about 6 months time.
3:14:48
And it seems like there's some degree of muscle sparing.
3:14:50
It works phenomenally well for weight loss.
3:14:53
The bodybuilding community has been onto this for a long time, right?
3:14:59
Bodybuilders always get there first.
3:15:01
Then what happens is in Florida and the United States, doctors who know who work out in gyms with people who know how to gain muscle and lose fat quickly start experimenting.
3:15:12
Then it goes into their highle clients.
3:15:14
Then it shows up in Hollywood.
3:15:16
Everyone lies or avoids answering the question of how they got so jacked.
3:15:19
They talk about eating chicken breasts and they're actually taking growth hormone and cipionade and winstrol and ratitide reat, excuse me.
3:15:27
And then and then everyone hears about it and they start you know whatever looks maxing it and then it's a big giant mess out there. Here's the deal.
3:15:34
We are looking at a potential change in the laws around peptides such that buying peptides would become illegal.
3:15:40
I actually think this is a terrible idea.
3:15:45
But the the motivation behind this is largely because Lily owns the patent.
3:15:50
People are already injecting reati.
3:15:50
They can buy it out there.
3:15:53
They're taking it at maybe a third of the dose that's recommended in the trial and they are seeing phenomenal results at fat loss.
3:16:01
Not a week goes by where I don't get a hundred questions about reutide to my phone, my personal phone, let alone email and etc. Wow.
3:16:08
>> So the reason why peptides might soon become uh illegal to purchase uh through even compoundingies, let alone gray market is because Lily would like to protect the domain over that patent.
3:16:20
This is going to be a trillion dollar drug.
3:16:21
trillion dollar, but people are already taking it.
3:16:24
I've never tried it, but people I know who have taken it said it works phenomenally well at a fraction of the dose.
3:16:28
So, today what what are what are some benefits that outside of weight loss?
3:16:36
Are there secondary >> reduced alcohol appetite, appetite, reduce impulsivity perhaps?
3:16:39
Remember that the the receptors for these things are all over the brain and body and you know the other GLP agonists have been looked at for alcohol use disorder, for binge eating disorder and for a lot of you know behavioral and impulse control disorders because a lot of the receptors are rooted in the hypothalammus which controls a lot of um impulsivity type behaviors and anxiety related behaviors.
3:17:02
So you know my stance on things is you know um do as you wish but know what you're doing but with the important caveat that if you are an adult and your body is fully formed you are in a completely different landscape than if you're 14 18 20 you know I worry very much we're hearing more and more about
3:17:21
young people especially women in the past but now also young guys who are thinking oh you know they've got to take TRT that's crazy you know I've talked about the fact that at 45 I'm 50 now 45 years old I started taking a micro dose of cipionate micro dose but micro so I
3:17:37
maintain fertility I've checked that I take hCG etc never in a million years would I suggest anyone else do that in their 20s or 30s unless they have a serious issue with peptides there's the risk of where you're getting it from purity right but there's also the risk that if you're augmenting growth hormone
3:17:52
in your teens your 20s you can really mess up your hypothalamic pituitary body axis all the organs of your body in major ways you shut down fertility um you get if you come off these things you you may revert to a state you know a depression etc. I I just think the whole
3:18:07
I I just think the whole looks maxing phenomenon is really really dangerous and foolish especially in people younger than 40, >> right?
3:18:18
Um it's just >> well especially because there's a way to looks max without hitting your face with a hammer, >> sleep diet, exercise, you know, taking uh >> get you 99% of the way there.
3:18:30
>> Also see you need to evaluate kind of where you are at the very end of puberty.
3:18:34
Puberty is a very protracted thing.
3:18:36
You know, I hit puberty when I was 14.
3:18:37
I didn't shave till I was 20.
3:18:37
I swear my head and face change shape, you know, over time.
3:18:42
You got to let people develop over a long period of time.
3:18:44
If you introduce exogenous hormones or you start blunting estrogen or you start um you know, cuz guys will say, "Oh, they have too much estrogen or something or they they start taking peptides.
3:18:54
I mean, you can really throw off the trajectory of your physiology and then you're the person in your 40s and 50s often who looks much older."
3:19:01
This is the thing that people don't realize that there's always a dynamic tension between the things that give you vitality and the things that extend your life.
3:19:08
For instance, if you take growth hormone, you'll feel more vital.
3:19:13
You'll gain muscle more easily, recover from injury more easily.
3:19:18
You'll lose visceral and body fat. Guess what?
3:19:19
You'll also age more quickly.
3:19:22
The reason bigger dogs die earlier than smaller dogs is they have higher dosing of of growth hormone and IGF-1.
3:19:27
It's it's well established.
3:19:27
So you get vitality but you lose time.
3:19:31
With testosterone you have to ask yourself what are you gaining? What are you losing?
3:19:37
If you can go from, you know, having no energy whatsoever to the ability to work and exercise, increase libido, etc.
3:19:43
Well, then you're probably gaining years if you're doing those things.
3:19:47
But I'm very concerned about people just, you know, trying to get a whatever an angle jawline, um, you know, abs, etc.
3:19:53
I mean, it's it's because of phones, right?
3:19:56
People taking photos of themselves.
3:19:57
I mean, I'm old school, so I feel like if you're a guy taking pictures of your abs and posting it online, um, like like you need to invest some money in a in a in a psychologist, not in a in a peptide.
3:20:07
But that's just, you know, that's a that's a generational thing, right?
3:20:11
I think I'm more impressed by what people can do with their physicality trying to hit it some numbers, some lifts, some running, some hiking, some rocking and and then but this overemphasis on appearance on social media, I think is going to cause a lot of um destruction for people that try and chase that. Mhm.
3:20:30
>> Uh, do you get questions from parents who have teenagers that are and they're considering giving their their teenagers HGH >> all the time?
3:20:41
So, people who have kids that are much shorter than their peers um nowadays often, I'm not suggesting doing this, will delay puberty in their kids.
3:20:55
>> Um, they'll sometimes give them growth hormone.
3:20:57
Um, we know that some of the drugs, the prescription drugs that help with ADHD can blunt growth >> um at high dosages.
3:21:04
Um, there's actually a very interesting drug uh that's went through phase three or is at the end of phase three now, which was designed, excuse me, designed for excessive daytime sleepiness called Senosi, >> which um is kind of a dirtier hit of the dopamine and epinephrine and a little bit serotonin receptors that is promising for ADHD.
3:21:23
that's a non-infetamine ADHD um drug um that's very interesting um that might be a good alternative to some of the aderall vivants etc.
3:21:32
Of course you need a doctor's prescription for these things.
3:21:36
>> Um I'll just throw out a name.
3:21:36
I have no uh formal affiliation to him but there's a YouTube channel a guy we had him on the podcast Dr.
3:21:41
John Cruz, K R U S E, who has covered every aspect of adult and child ADHD from behavioral um tools to the different drugs, um the amphetamine based drugs, senosi and all of those modafanil very very uh high information content, short videos, 10 to 15 minutes, nothing like mine, you know, mine will cure insomnia, his [laughter] will help people.
3:22:04
So I I send parents there um and look, I can say, "What are you doing giving your kid growth hormone?"
3:22:11
But like I don't know, I was never the tallest in my class.
3:22:13
I was never the shortest.
3:22:14
I'm, you know, 6'1 if I stand up straight, you know, 5'11 if I if I slouch.
3:22:18
And so I I do understand why people feel like they need to do this for their kids based on what you hear out there.
3:22:25
But when you start playing um atome endocrinologist and now that it's very easy to find a physician that will work with you on these things, you are you're you know, you're you're playing a dice game and you're playing it.
3:22:38
So, you have to think real carefully about when when and if to do these things.
3:22:42
But ultimately, you know, parents decide and kids decide.
3:22:47
>> Switching gears, I want to talk about cannabis. >> I was about to ask. >> Yeah. Are you guys fans? Be honest. >> I'm not a fan. >> Yeah, me either. It used to be.
3:22:56
>> I mean, I grew up in I grew up in Northern California, so it was uh always It's funny.
3:23:01
Wine country is like cannabis country, you know?
3:23:03
Oh, it's less visible than, you know, the vineyards and things like that, but >> it used to be expensive, difficult to obtain, laced with oregano.
3:23:11
Now it's like industrialized, available at the push of a button.
3:23:17
And and I mean, the New York Times is writing, you see these charts like the number of hospitalizations that was never a thing in the 90s that I remember.
3:23:25
Um, and it feels like we're on the cusp of really dis redising this as a society on are we in an epidemic?
3:23:33
do we need to have a more national conversation?
3:23:35
Um, but what are you seeing um in the research and and in your world about how big the problem is? >> Yeah.
3:23:43
Well, as a personal stance, I'm I'm largely for um legal but regulated.
3:23:48
That's just as my personal stance.
3:23:48
I want to just throw that out there, but that's just me personally.
3:23:50
Um I've hosted two guests, one clinician, one scientist about this and I did a solo episode.
3:23:57
solo episode. I did the soul episode and in addition to making some small errors in there about receptor affinities and things like that, I made a um a statement which was based on the literature I was seeing which is that
3:24:08
there is a population of people especially young males who have a predisposition to schizophrenia or bipolar meaning they have a first relative um with those conditions that are at very high risk for uh cannabis induced psychosis some of which doesn't reverse. >> Yeah. >> Yeah.
3:24:23
>> Um which is really scary.
3:24:23
I got accused of misinformation >> by traditional media.
3:24:28
Not, you know, not like being a misinformation.
3:24:30
And I'm going to keep my hands under the table because my response to that I think they can understand. >> Yeah. >> So misinformation.
3:24:37
Those are the general American medical. >> Yeah.
3:24:41
>> I I I've witnessed >> and at this point everyone has the the anecdosis. Okay.
3:24:53
Look, so there's more data.
3:24:53
Now I get it right and the media game is a whole discussion into itself.
3:24:56
So um the g I had one guest come on very thoughtful guest researcher from Canada who explained that the psychosis is often caused by an overingestion of too much THC >> and in his words just to be fair to him because he's a serious researcher serious scientist said that when people smoke so if they inhale it by vape or smoking they tend to hit a plane of high that's not excessive and and the the risk for psychotic episode is is lowered. Mhm.
3:25:25
>> Then the clinician I brought on said that he's not observing that.
3:25:27
He's a clinician, Keith Humphre. He's from Stanford. He's not observing that.
3:25:29
He's an expert in addiction of all kinds.
3:25:31
And so what we do know for sure is that THC concentrations are very high.
3:25:35
It's very easy to overingest through an edible versus through an uh inhalation.
3:25:39
Look, I think that the issue is how often and how much and on what genetic background uh is somebody doing this.
3:25:48
Now the problem is you don't want to be part of the experiment.
3:25:53
You don't want to be the person that discovers that you have a predisposition to psychosis by smoking high THC weed.
3:25:58
We see the same thing withratom.
3:26:01
So withratom you've gotratom products that people will say oh you know helped them get off opioids and I don't deny that right you've got activist groups that are proratom.
3:26:09
You also have people who will say they took so-calrum products which are isolates of the molecule.
3:26:16
This is what we tend to do in this country.
3:26:17
We find an interesting plant that has some psychoactive properties and then I don't know maybe pull the nicotine out of tobacco, concentrate it.
3:26:26
Next thing you know they're making money and you're putting in, you know, five pouches a day.
3:26:32
>> They'll take the plant which is chewed. The leaves are chewed.
3:26:34
People say, "Oh, it's kind of a balance between alertness and calm." Great. They'll isolate them.
3:26:39
Then they'll doratum isolates.
3:26:39
And now you're getting people who are having opioid like addiction toratum isolates to synthetics.
3:26:45
Likewise with THC, you had a balance between THC and CBD in the plant, different strains, etc. , etc.
3:26:50
It was a kind of a a community for a long time.
3:26:54
This has, you know, uh it even has religious uses, etc.
3:26:57
, and and cultural, you know, uh relationships.
3:27:00
And then all of a sudden, it's like, no, we've got really high THC concentrates that you can buy at the convenience store, and then people with a predisposition are having psychotic episodes.
3:27:09
Here are a couple things.
3:27:11
I'll say one thing facicious, and then a couple things serious.
3:27:13
One, I love to compete with people who like cannabis because unless they're a martial artist or a creative like a musician, >> you're going to blast past them.
3:27:23
So, you know, it's kind of like recommending um great ne Netflix series to your competitors, right?
3:27:31
>> Remote or remote work.
3:27:31
I recommend >> you're going to win. Okay.
3:27:33
Now, that's the facious side.
3:27:35
The serious side is, you know, couple guys or friends, gals, two, getting together and and smoking some weed and one person is like, "Oh, this really calms my anxiety, makes makes them feel as if they can engage socially."
3:27:48
And the next thing you know, they're losing motivation for a bunch of other things.
3:27:51
Layered on top of >> social media, the challenges of life as they always existed, but also nowadays.
3:27:58
And the next thing you know, you've got somebody with some serious mental health issues and a full-blown addiction.
3:28:01
I do know that my colleagues that work on addiction, Dr. Dr.
3:28:04
Anna LMK, who's an MD, who wrote Dopamine Nation, Keith Humphre at Stanford, they'll tell you that many, many more people are coming to them of all ages saying they struggle with cannabis.
3:28:14
I will also say that you're not going to get REM sleep if you are ingesting cannabis, certainly within a few hours of sleep.
3:28:20
Anyone that quits cannabis will tell you that their REM sleep, their dreams are just comes back like crazy for many months, if not years.
3:28:28
And you need REM sleep to offload the emotional load of prior day experiences. So, I get it.
3:28:32
And then the last thing is that there's always the comparison to alcohol.
3:28:36
I actually did a a post about this on X.
3:28:38
It's but it's crazy, right?
3:28:40
I mean, I've talked about alcohol before. Zero is better than any.
3:28:43
Two per week are probably fine.
3:28:43
If you're going to drink more, do more things for it to, you know, support your lifestyle. I'm not a tea totler.
3:28:48
Do as you want, but know what you're doing.
3:28:52
But people always say it's better than alcohol.
3:28:53
As if you had to pick one. And maybe you do.
3:28:56
Maybe in order to socialize, people feel like they have to pick one. I don't know.
3:28:59
I've always been able to say whatever the heck is on my mind without any alcohol or drugs.
3:29:02
So, you don't want to be around me if I were drunk.
3:29:07
Um cuz I'll tell everyone everything, you know, but now I do it anyway.
3:29:11
And I think it's just an issue of >> I get it.
3:29:14
In your teens and 20s, late teens and 20s, college, 30s, when you go to a bar, you don't want to sit around and and um necessarily sober. I get it.
3:29:23
So, what are you going to do?
3:29:23
You going to have a couple drinks or weed? Which one is better?
3:29:26
It really depends on who you are.
3:29:27
Um, and the comparison is just a kind of a fool's errand.
3:29:31
You're you're never going to solve what's better.
3:29:32
But it there are many many more people smoking or ingesting high THC cannabis.
3:29:39
And you have examples of people who are extremely successful, extremely ambitious who use cannabis, but those people probably need it to not go over the the the peak of anxiety from their from their ambition.
3:29:50
They're using it to stay in a healthy zone.
3:29:52
Whereas people that are, you know, have more apathy, they're less motivated, um, they should stay away from cannabis in my opinion if they want to make something of their life. >> I completely agree. That's great. >> So, yeah. >> Yeah.
3:30:06
>> Hard to fit into a post on >> So, uh, there's been a lot of studies on the impact of social media on people's mental health, >> right? Young people, teenagers.
3:30:16
What about uh and then right now one of the most popular words in the world is BR or phrase is brain rot, right?
3:30:23
Uh >> are there any largecale studies being done on the impact of consuming an hour or two hours a day of short form video?
3:30:35
Like what what are you tracking?
3:30:35
Uh because if you just look at the data, YouTube has I forget the exact number, but the average person in the world is watching like a meaningful amount of shorts every day.
3:30:45
So some people are watching >> hundreds and hundreds, others are watching zero or or maybe a few. >> Yeah. >> But there's a range. >> Yeah.
3:30:52
Well, I'll just offer a tool that I think is very useful.
3:30:54
I mean, I don't have problems with behavioral restraint, but I but I do have a lock box for my social media phone.
3:31:01
So, I took >> my an old phone.
3:31:03
I put social media on that phone so that if people text me social media links, I can't open them.
3:31:08
>> It [clears throat] also means that when I take out that phone, I'm posting or commenting or or doing that.
3:31:12
And I keep it in this lock box which is like a supermax prison.
3:31:15
You can't even code out of it.
3:31:17
Has a no code out setting.
3:31:17
And I'll leave it in there about 20 plus hours a day. And it's just great. >> It's just great.
3:31:22
I mean, your productivity will go through the roof.
3:31:24
I mean, it's getting very easy to blow past your peers now just by avoiding being on social media a little bit more.
3:31:31
I mean, David Gogan said this.
3:31:31
It's like it's easier than ever now to to be successful if you just don't do certain things.
3:31:37
Whereas in the past, it was like, what do I have to do to be successful?
3:31:40
like the shortest productivity self-help book will be, you know, lock your social media phone in a in a box for 23 hours a day.
3:31:46
And then I do think that so I'm not aware of any studies looking at 1 to two hours per day.
3:31:52
I do think that when you approach social media or you're getting online, I think it's important to realize that the brain has an absolutely insatiable appetite for short form video, but it's appetite for long form audio is also very high.
3:32:10
This is why podcasts continue to grow in their reach.
3:32:12
People don't want to hear audio turning over quickly unless there's video associated with it.
3:32:18
So, I try to get on social media at least once a day, maybe half hour, 45 minutes at the most.
3:32:24
Um, but I'm posting and commenting.
3:32:27
I think at the moment where you forget time >> and you're just like and it's sucking you in is the time where it needs to go back in the lock box.
3:32:34
But listen, Joe, I listen to you guys.
3:32:37
I I listen to news on social media.
3:32:39
And I think that's the world we're in.
3:32:41
I think that some people also can be on social media and experience some distance from it.
3:32:44
other people are more kind of pulled in emotionally to what's going on.
3:32:48
It depends if you're a creator and people are, you know, commenting about you or you're actively engaged in the conversations or if you're simply an observer.
3:32:56
But I think an hour and a half a day should be the upper limit for anyone that actually wants to accomplish something in life and probably more like 45 minutes to an hour unless you're watching news.
3:33:07
If you're learning, how do you know if you're learning?
3:33:09
What we know from neuroplasticity studies is that if you reflect on something the next day, you're likely to not forget it.
3:33:18
I It sounds sort of trivial, but most of learning is anti-forgetting.
3:33:19
In fact, there was a beautiful study done where they have people read some a passage once, twice, three times, or four times versus reading it just once and then self- testing on it in their mind at some point later.
3:33:30
And the self testing in your mind, what do I remember? What don't I remember? I forget that part.
3:33:35
extended the memory of what was in that passage out 6 months to a year in some cases.
3:33:40
So the reflection on what you saw is the key.
3:33:43
So if you think, okay, what did you see yesterday that was of interest where you actually learned something of value, then you're actually learning from social media or a podcast, but if you're just getting inundated with sensory information, forget it.
3:33:57
That's called entertainment.
3:33:59
You're just you're just numbing out or they're rage baiting you.
3:34:00
And you know, I mean, between what happened between like the Minnesota stuff, right, >> then the Epstein thing, >> and then you know, the next political whatever, >> you know, you the amygdala is just getting hammered.
3:34:16
And we know when the amydala is activated, you're getting many more sort of bits of memory stored.
3:34:23
So, but you're just getting the peak of that information.
3:34:25
You're kind of getting a summary.
3:34:26
So, >> look, your your brain real estate is your most valuable real estate.
3:34:29
What are you going to put in there?
3:34:30
what what are you learning?
3:34:31
And so I would say take a walk and ask yourself what did I learn on social media today or yesterday?
3:34:36
And there's actually some learning to be had.
3:34:39
Like I've reflected on what I've seen in this Epstein thing.
3:34:41
There were a lot of scientists there.
3:34:42
There were a lot of people that were um that buried his offenses so that that he could come out as this kind of like science finance liaison buries.
3:34:50
I mean it was wild how this could happen right after a conviction.
3:34:56
And so there's some some consideration to be had, but you need to get away from the content in order to think about that >> and then return to it in a way that's that's more I think nuanced and that can serve you as opposed to just getting you know rage baited into it all. >> Yeah.
3:35:10
>> Never get rage baited.
3:35:13
What excites you about AI? >> Anything? >> Yeah.
3:35:17
I'll tell you where I'm using AI. I've wrapped my book.
3:35:19
I'm doing edits on it right now.
3:35:21
And I didn't use AI to write my book.
3:35:23
Um, but what I think AI is great for is for designing little exams for yourself for exactly what I just described.
3:35:30
So, I will have I like Claude, so I'll just ask Claude to say um to test me on um give me a a 20 question multiplechoice test on um the microbiome as it relates to the solo episode that I did plus, you know, some recent work from my colleague Justin Sonnenberg, for instance, at Stanford, perhaps the best researcher on the human microbiome in in the world.
3:35:53
And and Claude will give me questions and I'll go, "Oh, that's cool. I remember that.
3:35:58
Oh, wait, that's where my knowledge falls off.
3:35:59
What percentage of the vagus nerve is is uh motor descending?
3:36:01
Okay, those things like are important to me, maybe not you.
3:36:05
And so I'm trying to cons I'm using it to consolidate information that I've learned prior that I might otherwise forget and update my knowledge and then I'll go to the papers, make sure the papers are actually what's linked there because, you know, as we know AI can hallucinate.
3:36:18
I think that's fantastic. Um so it's a teacher.
3:36:20
I'm using it to self- test which I know based on a lot of literature can help stamp down memories.
3:36:26
So, that's one way that I'm using it.
3:36:28
Um, I think I love the AI that's woven into eight sleep that is now I'm no longer adjusting the temperatures. It's doing it for me. >> Love that. Awesome.
3:36:37
Um, I'm giving it my blood tests.
3:36:41
>> Um, and it's helping me navigate that.
3:36:41
I mean, I had a blood lipid issue that was very slight, but I wanted to keep it in range.
3:36:47
And so, I started taking nattokinise.
3:36:49
I have no relationship to any company that sells it.
3:36:51
Boom, right back into range.
3:36:52
and AI helped me understand what nattokynise what sort of shifts can occur.
3:36:57
I mean I I love Claude and I don't you know I mean I'd love to work with them but I don't I don't have any relationship to them now.
3:37:03
So love love love what they're doing and and um I just think it's super exciting.
3:37:08
I'm not afraid of AI but I'm listen I'm born and raised in Palo Alto so >> like uh heretical to say you don't you know aren't excited about the most recent tech.
3:37:18
>> Last question for me and we'll let you get out of here.
3:37:19
Uh tell us about the fish behind you. Oh yeah.
3:37:21
So those are my discus fish.
3:37:24
They're freshwater discus.
3:37:26
Um and next to me on either side it was octopus on this side and there'll be another octopus on that side. Very cool.
3:37:32
>> Octopus died after 2 years. They lay eggs. They die.
3:37:34
That's just what they do. It was very dramatic.
3:37:35
Really cannibalized her tentacles. Everything. They're big drama. >> Yeah. Yeah.
3:37:42
They remind me of some online influencers who are always [laughter] trying to call attention to themselves.
3:37:45
But in the end >> it self-destruct. >> It self-destructs. Is it relaxing to you?
3:37:51
you? is there about why fish and yeah >> so I've long been a fan of a aquaria since I was a kid um and it's just the goal of setting up an ecosystem that's really in balance and it's it's made me very tranquil I've always had them in my office in my lab um and then some years ago I discovered a guy named Tekashi
3:38:08
Amano who unfortunately passed away some years ago but he developed this thing of aquacaping where it's really emphasizing the plants and the fish and the lighting so we built these blackbox rooms I'm living so I moved into an art gallery I converted a three-story art gallery into a living space. So, I have a gym
3:38:22
So, I have a gym [laughter] with my fish tanks, octopuses, >> a loft upstairs, and then in my basement is a no phone, no electronics zone where I draw, do illustrations, and I prepare for podcasts.
3:38:33
>> And I have it set up with these warm incandescent red lights.
3:38:35
My girlfriend and I like to go down there and listen to music and just hang like get totally away from the world.
3:38:40
Um, and I'm I just love that feeling of kind of being underwater a little bit, and they it relaxes me.
3:38:47
And I've got a bulldog puppy that was born a week ago and he's going to be here soon.
3:38:51
So >> more flora, more fauna is always good.
3:38:55
>> Let's bring the gong for the bulldog. >> Gong for the bulldog.
3:38:59
>> Bulldogs are the best >> name. Got a name yet? >> Yeah.
3:39:03
Strummer after the great Joe Strummer from the Clash.
3:39:05
I went to the the uh to New York last weekend.
3:39:06
My girlfriend and I went to go see Meet the Breeds. 140 purebred dogs.
3:39:10
everything from the little teacup ones to the bull mast tips that look like Bert Chryser and [laughter] everything in between.
3:39:18
And after we left, we just looked at each other and we're like >> mudded bulldog, which is what I had before.
3:39:24
Um, I love all the different breeds.
3:39:26
Uh, they teach you a lot about nervous systems and temperaments.
3:39:28
And, uh, Bullmastiff is the perfect dog for bird.
3:39:32
I think he's got a couple of them, but for me, it's always going to be the mudded bulldogs.
3:39:35
U, I don't think I'll ever deviate. >> Very cool. >> Amazing.
3:39:40
>> Thank you so much for taking time. Great to see you. Great to catch up.
3:39:43
>> Great to see you guys. Thanks.
3:39:43
And congrats on the um on the Super Bowl.
3:39:47
>> I hope to see you in the Olympics. >> We will be. We are. >> We are.
3:39:51
>> We're going to be in the Olympics.
3:39:53
>> And so Matahena's logo will be in there, >> right? Yeah. >> Awesome.
3:39:57
>> Maybe we got to get Mate Matina into the Olympics. I got to talk to Rob.
3:40:01
>> I I have an idea a minute, but Rob, >> the whole team The whole team has Look, the whole team has There we go. >> Love it. >> Yeah. Yeah. We open the show.
3:40:08
You're watching TVPN with these every day.
3:40:12
Thank you so much yesterday. We'll talk soon. Take care. >> Goodbye.
3:40:18
>> And with that, we do have to jump on with Tokyo.
3:40:21
So, is there are there any post >> any other news, Tyler? Did we miss anything?
3:40:26
>> Andrew Reed said, "Literally, everyone's capitulated on the AI trade.
3:40:27
There's definitely no bubble.
3:40:30
Surely now is the perfect time to declare victory."
3:40:35
>> Cloud code is growing 100% month over month. >> Wow. Wow. >> 2. 5 billion run rate.
3:40:40
>> 100 100 100% month over. Is that good? >> Yeah.
3:40:44
>> There's also the uh the codec spark came out today.
3:40:46
That's like uh the first >> Oh, yeah.
3:40:48
>> Open AI model running on Cerebras. I tried it out. It's extremely fast. >> You got it working.
3:40:52
So, you're on the pro account. Is that the $200? >> You need pro? Yeah. $200 a month. 5. 3 Spark. >> Okay.
3:41:00
>> Um I was on X high, which is like the highest, you know, they listening tiers. >> Okay.
3:41:04
I think I need to hit the update button. And >> it's super fast.
3:41:06
It's like so much better than um really like I I tried >> Codex Spark. Here it is. Extra high. >> I think I tried 5.
3:41:12
3 in in Codex like maybe a week ago.
3:41:14
That was the last time I used it.
3:41:15
It's like unbelievably it's so much faster. >> That's amazing.
3:41:18
Okay, >> so this is apparently what they had internally.
3:41:21
I think Run posted about this a couple weeks ago. >> Yeah. Yeah. Yeah.
3:41:24
No, he he was really baiting it because he was like, "It's slow. It's slow."
3:41:27
Like, you know, this could take us years to figure out.
3:41:29
Then it ships a week later. Classic. I love it.
3:41:33
Uh well, Tyler, build me an app. An amazing app. The best app ever.
3:41:39
>> One that will win awards and make me a legend in Silicon Valley. That's my prompt.
3:41:42
And I'm waiting for Codeex 5.
3:41:42
3 Spark Extra High to oneshot it for me.
3:41:50
We'll we'll check in with that tomorrow.
3:41:52
Anyway, thanks for watching.
3:41:52
Leave us five stars on Apple Podcast and Spotify.
3:41:57
Sign up for the TVPN newsletter at tvpn. com.
3:42:01
And we'll be back tomorrow for a Friday show.
3:42:05
>> I got to press it twice. Bomb's still going.
3:42:08
>> Uh we love [music] you guys.
3:42:08
Thanks for hanging out with us.
3:42:12
>> Can't wait for tomorrow. >> Great show. >> It's been an honor. >> It has.
3:42:15
>> Have the best evening of your entire life. >> Yes.
3:42:17
And we'll see you tomorrow, Friday. >> Goodbye. >> Nice work, brothers.
3:42:25
I'll see you on the next one.