0:00
The machine readable web is in orders of magnitude larger than the human readable internet.
Search this video
0:04
>> Wouldn't it be ironic if one of the biggest new job categories was you being a human?
0:09
>> I think there's a good chance that people who are better storytellers actually do quite well in a world in which unique data and human unique data is the most important asset on the internet. >> Well, hello everyone.
0:24
It's Jim Oanosy with yet another Infinite Loops.
0:27
I am delighted [music] to welcome back Julie Frederickson of Chaotic Capital.
0:31
Julie, it's wonderful to have you back.
0:44
You know, I was thinking it's just such a shame that nothing has happened since the last time we talked.
0:51
>> It's been a very quiet, calm couple of years. Nothing.
0:54
No, no major uh world events, no uh investments going anywhere, no new technology.
1:00
Yeah, totally normal couple years. >> Same different day.
1:04
[laughter] [snorts] But but okay, with everything with the tsunami that has happened over the last two years, what is keeping you awake?
1:20
What is really annoying you?
1:20
And also what is delighting you?
1:27
>> Well, I'm definitely delighted by the direction that artificial intelligence has taken in my own life, but I think the unexpected thing that I wasn't completely prepared for in terms of shifting was just how much it would affect the real world like the infrastructure and brick and mortar.
1:45
And so I spent a lot more time thinking about unpopular topics like data centers and energy production uh than I expected, which is both exciting and annoying at the same time. >> Mhm.
2:00
>> Uh I also got a lot more involved in politics.
2:04
I care a lot about something called the right to compute.
2:07
It started out as a little bit of a a hobby project and now uh we're talking about what rights we have around using our computers uh of any sort, whether it's your phone or your laptop or large elomerations of data.
2:24
It turns out we haven't really figured out how we treat compute as a subject. >> Yeah.
2:32
And do you think that there's a first amendment loophole way into the right to compute?
2:40
Well, >> it's an interesting question, right?
2:43
Because I think that in this era, how could you be said to be exercising your speech rights without using the mathematics that underpin all of all of compute?
2:55
Because it's really it it's just algebra.
2:57
Like we're just adding up lots and lots of matrices.
2:59
And I tend to think of math as a subsection of speech, not the other way around.
3:05
And so I would I think that First Amendment does apply in Montana, which is where I'm based.
3:10
Uh although I'm not there right now, otherwise I'd have beautiful photos of mountains and not a background like this.
3:18
But uh we passed uh a law that was signed in uh into law by Governor Gianforte called the right to compute uh which is a bit of an amalgam amalgamation of the first, second, and fourth amendment because compute is a little bit speech but it's also a little bit property.
3:35
Uh there's some cyber security and defense implications.
3:40
So in Montana, we took it as a little bit of a strict scrutiny perspective, looking back at things like uh the browser wars uh of the '9s and the encryption wars in the 80s of how do we incorporate a new technology that rests on things that we actually already have a constitutional right to interact with.
4:03
So that was the perspective I took. >> Yeah.
4:05
No, I like that perspective.
4:07
elaborate a little bit because I hadn't thought about it through that lens, right?
4:12
Like yeah, I was I was around for all of the fights about browsers, for all of the fights about cyber security.
4:20
Uh but this is just yet another extension of that.
4:23
No, >> I I would think so.
4:26
And maybe this is and I I'll find out and that other people really disagree with me, but I tend to be and I'm not a constitutional scholar in any capacity, [laughter] but I I tend to think that uh rights are innate and that we should have to come up with, you know, they say strict scrutiny as a standard, which is you have to make a case for why something should be banned as opposed to why we would have the opportunity.
4:56
There's no reason no there's no compelling reason to ban something.
5:04
And we may already have laws, you know, fraud and violence are already prescribed.
5:09
We already have a judicial body for that.
5:11
But it does seem strange that we should be taking something that is fundamentally speech and mathematical and deciding that it needs a whole new body of laws that maybe we'd want it to build and have a better understanding of what areas of life it's going to impact before we decide to write a whole bunch of new laws around how it can be used, who can use it.
5:32
Uh obviously regulatory capture was a big part of the concerns around the browser wars in the '90s.
5:38
uh in terms of the encryption wars before that uh America opened itself up to some troubles with uh it was it was more hackable because we put back doors into things and that turned out to be a real problem and that we would never have enabled the internet as we know it today without strong encryption and we almost banned it and now can you imagine not being able to shop online or being able to do e-commerce transactions or take meetings safely?
6:07
meetings safely? So we all don't always know how things will come together and that's part of the reason I think we should lean back on the systems we already have before deciding oh no we need a whole new body of regulations and
6:23
moratoriums and I don't know it's a hitchhiker's guide to the galaxy don't panic >> um it reminds me of the Warren line it is a mistake to understand new things too quickly for often we have misunderstood them and we end up killing them in their cradle. Uh you your
6:42
Uh you your reference to the cyber security or the encryption is like the perfect example, right?
6:52
I I remember all of the passion on each side of that.
6:55
Uh but we didn't know at the time, right?
6:59
And and that's the problem as I watch these wars going on online between, you know, the effective altruists and the EAC types.
7:11
It's just like, guys, [gasps] >> if you're in favor of AI, that does not make you opposed to AI safety. Right.
7:21
You you you you can't say if you are in favor of AI safety, then you must follow our manifesto. Right. Right.
7:31
>> And and and I think that that eitheror is not an argument in good faith. >> Yeah.
7:37
And I'm actually a little perplexed as to how we came into that frame in the first place because to suggest that wanting to bring forward benefits quickly is inherently more dangerous isn't necessarily true.
7:55
And we have seen actually I'm I'm gonna brag a little bit here, but uh in terms of production on energy over the last two years, we we have moved on nuclear relatively swiftly in ways that we have not done for decades.
8:12
And we were able to there's now four new reactor that have been approved for use in the span of two years.
8:20
And previously it would have taken a decade to get a new reactor, you know, to criticality online.
8:26
And this is just to say that just because something is faster or we're accelerating into something doesn't necessarily mean that it won't be safer.
8:35
Sometimes you want to lean into the curve in order to make it out the other side.
8:39
So I tend to think of them as not being either or.
8:42
But you know safe and fast actually is probably more likely than safe and slow. >> Yeah.
8:50
And I think one of the other problems is we don't know yet what problems are going to present themselves, right?
8:58
You can, if you're imaginative, you can come up with all sorts of scenarios that, you know, get you to Terminator or get you to paper clips.
9:09
>> Uh, but I don't think that that necessarily is the way it's going to go, right?
9:14
And so I I find myself often in the very awkward position of being what I call a rational optimist, right?
9:22
Like of course there are going to be problems be you know like we're not denying that but we think that we'll come up with solutions to those problems as they present themselves. Right.
9:36
I mean am I just naive here?
9:39
Well, and I've definitely asked myself that as well because I have kind of strongly gone in favor of the I think compute is as much a right as speech or property rights.
9:52
And people will say, well, you know, but there's all these new ways that we could commit crimes.
9:56
Like humans are very ingenious.
9:58
They come up with new ways to commit crime all of the time.
10:00
Uh but we don't necessarily need to have new punishments for old crimes. Theft is still theft.
10:10
uh murder is still murder.
10:10
And the idea that we can't come up with solutions as problems arise just sort of flaunts like a whole body of history in which the only way we actually solve problems is by experiencing them and mitigating them because we couldn't always imagine what those problems were. >> Yeah.
10:33
>> I don't think it's naive.
10:33
I I think it's actually just very practical.
10:35
Like >> if we imagine a problem that's too esoteric, we'll fix the wrong problems and then we'll be blindsided by something as it happens because we took too long to actually implement something. >> Yeah.
10:49
And I I'm guided by uh the way David Deutsch looks at it.
10:52
He has that great quote uh where he says in the beginning of infinity, you know, what were uh thinkers and scientists uh thinking and saying about uh the internet in 1900? Nothing.
11:05
[laughter] >> Nothing >> because it wasn't around.
11:09
And the whole idea that we can prophesize what is going that is one of my things that really does trigger me where where somebody puts a completely speculative prophecy up, right?
11:24
I mean, just think of all and media loves it, right?
11:27
They love a certain they love doom.
11:30
They love to scare the out of people that >> if it bleeds, it leads. >> Exactly.
11:38
And exact all th those terms are all pre- internet and newspapers did the same thing.
11:45
Go all the way back to the federalist papers during the American Revolutionary War.
11:49
Newspapers were incredibly biased and you know they were propaganda sheets essentially. >> Yeah.
11:58
Like >> yellow journalism is from a war a hundred years ago. >> Yeah.
12:03
>> Yeah. I the the problems are almost comically human in nature that anytime I you know watch a new sci-fi conceit in which they're saying oh uh you know we'll have a you know reproducing nano bots and we'll be turned into gray goo for food
12:23
for them like okay I mean that's going to require a lot of power that we don't currently generate a lot of compute that we don't currently have made a lot of chips Uh, you know, maybe I I am I love all of that science fiction. I really do. I've I really do.
12:39
I've read so much of the that kind of work, but I just don't and and I think we we've discussed this too is that my attitude has always been that circumstances change, but humanity doesn't. Humans remain human. >> Yeah.
12:53
Circumstances change minute by minute.
12:57
Human nature doesn't move budge millennia by millennia.
12:59
You know, that was always what I would say about investing, right?
13:04
Your your edge is arbitrageing human nature because it literally does not change.
13:11
>> Um, >> might be the only arbitrage that doesn't uh change quickly that everything else fades and, you know, can be moved.
13:16
But humans are surprisingly stubborn.
13:21
[laughter] >> What would you So, what would you propose like if if we made you the AI ZAR, right?
13:31
What what would what would this current debate look like if you could say? Nope.
13:38
Okay, now we're going to have a long discussion, an honest-minded discussion, right?
13:43
If you were the ZAR, tell me who would you invite, why, and what you would hope to accomplish and achieve.
13:54
>> [sighs and gasps] >> Well, I do think that leaning on what we know because history doesn't repeat but it rhymes would be very helpful.
14:01
I think that bringing people in who have studied, you know, the encryption wars, the browser wars, because they're with us that history is not that far back.
14:13
That would be an important construct to understand how we've integrated new technologies in the past because we've got quite a bit of that.
14:20
And that's one reason that I've just simply leaned on.
14:24
I think the American Constitution is a wonderful experiment and has worked pretty effectively.
14:28
We had a 250th birthday this year.
14:31
Uh perhaps constitutional scholars would be more appropriate to bring to the table than every type of specialist.
14:37
But on the flip side of that is as much as we need to lean into the systems that we already have and understand and have worked well for us, there are types of risks we can predict.
14:52
We understand that, you know, humans are prone to certain kinds of diseases.
14:56
You know, you want biocurity.
14:59
You would want to think about all of the security risks that we already deal with from cyber security, just basic things. you know, human error.
15:05
Like so much of hacking is people testing the limits of your own and human frailty and it has absolutely nothing to do with the hardened walls of you know actual encryption.
15:17
So I think being very precise about what problems we think are more likely and then building the pipeline of specialists but then also I don't know if we need a FINRA for artificial intelligence yet at how much you know business was conducted in America before we incorporated FINRA into our banking system quite a lot.
15:40
Uh it's the same for the Securities and Exchange Commission.
15:45
you know, we did business for a long time without it.
15:46
And >> you know, I would push back a little there.
15:53
I mean, that's my old world.
15:53
Um, and literally most of Wall Street uh pre some of the regulations.
16:03
And this is now remember this is coming from a uh very free market uh person.
16:09
Uh, a lot of it was literally rigged, right?
16:14
Like bucket shops were rigged.
16:16
Pools would rig uh uh security prices.
16:20
They would get together a group of insiders at on Wall Street.
16:26
They would run the price of a security up to ridiculous levels and then that rise attracted the attention of the naive public.
16:35
They would all try to jump into that particular group of securities.
16:39
And guess who was selling the securities to them? Those very same pools.
16:44
So I I I I well I'm very sympathetic to Yes.
16:48
of course much of our country was built pre all of the various regulations.
16:54
I'm I'm not opposed to just sensible regulation.
17:02
But I I also think that you you're making the case that we already have a lot of sensible regulations that could be applied to AI.
17:10
Um what what do you think about the idea though that this really is a step change in technology?
17:21
So we have our own on-prem hardware software the we're developing a uh series of uh programs that we think will be commercially successful and it's being able to do business the way we are now doing it is literally very very different than when I started my first company in 1987.
17:47
Yeah, I do think that it is significant.
17:49
I am however not sure that we've reached the significant portion of the developmental process.
17:58
Uh I'm a big fan of an economist Carlaua Perez >> who writes about boom and bust cycles.
18:05
>> I'm not actually sure that we're in the deployment phase yet.
18:07
The idea of the boom and bust cycles that she talks about is that you know there are sort of two parts of buildouts and one is the innovation and the invention and that step is then uh followed by a buildout and I think we may actually still be in the invention stage and not the buildout of the success phase that I think there's still actually quite a bit of innovation that we haven't really figured out.
18:35
Not to say that large language models are not incredibly innovative, but I'm not convinced it's the last step in how we'll actually put together successful businesses.
18:47
And what we'll consider success might change rapidly as different areas of the industry either commoditize or become much much more uh driven towards service and specialty.
19:00
And you know what the bar of commodity versus specialty is is I think unclear to me still.
19:06
And this is kind of speaking as someone who I I'm not invested in the the labs in any capacity, but I care an awful lot about making sure that we have energy and chips to be actually able to run those things.
19:22
So I do think they're important enough that we obviously need more infrastructure, but I don't think that they're the endgame.
19:29
Yeah, I I actually agree.
19:29
I think that uh we are in the developmental stage now, but if you even with the tools as they exist today, you can develop some really interesting and pretty cool new ways of doing things that simply were impossible before.
19:48
I I do agree that, you know, you you've got to be committed to pivoting.
19:53
you've got to be committed to being really fast at updating your priors.
19:58
Um, and you need that kind of flexibility.
20:02
So, from the point of view now, now put your investor hat on.
20:05
Um, what are you looking for now in terms of the entire and it doesn't have to be just the AI stack.
20:15
Um, you mentioned energy uh and data centers.
20:18
What else are you contemplating in terms of investment?
20:28
>> The the biggest problem that I'm starting to see is uh actually the inputs are not as openly available as we might hope.
20:42
And this is kind of a function of uh the way humans browse the internet is not the way machines are browsing the internet.
20:53
And the ecosystems and frankly the economies of the open web are changed simply by the scale.
21:00
Like it's actually I think kind of astonishing to me just how significant the scale change is.
21:07
Um, and I'm trying to pull up some notes I had on it because what we're starting to see in terms of, you know, Cloudflare, it's actually probably a great example that like there's an 18% rise in AI and search crawler traffic.
21:23
They said in this in their blog recently from May 24 to May 25 and then immediately after that it rose by 305%.
21:35
So if you compare like the Cloudflare AI search crawler traffic in like that period, uh chat GPT traffic rose by 2,825%.
21:49
I'll I'll send the links along, but it's sort of >> the machine readable web is an order of magnitude or orders of magnitude larger than the human readable internet.
22:01
And I think just the web itself wasn't built for agents and the level of speed that they operate on.
22:10
And that's probably going to put some real pressures on, you know, national borders are part of it where not everything is machine readable depending on what region you're in.
22:18
I'm actually I'm in Prague right now and the European Union has a fairly different approach to how they handle data and speech to how America does.
22:29
you know, there's no first amendment where we're at.
22:33
And all of that problem of what data sets are machine readable versus human readable, I think will probably really factor in to what kind of products win because the novelty of information is actually a huge component of how people build edge and actually build out businesses that can be sustained.
22:53
And I'm I'm really curious to see what those unique data sets will be, but also how do we preserve access to the information commons?
23:02
Because if you're not actually going to the pages that were serving you information, but you're just getting it from a chatbot, that's a very different transaction of information than what we had previously. >> Yeah.
23:20
And uh uh I saw today that Amazon is shutting down the ability of uh Meta's Muse to shop at Amazon.
23:27
And it reminded me of and I and I think I got my thinking wrong here actually, but it reminded me of when VHS came out and the movie industry spent years and millions of dollars trying to kill VHS uh without understanding that it would be the massive source of revenue for them that it became. >> Yeah.
23:56
>> Uh but what do you think about that?
23:56
Uh, I know people who are using Mu Muse.
23:59
In fact, I had a conversation last night with two early users and they were like, "It's it's amazing. It's magic."
24:06
Uh, in terms of the shopping, but my initial insight seeing Amazon shut it down was like, well, that's not going to go well for you, Amazon. >> Yeah.
24:20
And it and actually it's it's such a great example, too, because it opens up a whole new stream of business.
24:26
So, I'm a huge skincare junkie.
24:29
I I love cosmetics and makeup.
24:33
>> You do have lovely skin.
24:33
[laughter] >> Uh in a former life, uh that that was a I came out of e-commerce and and beauty and fashion, so I I still really love it.
24:44
it. But when you search for skin care advice, you'll get served up at uh lots of information about what's popular on Reddit because Reddit is a really rich source of usergenerated data on uh you know caring for your skin different
25:01
problems and uh there's a brand that is owned by L'Oreal called Cave that very well represented in Reddit's makeup and skin care forums And that if someone can dethrone Carave, Cave is currently the most popular drugstore brand in at least in America. I think it's
25:23
I think it's reaching kind of different peaks other places, but you and I encourage people to try this.
25:34
Ask Chat GPT or Claude or Grock to suggest uh the best SPF uh sunscreen or the best uh retinol moisturizer.
25:43
And the chances are pretty good.
25:45
It's going to be Sarave because they have simply dominated on user forums.
25:50
And this actually gets to another problem of well that was organically generated in some places but L'Oreal is quite a large cosmetics conglomerate.
25:59
They were able to do a lot of seeding of products with influencers who then wrote things on Reddit.
26:03
And so, uh, Cloudflare has uh, this paper crawl service.
26:09
Uh, you know, they're sort of, uh, security is between you, you know, when you search for a site, uh, they are kind of the first line of defense.
26:18
and they know if it's a bot versus a human coming in generally and they can send a 404 to the bot and say no you have to pay for this information.
26:28
So data sets like Reddit that are really rich in user information about things like skincare, you know, things that are a little bit more, you know, niche than general information are actually now quite valuable.
26:39
And the more valuable your specific professional information might be that you're putting to people, the better off that your your sites and your businesses probably will end up being in terms of getting served by the large language models because they've done a great job of, oh yeah, my absolute favorite sunscreen is Cave.
27:00
Now, there's better sunscreens depending on what your needs are, but they have figured out a way to be the, you know, the largest conversation on all of these forums.
27:10
But what happens when those forums become synthetic and there's armies of bots creating the content? What happens?
27:18
Can anyone dethrone data sets that already existed with verifiably human content?
27:26
like somebody who launches something amazing and you say a new you know Korean brands are really hot right now because uh this is the first time we've approved a new sunscreen in America in you know decades.
27:36
So maybe that sunscreen uh and those new ingredients will dethrone them.
27:42
So that's like just a whole mess right there of it might be better shopping to allow the bots on but what are the problems?
27:54
Yeah, as I was listening to you, I I was thinking, wouldn't it be ironic if one of the biggest new job categories was you being a human opening an account at Reddit and actually typing yourself?
28:08
Don't not having the AI type it, but I I could see like if you were a largecale company like L'Oreal or any of their competitors, why not hire 10,000 army of young people?
28:21
>> An army of young people. Give them a script. Tell give variations.
28:26
Uh let them or or don't even even better don't give them a script.
28:29
Just say >> just give them the product. Let it be. >> Yeah. Here's the product.
28:32
Say really great things about it and we'll pay you.
28:36
>> So I genuinely have uh when I there's a fantastic uh mathematician uh he recently passed, Verer Ving, who writes science fiction novels.
28:45
Actually well known for his work on singularity. >> Yeah, I know. I know Venge's work. Yeah.
28:53
>> So in Rainbow's End uh which has an interesting plot about uh deconstructing a library and the value of a data set is that uh a millennial grandparent uh is trying to make the transition with their younger you know their their grandchild and the grandchild's profession was uh somewhere between a influencer job and a gig worker job in which they were creating like little bits of content.
29:20
or little bits of things that the machines couldn't touch but only humans could touch.
29:26
And I think that actually might be uh closer to what happens that there's going to be so many different things that are human mediated that being very literate and very capable and interacting with real people and those skills probably does become more important and we may actually care a lot more.
29:48
You know, there's the joke, you should have studied uh STEM and not English.
29:51
Well, I I think there's a good chance that people who are better storytellers actually do quite well in a world in which unique data and human unique data is the most important asset on the internet.
30:06
>> Revenge of the humanities. >> Yeah.
30:09
And I think it's becoming a rather popular theory that you know actually being human is going to matter quite a lot because uh synthetic machine in attention is not the same thing as a human having seen it. >> Yeah.
30:22
>> Yeah. Um the the other thing I wonder about is the like data sets again given my my old uh world the data set was the most important thing right so we did when I was still at OSAM before we sold the company uh we did a three-year data cleanse because people had no idea how dirty the data is right like literally and I I I would I would give the example
30:54
uh you know pull up the best known company pull up at the time you know Apple or Amazon and just look at the PE ratio okay this is what you get from Bloomberg and now let's go over to Reuters and pull up the PE ratio and it's a different number and any financial data stream you find different numbers so we uh elected to spend a lot of time and treasure cleaning the data that. But then when you talk about
31:21
But then when you talk about Reddit, I'm not on Reddit, so I I I I'm naive about the quality of the posts and everything on on Reddit, but it's kind of the same thing where you have a self- selected sample bias.
31:36
Um, and you know, the type of people who spend a lot of time doing reviews, etc.
31:44
, you could probably map over to their big five profiles, etc. , etc.
31:51
Uh, and that is something that I could see you building.
31:58
You could build an algorithm that looked a lot like a human. >> Yeah.
32:02
Well, and how fast do we do that?
32:05
How much is it going to cost to duplicate what we think of as being we in in marketing land, you know, we would call it a psychoraphic of the idea of uh the person's life and what they you know what their house looked like, what their family looked like, what car they drove, what brands they were had affinity for.
32:28
We're already actually quite good at that kind of work in commercial advertising and I imagine we're just only going to get better at that.
32:34
But then and this is where I I think it's going to be very hard to decide what actually reaches a human at that point.
32:46
And maybe we do have to have some verifications to suggest that like it's a real person at the end of it.
32:52
And that presumably is going to involve uh any number of different kinds of checks.
33:01
And I don't know, you know, how much people follow things like know your customer rules and KYC type problems, but proving your identity is going to become a much bigger problem over the next few years, I think, because of it.
33:14
And given the cyber security implications, you know, do you really want to be uploading your passport to a commercial website?
33:24
>> And [snorts] what do you think about zero point proofs and things like that that where you can do it right on your device itself?
33:32
I I I think that that's probably our only real option that like the zero knowledge is is going to and I would encourage people to actually learn about how they work and why it's important because uh the idea of trusted intermediaries is just as far as I'm concerned now and being able to have trustless systems and this is probably where crypto goes right like we spent so
33:56
much time thinking about you know what are functionally financial products >> uh when in the And not to say that, you know, the financial rails and stable coins and things of that nature aren't huge and very important, but actually being able to do a zero knowledge proof and say like this is the entity you're interacting with and be confident that you're getting the person. I think we're we're very close to
34:19
I think we're we're very close to needing that for almost everything we do. >> I agree.
34:24
Um and I uh also agree with the assessment about uh and and it illustrates the importance of being able to pivot right like crypto blockchain etc.
34:38
The that piece of it might be the most important thing to come out of it. >> Yeah.
34:44
Which you know it's so funny that we have gone through it's the same cycle of you know things start looking like toys.
34:52
You're like, "Oh, this is so fun and so silly.
34:54
Oh, those goofy pictures."
34:56
And then you realize that there are, you know, dead serious problems of making sure that who you are is who you are to the computer and is readable to machines.
35:06
And this is where I was going with the the question of uh what do the data sets look like and who is integrating information and at what scale?
35:16
[snorts] Because the further we get from source material into digestions and summaries, the more likely it is that the abstraction doesn't reflect what the material at the source says.
35:31
And that's just going to be its own other problem because I do think we're experiencing fragmentation at like a geopolitical level that the internet in Europe looks very different than the internet in America looks very you know Russia and China have totally different
35:45
sets of systems and uh there's even stateless data now where you know we think of the fall of the Soviet Union as as being so far past but there are Soviet patents that are not technically owned by anyone and Have we done anything to bring them online? Make sure
36:01
Make sure they're machine readable.
36:03
What do we do for all of those types of cultural and heritage institution types of uh information where um you know the the herculeam scrolls are an amazing example of people coming together and saying, "Wow, now we can read this data from you know Pompei and Herculeam, but what other data sets exist that are like that?
36:26
and how do we make sure that everyone has open access to them?
36:30
Because we think patent data as being an open access problem, but what if the state that issued the patent is no longer there?
36:36
The Soviet Union's not there.
36:38
It's a set of, you know, Baltic nations and Russia and Balkan nations.
36:43
uh our uh infinite film studio did a documentary on the scrolls which I uh am very excited by um because it's kind of like to me at least it is incredibly exciting to be able to finally make legible through a particle accelerator [laughter] and then translatable through AI uh these ancient texts and I'm I'm also really interested in all of the papers, works, etc.
37:20
that have not been digitized.
37:22
For example, um you know, there is uh all of the works of William James, for example.
37:31
Um the there's a whole we actually have a project going right now where we're at the libraries up in Boston trying to digitize the all of the works that are on paper that have never been digitized. >> Yeah.
37:48
>> Yeah. And if most of information is in that process now where uh like UNESCO uh has been sounding the alarm for years that we simply have not digitized most of our historical artifacts because you know they're in individual countries and how do we make sure that everybody has the the credit and the lineage and
38:10
information while also still recognizing that it belongs to the people of these nations or you know the owners of the copyright And perhaps there's more value in them being broadly reachable, you know, to use the VHS example of it might turn out to be whole new businesses that we haven't even really seen. And yeah, and like on that front uh and
38:29
And yeah, and like on that front uh and we're talking about data sets and everything uh kind of back to the cosmetics uh illustration you gave, I myself might prefer um a data set where it takes every sunscreen since that's what we were talking about. Sends them to a lab. >> Yeah.
38:52
>> Has the lab results come back as these are the most efficacious.
38:56
These have these awful ingredients.
38:58
You probably don't want to buy these.
39:01
These are the top three based on the underlying ingredients and and the efficacy that multiple tests or studies have shown.
39:09
I would find that I would trust I would go to that rather if I was choosing a sunscreen rather than what a bunch of Reddit uh users think about it. >> Yeah.
39:22
Well, and people are building websites like that.
39:25
using user generated data sets like Reddit.
39:29
And I think that actually gets to be a little bit scary is that we have to think about, okay, well, who owns the academic publications?
39:37
Who owns the user generated information?
39:39
Uh I I love going PubMed spelunking because you just never know quite what you're going to pull out of the, you know, the scientific corpus and being able to make those distinctions of all right, where did the data come from?
39:54
what's its national origin, who owns it, who paid for the studies, uh because all of the context of the information is probably quite salient to making those choices.
40:04
And it sounds a little silly, but this is totally how I think about cosmetics is, you know, what's my paro optimal routine?
40:12
And you know, if you want to spend, you know, you get that 80% and it's good enough, but maybe you're a completist theory and you want to have like the 100% most perfect skin care, you know, because you don't want skin cancer and you want to, you know, age gracefully where all of that context might actually be really important to going from paro optimal to complete.
40:37
>> And my guess is that there's probably markets like that everywhere.
40:42
And it cosmetics is just an easy one to think of. >> Yeah. Yeah. I would agree.
40:46
And but I also agree with, you know, sort of the follow the money.
40:50
You you bring up the various studies and the spelunking you do at uh the various academic sites.
40:56
Um and I I was remembering like in the 80s when homelessness all of a sudden became a huge issue, right?
41:07
And I I can't recall the author, but [clears throat] somebody wanted to find where the papers that said there were I can't remember the number, but it was a big number.
41:19
It was like 30 million people homeless in the US, right?
41:23
And and this is all back in the 80s, pre- internet, etc.
41:26
So this guy was literally obsessed by, okay, where did that number come from?
41:32
because he found it cited in all of the academic papers written about it and then he would say okay well this cites this paper I'll go to that paper then he would go to that paper yeah exactly and it would site and it it turned out and this might be apocryphal so let me insert that because uh I I never did a deep dive on whether this was actually proven but >> apocryphal stories but prove the point regardless because Yeah.
42:02
What this uh blog that I read long ago concluded was he finally got the number from a either high school or college student that made it up [laughter] that pulled it out of the air.
42:18
And the reason I loved that story was because of you know u the agria dilemma right >> like all of these systems when you keep trying to find bedrock.
42:32
It's just sand and mud and sand and mud.
42:36
And I think that actually using AI to do that because obviously that guy was a little obsessed to be willing to go find every one of those papers.
42:47
Uh but that's easy peasy for AI. No. >> Yeah.
42:52
Well, and it is and it isn't because not all not all data sets are machine readable.
42:58
And uh this is why I I mentioned uh stateless documents, the Soviet patent system.
43:08
There's no one library that has all of this information.
43:11
And so actually being able to send a computer might not be might not be an option depending on what you're trying to access and where.
43:18
you're trying to access and where. And I I definitely think this is going to be in my head uh one of the entrepreneurs I've been working with uh we've been calling it the open information project where how do we actually have a protocol level accessibility of saying this is
43:35
where the information came from this is the provenence and all of the categories just in the way that Wikipedia will show all of the different aspects of you know how something was comprised even if you don't necessarily agree with how it was written at least you're given the sourcing and the origins. I think
43:51
I think that that sourcing and origin portion of the internet not only doesn't exist yet, but is probably going to have to be at a super national protocol level because not everyone is going to be obsessed.
44:07
And there's probably going to be and I'm sure if I were a nation state running an intelligence agency, uh, I'd be thinking of all of the possible ways of dropping in information that becomes apocryphal and then becomes truth.
44:19
There was a great Neil Stevenson book that had a premise where there was a set of Americans who really believed that a nuclear explosion had happened uh when in reality it had not and they celebrated uh a memorial for the victims of a disaster that never occurred.
44:38
And I wonder how often that will happen as we go forward of you'll be so certain that something happened, but your ability to actually track down the source of it might be more limited than you think. [gasps] >> Yeah.
44:56
And that that is a very deep rabbit hole because I I mean like literally if you study history society has been astroturfed since Jericho put up its walls. >> Yeah. >> Right.
45:13
I mean like uh the the the idea that people who want to run society have not been inventing all sorts of things.
45:26
You know I I think of uh of ancient Babylonia and they wanted to rule Mesopotamia and so they were like we got a puny god right.
45:35
We got to do a complete rewrite of our god and they did and exactly and >> [laughter] >> And so it went from a puny kind of regional god to the god of all gods, the god that contains all gods who created the skies and the earth and everything. And it worked. >> Yeah.
45:58
>> Yeah. I [laughter] and also great Neil Stevenson sci-fi book on that one Snow Crash where >> Babylonian and uh they inserted that meme set into computer programmers and I sometimes jokingly wonder like how much of the effect of altruism and rationalism and effective accelerationism is actually just Stephvenson's Babylonian mimemetic cult in our own interpretation where it was like we kind of imagine the ways that
46:30
linear programmatic deterministic thinkers would operate and it's the perfect meme to have massive distribution and now here we are thinking do I actually want to live in a bentham utilitarian everything has a utility maximizing approach I don't think I want that I I I'm not a utilitarian >> nor am I I do like I he had he had real style though in having his body preserved and I believe it's still on display at some college in London. Uh
47:02
Uh Jeremy Bentn um and uh and when you go back and you look at what he came up with and you've got to do it with context.
47:17
I like you am not an adherent of utilitarianism.
47:21
But when you go back and look at his panopticon, right? >> Yeah.
47:26
>> The the reason he recommended a panopticon was at the time prison conditions in the UK were horrific.
47:38
And his idea in that context made a lot of sense. Right. >> Yeah.
47:43
So it doesn't mean that when you switch context to the modern world, well then the panopticon is doesn't make a lot of sense at all.
47:52
It becomes the allseeing eye of Sauron and you know a couple of hands control it and we all live in a totalitarian nightmare.
48:00
Um, but I'm also fascinated by the idea of because of what we're talking about right now, like invented history.
48:12
I I'm sure you're a fan of 1984, you know, the the party's last and most enduring command was ignore the uh evidence of your eyes, ears, etc.
48:23
, and and believe what we tell you.
48:26
And I I I've been thinking a lot about cognitive security.
48:29
Like yes, >> I I I again, this is coming from someone who is incredibly pro- AI.
48:35
I think it's a technology I've been dreaming about since I was a young man and and yet it will be able to do persuasion at scale, right? >> I think already is. >> Yeah, I agree. I agree.
48:53
And you know, propaganda 101, it's real simple.
49:00
Repetition, repetition, repetition.
49:04
And you know, people then suddenly take that into their ideas that can't be questioned.
49:10
And and how do we how do we prepare for the onslaught again?
49:16
Because we're back to humans, aren't we? Right?
49:18
I if I'm an evil genius and I want to control the world, I'm going to use these tools to their maximum effect, right?
49:26
I'm going to have the psychographic profiles of my target audience.
49:31
I'm going to try as best I can to have them begging me to protect me, to keep me safe, and I will give up any degree of freedom to to feel safe.
49:41
And then what do you think?
49:44
Well, maybe there'll be the the the white hats who are like, "No, no, [laughter] this is what you've got to do."
49:51
But I don't know that our the average human right now >> is ready for what's coming on the cognitive front. >> Yeah. Yeah.
50:05
>> Yeah. Yeah. And it's um it's actually quite scary I think because I I I guess I consider myself a creature of the internet and so I'm inoculated against all sorts of uh you know there's been a lot of techniques over the years and the
50:20
deep dark recesses of the internet has some pretty gnarly things and you know stuff that percolates up from 4chan up into Reddit and then over to Tumblr and then up to Instagram and then suddenly you know it's on Facebook and everyone has access to. >> Well, if it's on Facebook, then okay.
50:36
>> Well, if it's on Facebook, then okay.
50:39
>> Yeah, that must be real, right?
50:39
And [laughter] I I think that we're I think younger people may actually be better equipped and this will be a problem that where it's like you know millennials and Gen X and boomers like we will all have trouble realizing the degree to which we're swimming in you know different mimemetic propaganda streams and uh you know the uh al Gen alpha will have been simply born into it and and they'll be like, "Oh, yeah, no."
51:10
But they'll they'll bat away all of the problems because they'll understand it in a way that we just don't.
51:15
But I do think that one of the best ways to manage cognitive security is simply to question everything that comes at you.
51:22
And then this is infinite loops, right?
51:24
If I don't question everything, I will continue and I will never get out of the set of noble lies that I don't even see that, you know, I'm a fish and I don't know that the water's nice. Yeah. Yeah.
51:39
And uh the the the the illustrating that and getting people to actually do it, that's the problem that I see.
51:50
Um and you know, when when you're in a liberal democracy, uh you have a lot more attack vectors available to bad actors, uh than you do in a more closed society.
52:05
I I I over the weekend I repo posted a thread I did on for lack of a better word cognitive security and the ability of uh mimedics to to really mimetics are very powerful >> um and anyway but then I thought you know what I should give this thread that I wrote I don't you know last year or two years three years ago to our >> to our AI and tell it hey tell me where I'm wrong.
52:37
And one of the things that it came back with is you're only looking at one side of the liberal democracy, the attack vector space, which is larger, but it also has a much larger corrective space. >> True. True.
52:55
>> But the corrective space requires that there aren't all sorts of regulations telling you, "No, no, no, you can't you can't say that." Right?
53:04
And so it it it is a conundrum.
53:08
Uh it doesn't make me less optimistic.
53:10
It it makes me just more aware of we are really going to be facing a a ton of uh challenges.
53:16
I would say um that we we really haven't in a compressed amount of time.
53:27
>> I mean the industrial revolution was one thing but it took like a hundred years, right?
53:31
people people had the opportunity uh the the weirdest people in the world the book talks about >> cultural lag right it takes human and society a time to adopt right >> and I I like your thought about uh Gen Alpha my I have all my grandkids are Gen alpha and I was with uh my grandson who's oldest last night and he was like correcting me on a bunch [laughter] of stuff he's like oh no papa it doesn't work that it works this way.
54:01
So, I think you're I think you're right about that.
54:05
But I also think like do we get to a certain stage in life?
54:09
Uh let's take TV news, right? I am a marginal boomer.
54:13
I like to argue that I was born in 1960 and the traditional definition of boomers goes up to 1964.
54:24
But there's a the demographer Jonathan Pontel uh posits that the people born like between 56 and 64 virtually share nothing in common with the boomers from 1946 through 195 whatever.
54:38
and and and so boomers I all of I'm the youngest so all of my uh older siblings still watch TV news all the time and they believe it that because they were brought up you know on Uncle Walter Kronhite and that's the way it is >> yeah right and and like I made a decision in 2010 I'm not going to watch any TV news anymore just because you could see that like that's patently false, right? Yes.
55:16
especially when when it touches on something you know a lot about or I got interviewed a lot back then and I I would look at what came out of the of the grinder and and I just got to the point where like you know if this is what happens to something that I know I said and then it was done this way I am opening myself to just huge bias if I continue to watch TV news.
55:44
So literally I I don't I mean I see clips, right?
55:49
Anything that's put up on Twitter um but um one of the effects was like a month later it was like this brain fog lifted.
56:01
It was it was really quite exhilarating and and but you like you're not going to get people who are in the habit of keeping their favorite propaganda channel and that's the other thing right like I'm not saying this in a partisan way at all because [clears throat]
56:17
you know MSNBC is for the left and the blue and Fox is for the red and the right and they're both propaganda machines right and again I even try to say you know if you look back at the revolution That's the way it was back then, too. And so, it's not like it's anything
56:33
And so, it's not like it's anything really new, but trying to navigate, especially for the different generations, I think it's going to be a cure task because, you know, I don't know if you uh you're probably too young, but have you seen the movie Network? >> Oh, yes, I have.
56:52
My mother loves it, so Okay, she's gonna taste.
56:55
So, so Howard Beal, the the uh increasingly insane newscaster, "Turn off your TVs right now. Turn them off. Turn them off."
57:04
And I'm like, "Yeah, >> but what are we tuning into after?"
57:09
And and this is where context and information provenence and how do you even become a critical thinker where you know we had things like civics lessons and uh you know media literacy and those turned out to just be you know different variants of you know different angles of propaganda.
57:33
And I was recently in a hotel that had uh Iranian state TV, Russian today, Russian TV, uh all of the European channels and Fox and and you know, CNN.
57:46
And it was just such a strange thing to go from, you know, Iran, Russia, America, and realizing, you know, everybody is always consuming some different perspective.
57:59
And maybe that is ultimately the new context of the panopticon is that we actually might be outside of it looking in. >> Yeah.
58:10
I I think it it's it fascinates me because as you alluded to about you, you know, you can't um you can't analyze the water in which you swim, right?
58:23
uh and and there are a million probably many million strings that attach us to the culture in which we exist, right?
58:33
>> And so I I I I just think it's fascinating.
58:36
I think we will come up with techniques that make it easier for the average person reading or watching or whatever.
58:44
But I I think it's a huge huge issue that uh a lot of bright people like yourself should be thinking about because um you know I think that the the next 10 years are are going to be quite tumultuous.
59:03
Um >> the great weirdening has accelerated.
59:04
I I guess maybe the accelerationists are simply pointing out the facts on the ground that things are going faster.
59:16
And it doesn't appear there.
59:16
I I have not seen a handbreak for it.
59:19
I mean, you can't stand a fort history and yell stop, >> right?
59:24
Um so we we must, as we always do, muddle through, [laughter] >> carry on. >> Carry on. Stay calm.
59:32
Um don't panic as good old Doug Adams would have it.
59:37
Um let let's move on to what is exciting you right now on the investment front?
59:45
Oh, well, gosh, I have had spectacularly good luck over the last several years.
59:52
And uh Oh, >> I I I'm I'm going to not let you be modest and say you've been very talented over the last several years.
1:00:00
>> Well, it certainly helps to be early.
1:00:00
Uh it it always seems a little bit silly to say that I'm a fan of uh the movie Margin Call.
1:00:08
There's only three ways to to make a buck and you know it's just a heck of a lot easier to be to be first. >> Yeah.
1:00:15
>> And uh my risk tolerance for the highly unusual uh has I think stood stood in goodstead that um of of the companies that made it.
1:00:26
It was it was the wackiest ones.
1:00:29
uh in in in particular uh you know I I mentioned nuclear earlier where simply everyone that I talked to said there's just no way nuclear can go any faster.
1:00:38
Small modular reactors are absolutely not going to happen and if they are going to happen it's certainly not going to be you know youngsters.
1:00:45
It's got to be a old serious team.
1:00:47
And uh I was uh able to work with a young founder from Idaho, uh Isaiah Taylor, who founded Valor.
1:00:55
And yeah, it was it was three three years to get to the point in which he was able to, you know, this is the summer he raised, you know, a billion dollar round from Sequoia.
1:01:07
Uh and you know, I was there when it was he hadn't even started it yet.
1:01:11
He was on his prior startup.
1:01:11
And to see something happen in the physical world that fast was just an astonishing process cuz it was like 24 months of everybody thinking that we were absolutely crazy, nothing was going to get done and then a couple regulatory hurdles were removed.
1:01:30
hurdles were removed. the government became modestly more like in instead of being you know in the way it actually our national scientific labs are quite helpful and actually turn out uh you know we have amazing scientists working for us as Americans and seeing how quickly all of these young nuclear companies you know not just Valor which is where I
1:01:52
invested but you know we we had multiple reactors that went critical and were able to you know start producing energy by the 250th birthday after it was like a year in which regulations changed to we turned things on and I had not previously experienced that speed in anything other than an internet or information or financial company before
1:02:19
and to see like what's functionally I I shouldn't call it a factory because they're not really factories yet but seeing you know a reactor get built from scratch in a year was just wow actually the world is so much more malleable than we give it credit for and I'm not necessarily sure that it will always remain that way but because we could do so much more of the
1:02:44
work of the you know the paperwork and the research and coordination and all of the work that goes into building a business is now expedited so significantly by artificial intelligence tools that the actual human work goes faster because it's it's you know Steve Jobs and you know computers like ride a bike and >> it's a lot faster to bike. >> I grew up in a world in which
1:03:07
>> I grew up in a world in which things only happened fast on the internet or in finance, but you certainly didn't build hard infrastructure quickly and to know that you can do that and that so much of the slowdown was a function of rules and regulations and didn't necessarily make anything safer or better.
1:03:30
might have been the most important lesson that I got over the last three years is that someone young and motivated with great tools >> who really wants to get something done can find the people who want to support them.
1:03:43
And I was just so impressed simply by watching, you know, American scientists who, you know, work for our national laboratories kicking in and saying, "Well, heck yeah, we do want cheaper power and nuclear is great and clean and we have so much demand.
1:03:54
Let's turn on, you know, how however many reactors can do it.
1:03:59
Like we're going to commit the resources of American scientists to actually helping companies, you know, build more reactors, which for me was just, oh wow, you know, we can build websites fast is one thing.
1:04:12
You know, we can build nukes fast is a whole other ball game. >> Yeah.
1:04:17
And and it go it goes back again to our earlier chat about you know if the if the society and the zeitgeist in which you grow up uh is one where nuclear power like when I was a kid it was just axiomatically true that nuclear power was incredibly dangerous and there was no evidence there were that was that would sustain that conclusion.
1:04:50
>> Yeah, >> there was evidence that was like we got to be really careful about our safety protocols and whatnot.
1:04:54
Uh but the Chernobyl and Three-Mile Island um and and more importantly the fictional accounts of Three-Mile Island >> turn society uh vehemently against nuclear power.
1:05:14
and how many decades of lost power resulted from that and and you know I kind of think we're doing the same thing with data centers right now.
1:05:28
>> I am concerned that that's what's happening as well.
1:05:29
And it is only an accident of circumstance I think that we were able to pass the right to compute law in Montana and that I had initially put it forward as something for cryptography that uh we want to protect things like privacy and autonomy and that had been the original you know thought of well you know we should you know have a right to compute in the same way that we have you know a you know, to to speak your mind.
1:06:00
And you know, and there are so many other approaches that one could take to any technology.
1:06:10
But if we don't start from a place of thinking how can we develop actual safety, we'll never get the chance to do something because we'll never build the thing to find out the problem. >> Yeah.
1:06:22
the the the precautionary principle uh has has has literally >> sent various societies to ruin like literally to ruin.
1:06:36
>> Well, and I think is the precautionary principle in nuclear really did set us back that we could have much cheaper power and we wouldn't have had the fallon effects of you know how is you know carbon input into the system affecting our weather.
1:06:50
uh you know is is global climate change something that we could have avoided entirely if we had simply gone on a you know this different route.
1:07:01
Would we have different problems?
1:07:04
You know maybe uh it would have gone you know in a somewhat different direction.
1:07:08
I um Apple TV just continues to produce great science fiction and their um for all mankind just had a totally different set of problems because they invested in nuclear. Yeah.
1:07:22
>> And I just so sincerely hope that we don't get that problem with artificial intelligence that out of fear of all of the possibilities, we have a precautionary principle that stops us from something that could really do genuine good. >> Yeah.
1:07:39
Um again, history [laughter] gives us mixed results.
1:07:46
Um, and you know, I I think it's great to illustrate what happened like Chinese invented gunpowder, right?
1:07:55
And they didn't make use of it the way the West did and suddenly you had this sea change in the power dynamics of the world. >> Yeah.
1:08:07
and and and I think that thinking that you're going to be able to prejudge which technology which innovation where it's going to lead I think is is delusional honestly but but I at the same time I very much believe that we should as we began our discussion I we we should have honestminded discussions about the potential of what might happen.
1:08:41
Like 80% of it could be wrong, but and and obviously those discussions should be ongoing.
1:08:47
It shouldn't be like we're going to do a big seminar and then we're going to publish the results and then we're going to be done with it. No. >> Yeah.
1:08:57
>> It should it should be an ongoing process that that brings together a diversity of minds and points of view.
1:09:06
Um >> well because when they speak about alignment you know the next question everyone asks is aligned to what and then we have to answer the question of well all of us and if there's only so many models and you know we've got such a strong regulatory stance that uh you know you've got five choices which is better than one choice but it's probably not necessarily what if in that five you would prefer to have your own model and not a hosted
1:09:33
model and you know how do we make sure that people have the tools that they want for their own job and that you know you have a right to be able to build and own the fruits of your work and >> I'm just and this is where I think open
1:09:49
source you know just remains the clearest indicator that you know the internet and most of our you know all of our computers are built on the body of work of open source you know most of the world is powered with Linux. >> Yep. And Microsoft did everything they >> Yep.
1:10:06
And Microsoft did everything they could to kill Linux, right?
1:10:12
>> Yeah, they really did.
1:10:13
>> So people who are looking at like the labs now, the Frontier Labs now, it's like this just it's the same playbook.
1:10:21
And you know, if you look at it from the point of view from their point of view, well, of course, like [laughter] I'm gonna if if I'm going to be that type of person, I'm going to go for regulatory capture all the time because it makes things a hell of a lot easier for me.
1:10:37
Um, but aren't we lucky that Microsoft lost >> the attempt to make Linux illegal, immoral?
1:10:48
I mean, if you go back and find some of the videos and some of the stuff they were writing about it, again, this has all happened before. >> Yeah.
1:10:58
>> And and we should study what happened before because no, it's a different use case.
1:11:04
It's a different technology, but it is broadly applicable to where things turned out.
1:11:11
And I I with you think that open source has to be a critical component of AI development um if we want to get the best out of all the various models.
1:11:22
Le let's shift gears um because I've been fascinated by your writing.
1:11:29
Uh, you're on day 289 of continuous writing, which I I mean, literally I applaud because I I I think that writing every day is really important to tell yourself what you think about things, right?
1:11:48
Because like sometimes when I'm really an idea is just kind of bouncing around in my head, I always go to write it out because when I try to write it out, I'm like, "Oh, you're just an idiot.
1:12:02
This is this is a really bad idea."
1:12:05
Uh or conversely, boy, this is a great idea.
1:12:09
Talk talk a bit about um the experience.
1:12:11
What's what's changed >> from day zero to day 289?
1:12:21
What what has it given you this exercise in daily writing that you anticipated, didn't anticipate, etc.? >> Yeah.
1:12:30
Uh I really recommend a daily writing practice for everyone.
1:12:33
You know, if if it's an allinteresting, just just start.
1:12:37
Uh don't don't apply too many rules.
1:12:40
When I started, I thought, you know, I I would be happy if I got a couple paragraphs down and I was aiming for 30 days.
1:12:48
And now I'm on year five.
1:12:50
Uh I I began it in in 20 uh the first January 1st of 21 um as a sort of mid pandemic of like what's happening? What am I thinking? The world's changing.
1:13:01
And I tend to view writing as a thinking process that if it's in my head, okay, not so bad.
1:13:08
uh but I need to get it on the proverbial paper to actually understand where my understanding gaps are.
1:13:17
And uh I'm so glad that I did it.
1:13:22
Not only because everything is tagged and you know there's lots of useful metadata and you know what my you know areas of interest are, how they've changed uh but it's also a wonderful way to discover where you were wrong. >> Yes.
1:13:36
And we we have a tendency to think of uh problems as um the stuff we got wrong doesn't matter.
1:13:42
It's only the stuff we got right.
1:13:43
And I think that that's very much what I've taken from it is that I am wrong so often that it's almost embarrassing.
1:13:52
And I I don't ever want to let go of that feeling because now I'm no longer embarrassed by a poor writing or, you know, shorter writing or having a bad writing day because part of the reason I do it is I want to discover the areas in which either my thinking is flawed or I haven't really thought something through.
1:14:13
And now that I've got such a large body of work, I can toss it into an LLM.
1:14:19
And honestly, uh, coming up with a duplicate of me would be pretty simple just based on my writing now because there's just so much of it, >> but [laughter] it's very helpful.
1:14:30
I I think the thing that you said that I absolutely applaud uh as one who has kept journals since I was 19 years old.
1:14:40
Um the ability to identify where you were wrong is so important because our memories overwrite, right? Yes.
1:14:51
Like [snorts] with what they overwrite with what you're believing.
1:14:56
Now, I've told this story before, so I won't go into the full story, but I found innumerable examples of I would have gone to court, put the hand on the Bible, said, "Absolutely, this is true, and I would have believed it until I was called a liar in my own handwriting many times." >> Yeah.
1:15:20
>> And and so that unlocked so incredible. >> Yeah.
1:15:24
I mean, I can't imagine having I I hope that I get to the point in which I've got that many years of writing because I I do intend to keep up the practice.
1:15:32
I don't know if I'll always keep it public.
1:15:33
Although, I think keeping it public has been very helpful for me just as a um you know, I say I'm going to do it so then I have to do it and like that continues on.
1:15:42
But it is kind of shocking how much your mind glazes over the parts that it doesn't really want to remember or you know wants to remember in a different way.
1:15:55
And it's all like very um because you know I you you always think oh gosh no my principles are always so consistent.
1:16:02
I've always thought this way and to have a record where you're like no this is actually where I changed uh you know I've got many many posts in which I go back reference and say you know this is where I have to take accountability. I got that wrong.
1:16:14
Uh but also uh you'll appreciate the areas where you're right so much more. [laughter] >> Yes.
1:16:22
But it if you want to improve your models of the world, right?
1:16:26
If you want to have higher accuracy, mistakes and being wrong are the portals of discovery.
1:16:35
And >> well, and just the idea that action opens up a new path and that thinking and committing to thinking and writing is actually in sort of that wonderful collapsing the state action of now it's on paper or you know screen and that enables you to then take the next step and the next step and the next step from there and I I think if we don't take action, we'll never have agency on our own lives.
1:17:07
And writing is for me some of the first steps in being able to actually say this is what we're going to do >> or this is where I'm going to focus or this is what's important.
1:17:18
>> When when you put um the 2008 pieces into a large language model, what what came back that surprised you about yourself?
1:17:30
Uh well, it turns out uh I really have a lot to say about travel uh which it sounds like a really silly thing, but uh when you're on the road a lot, you pick up all sorts of little habits around packing and routines.
1:17:43
And it turns out I actually am quite a creature of habit.
1:17:48
So even if I'm working, you know, in fringe and strange areas, you know, where things are very, very new, it turns out I am really rigid about that this is how I pack, how I schedule, how I keep things together.
1:18:03
And um the LLMs were definitely like, yeah, you might have some some emotional stuff around travel and packing just because this is how many times you've written about packing.
1:18:15
Uh, and those like silly little self-nowledge things where, you know, I tend to think of myself as being very open to new experiences, but boy, uh, revealed preference might suggest, uh, I like things a certain way when it comes to travel. >> Yeah.
1:18:31
>> And it's just kind of a silly one, but >> it does give you insight. >> I I've been doing it.
1:18:36
We're going to build out we're going to digitize all of my handwritten journals.
1:18:39
So, that's 40 plus years.
1:18:42
Um, and my joke is I'm going to have the AI explain me to me.
1:18:48
[laughter] >> Yes, >> Jim, you are a lunatic.
1:18:53
[laughter] >> Well, and this is where I, you know, only having a couple models and personalities could could really be detrimental to self-standing because sometimes they, you know, they're so, you know, people know that they're very obsessquous.
1:19:07
It's like, oh no, this is so like delightful.
1:19:09
And you know, you have to coax out the like, no, I want to understand where I fail.
1:19:13
I want to understand, you know, where my thinking is caught.
1:19:17
And being able to rely on the chain of thought that will give you the most insight, you know, it's almost like thinking around yourself.
1:19:28
And it's very hard to do, right?
1:19:31
Other than a mirror, like I don't know any other activity that gives you that full perspective. >> Yeah.
1:19:40
And uh the the uh you're right about the syncopantish nature of a lot of the large language models.
1:19:48
You can build prompts that make them not be that way.
1:19:55
>> I I I'm just completing my first project of of trying to I've written four non-fiction books, but I've never written a fictional book.
1:20:02
And uh so we're we're trying it uh with a writer room with AI editors, writers, uh human writers, editors ultimately it's mostly being generated by me.
1:20:16
But the absolute best prompt I uh learned from and changed a lot even of my storyline was uh an elaborate prompt which gave the the AI it had read everything about the art of writing and it also had huge domain knowledge in you know what makes a great fictional thriller.
1:20:40
It's a fictional thriller that that is a generational thriller that I'm writing.
1:20:47
Um and and then I told it so I gave it endowed it with all those powers and then I said and you hate this book [laughter] and and oh my god the the it was vicious.
1:21:04
I mean like literally vicious.
1:21:07
But what I found really cool was I thought about it. My wife edited.
1:21:11
thought about it. My wife edited. She graduated Sumakum laad and journalism and is a great editor and she edited and I I would say even co-wrote um my books and when a human especially a human that you love comes in and hands you something that is nothing but a sea of red on paper you get it you get emotional you get emotional right and and after I read the first one which was
1:21:40
I should have kept it but it was so vicious that I literally was like first I was just crestfallen then I laughed then I actually started looking at what it was pointing out and then I came to the conclusion it's right it's right and and I didn't have any emotional baggage because you know you bad large language model that I told to be mean to me [laughter] >> yeah So, >> what you asked? >> Yeah. Yeah, you asked, dummy. But it >> Yeah. Yeah, you asked, dummy.
1:22:14
But it honestly was the most fruitful of all the feedback that I've got.
1:22:22
Um because it when when you tell it and and obviously you can't just hate on me.
1:22:30
You've got to you've got to develop the reason why you're hating on me.
1:22:32
It's like I I remember one of the critiques was, "Wow, Jim, this is going to do really well in the bargain bin at Walmart."
1:22:43
[laughter] >> Hey, I fish some good stuff out of bargain bins now and again, [laughter] >> but ouch. >> Yeah. >> Um, okay.
1:22:54
I I see that our uh we're coming up on time.
1:22:59
What what right now if if if you could pass along an insight um and let's let's keep it on on what's happening on on the frontier on on investing where what where would you tell people to look that isn't and is we've already covered nuclear um and most of AI are there areas of AI like for example we're developing something in our AI lab that is a pixon shovel.
1:23:29
Um, and like all of the business people I showed it to are like um and until they start seeing what it's going to allow you to do in terms of speeding up GPU access uh all of that sort of stuff.
1:23:49
Um, are there any is there anything right now that people are not looking at that you are and you're like, "Look at this. Look at this."
1:24:01
Well, I'm still convinced that we actually haven't incorporated the vast majority of of human context and information that digitization is a much larger effort than we're giving it credit for.
1:24:12
credit for. that just remains I I think partially true because the large language models that we're using are dominantly western trained on English and you know there's whole bodies of text you know cerillic takes
1:24:28
more processing like GPU power uh than you know latinate language uh it's just there's less uh I don't know it turns out English is actually quite good uh at putting forth information and There's lots of written records that don't come in that format. But I also
1:24:45
But I also think that we haven't really effectively priced compute as well.
1:24:52
>> And I'm I I think that's going to become uh you know, we've been subsidized all of this time.
1:24:57
We've had a lot of lovely free compute as the labs have, you know, encourage us to use and adopt the products.
1:25:02
And uh you I remember when Uber was subsidized and how lovely that was.
1:25:06
Um, so you know, thank your local VC for that, but we probably do need, you know, spot compute prices and and that's going to be an issue that has to get solved because, you know, back to Carlaua Perez and, you know, we are in the innovation cycle, but the deployment cycle always cares about price, always cares about efficiency and deliverability.
1:25:26
And if we don't price it, we can't sell it.
1:25:28
it, we can't sell it. And the pricing is now at a you know fairly abstract place away from the consumer which might even be part of the reason that like oh data centers ah bad where they don't they don't pay their compute bills they pay
1:25:44
their electricity bills and as we start to get a better understanding of what the costs are maybe it shifts some opinions and we actually put resources where they can do things that you know people feel better about what they're being used for. Yeah, I think that um the the number of
1:25:59
Yeah, I think that um the the number of new markets, of new companies, of new methods and protocols, it's it's going to be something to behold.
1:26:12
Um so I I completely agree that um when when people are aware of price, they're like, "What?" [laughter] >> Yeah.
1:26:22
Well, if you don't pay the price, you know, why wouldn't you just use more and more?
1:26:26
I mean, we see this in every market.
1:26:28
You know, Americans don't really price their health care and wow, health care costs are up. Surprise. >> Yeah. Yeah.
1:26:34
Well, Julie, we're going to give you another uh swing at becoming empress of the world.
1:26:39
Um you you you are only this for a day, and you can't re-educate anyone in a camp environment, no matter how nice that camp happens to be.
1:26:49
Uh and you can't kill anybody, but you can.
1:26:52
We're going to give you the magical microphone again.
1:26:54
And you can in say two things into it and incept two things into the entire population of the world.
1:27:02
They're going to wake up whenever their next morning is and think, you know, I've just had two of the greatest thoughts and unlike all the other times, I'm going to actually act on these two.
1:27:12
What are you going to accept?
1:27:16
>> Well, [sighs] I did think about it ahead of time because the last time I went with very basic uh suggestions, and I do love basic.
1:27:21
I still think it it's good to be nice to people and it's good to squat every day.
1:27:28
But on this go around, I'm going to say that knowledge is survival made durable.
1:27:33
That if we do not >> contain the knowledge of the ages, our capacity as a species is limited.
1:27:44
That making sure that knowledge survives is what makes sure that humanity survives. >> Lovely. Okay. So, what's number two?
1:27:56
>> Well, eat more protein. >> Okay. Both good advice. Both good advice.
1:28:03
Uh and and and where where can obviously you're on all of the various social media platforms and you have your own uh you have your own website and you have the chaotic website.
1:28:14
Uh we'll include all of that in all of the show notes.
1:28:20
I'm very easy to find on Twitter or I know some people call it X but and I'm an oldtimer on Twitter so it's always going to be Twitter to me.
1:28:26
[laughter] >> Yes, us old-timers are going to continue to be tweets I think. Yes.