0:00
I'm now recording.
Search this video
0:00
Jesus, you look terrible. What's going on?
0:03
Did you sleep last night? Do I look tired?
0:05
You do look a little tired. look unshaven.
0:06
Yeah, what happened last night?
0:09
I had a holiday party at my house last night for my office and You look destroyed.
0:12
Yeah, what's going I mean, did you drink? I did drink, yeah.
0:16
What does vodka and oat milk taste like? HE HAD A WHITE RUSSIAN.
0:22
A WHITE RUSSIAN, BIG LEBOWSKI STYLE. WITH OAT MILK.
0:24
HE'S LIKE, "I WANT a white Russian with oat milk." Oh my gosh.
0:27
Actually, my white Russian is made with oat milk.
0:29
And please compliment that with a huge punch in the face. Give me 5 minutes.
0:34
I'm going to go I'm going to go shave and hair of the dog. Yeah.
0:37
You have a little Bloody Mary.
0:40
You're better off with a beer. Better off with a beer. Should I go get a beer?
0:42
I'm going to get a beer just to beat this hangover. Go do it. Yeah, go get a beer.
0:46
Hang on, I'll be right back. All right, listen.
1:05
We have to start with scam, bank run, fraud.
1:07
I mean, Sam Bankman-Fried.
1:12
I get that wrong sometimes.
1:12
He was interviewed by Andrew Ross Sorkin, the suit, at DealBook, who gave him softball after softball.
1:20
Then, the next day he was on Good Morning America.
1:24
passive-aggressive right now. bit.
1:26
You're throwing a little shade here at our friend Andrew Ross Sorkin. shade.
1:29
He's sticking the knife in.
1:29
He's sticking the knife in, but J Cal's got a point. He's got a point. A little bit. A little bit. A little bit. A little bit.
1:37
Anyway, The New York Times continues to embarrass themselves by handling Sam Bankman-Fraud with kid gloves. Wow.
1:44
George Stephanopoulos, my Greek brother, Stephanopoulos the Spartan, came in and absolutely fricasseed and filleted Sam Bankman-Fried on Good Morning America.
1:55
Very important to note that Good Morning America, the segment was between holiday cocktails and the cast of The White Lotus.
2:02
And Andrew Ross Sorkin was at the DealBook finance conference.
2:07
But you know, that that that Stephanopoulos interview was 2 hours, but they reduced it down to 10 minutes.
2:12
So, there were probably a lot of sort of cordial conversation, banter, and then softball questions.
2:19
And then, he does stick the knife in and and does the fricasee, but he cuts out all the other stuff, so he just gets it down to the 10 minutes. Absolutely.
2:28
What Stephanopoulos did to him was extraordinary in that he said over and over again, "In the FTX terms of service, you cannot touch the user accounts, but you at Alameda were taking them and you were loaning them out."
2:43
I don't know who is advising Sam Bankman-Fried at this point, but why he is talking so much.
2:48
He was also on a 2-hour Twitter Spaces after all this.
2:57
At this point, Sacks, what do you think is going on here in the mind of Sam Bankman-Fried and also the media, which seems to have a very variable way of dealing with this obvious fraud and crime? Right.
3:10
Okay, well, you know, I can speculate about SBF.
3:12
I think if there is a strategy here, it is this.
3:15
He is basically copping to criminal negligence in order to avoid the more serious charges of fraud.
3:25
And I think, again, if there's a strategy here, it is he saw himself being defined as Bernie Madoff 2. 0 in the press.
3:31
And if that image, which may well be true, cemented around him, then prosecutors would never stop.
3:42
They would never accept a plea that basically gave him anything less than a Madoff-like sentence, which would be, you know, decades in prison, maybe a life sentence.
3:52
So, he is out there doing what lawyers would tell you never to do, which is basically incriminate yourself, create more of a record, but he's doing it to change the public perception, maybe muddy up the public perception, get people thinking that, okay, he, you know, this that he he's admitting he did something wrong, but it wasn't deliberate, it wasn't fraudulent, it was just basically carelessness or sloppiness.
4:16
And if he succeeds in muddying the waters enough, then maybe the prosecutors will give him a plea deal that allows him to have his life back at some point.
4:23
I think that would be the crazy like a fox explanation of what's happening.
4:30
Now, there is, you know, an alternative explanation as well, which is I just think that these types of guys, you could call it, you know, a narcissistic fraudster, they think they can talk their way out of anything. You know? Because they have. They have.
4:46
You know, And they've talked their way into getting hundreds of millions of dollars of investment. Billions. Billions in some cases.
4:51
And so, they just feel and they've been trained by employees, partners, investors, the press that they can talk their way out of any of this stuff.
5:03
Yeah, the average person is not really used to dealing with one of these personality types who is I mean, they clearly are smart and they're articulate articulately know what to say and they're crafting their words.
5:14
What they've learned through their life is that if they use the precise magic words with a person, they can pretty much, you know, convince them of anything, get them to do anything.
5:23
And in particular, I would say investors tend to fall for this, not because investors are dumb, but because investors are so clear about what they're looking for and what they want. They're predictable.
5:32
Yeah, like we're, you know, VCs especially, we're looking for the 100x outcome or whatever.
5:35
So, it's easy for this type of personality to construct a story to essentially stroke the erogenous zones of a VC. Whoa.
5:45
Yeah, and and and sort of trick them.
5:48
And so, there is probably a positive reinforcement loop that gets created in the minds of one of these people who and they start to think that they can basically talk their way out of any situation.
5:57
So, I think that would be part of what's going on here and you know, if you look at sort of the tactics that he's using to do this, you know, all of a sudden, he's trying to portray himself, you know, before this, he was portraying himself as the smartest guy in the room.
6:11
Now, all of a sudden, it's this babe in the woods impression where, "I didn't know. It was my subordinates.
6:17
I wasn't really in control. That was somebody else." Right.
6:22
Each individual decision, he says, looked sensible to him.
6:24
It's just it all added up to something he didn't anticipate. Like, really?
6:30
You know, loaning yourself a billion dollars of basically the company's money, which was basically customer money.
6:36
That seemed reasonable to you.
6:37
I don't know how you defend that individual decision.
6:39
And there's many like that.
6:41
But but this is sort of the narrative that he's trying to construct.
6:45
And let me stop there, but I think there's a lot more that can be said to dismantle the narrative he's trying to create.
6:51
Sacks, what's your what's your take on this and then Freeberg and Can I make a comparison to SBF and Trump through the lens of the media?
7:00
So, if you go back to 2016, you know, Donald Trump violated every single establishment bias that these left-progressive journalist elites had.
7:14
And so, they basically just attacked attacked attacked attacked.
7:18
But then you went into the election and there was a very clear data point that said whatever you thought was at was at best limited and you missed the tone of the country.
7:31
Because 50 plus percent of the country held a very different view about this person.
7:37
And instead of taking a step back and and then the left media, the mainstream media, re-underwriting and learning and then saying, "You know what?
7:46
Mea culpa, I got this wrong." They just doubled down.
7:50
And they said, "No, it still doesn't meet our priors and so we're just going to ring-fence this problem and we're going to just try to destroy this issue because, you know, we want to control the narrative and and and by result, we want to control power."
8:05
Now, you look at SBF, it's the exact opposite.
8:07
He went to the perfect elite private high school.
8:12
Then he went to one of the most prestigious elite private universities. MIT.
8:17
His parents teach governance of all things at one of the most elite liberal institutions in America. Stanford.
8:25
They are in the establishment of the progressive left.
8:30
And what happened was, he took customer funds and all of this money, he made tens of millions of dollars of political donations, he wrapped himself in this blanket of a progressive left-leaning cause called effective altruism.
8:45
And all of the mainstream media fell for it and embraced him, as well as some politicians, because it met everything that they themselves also bought into. Yes.
8:58
And now you have this cataclysmic event, a multi-decabillion-dollar fraud or bankruptcy, millions of customer accounts who are frozen, you know, tens of millions to hundreds of millions to billions of dollars lost and stolen from them.
9:14
And they refused to re-underwrite this kid.
9:17
And the reason is because in order to do so, it's like eating your own tail.
9:21
And that's why they don't want to do it.
9:22
And so, this is why you have the media basically allowing him to do an apology tour.
9:29
Now, this is his second time at manipulating them.
9:31
The first time he was able to manipulate them by basically being one of them.
9:37
And now, he's allowing them and their desire to basically protect themselves so that he can create some kind of a defense for himself.
9:45
And I just think the whole thing is gross because it misses the entire mood of the nation.
9:49
This is an enormous financial fraud that was perpetrated on tens of millions of people and there's no accountability because in order to do so the media would effectively have to admit that they missed it and they got it wrong and they refuse to do it.
10:07
And I think that that is the really big problem that nobody is really speaking out about is like well if these folks are meant to be the last stop to make sure that there's truth and honesty and transparency in society and you can't count on them and in fact they're just going to reflect their own narrative, what is one supposed to do to learn the truth?
10:28
In a way what you're saying and then we'll go to your freebird is this fraud was encased in all the gilded facade that America hates right now.
10:38
the institutional rot of America.
10:38
It reflects every single aspect of institutional rot that every non-elite talks about all the time but elites when they have those labels will refuse to give up.
10:51
And just to add to that the the the thing that's missing I'd say one of the big issues with the institutional rot in our country is the lack of accountability when somebody gets it wrong.
11:01
We saw this with COVID, right?
11:03
The health establishment is saying that they want amnesty. Yeah.
11:05
And Atlantic magazine was willing to give it to them.
11:09
So the point is that this this class of people think that when they get it wrong that they're the experts but when they get it wrong there should be no accountability and so Chamath to your point the media and these institutions are not willing to re-underwrite SBF when he so clearly is a fraudster.
11:26
Uh freebird what do you think the of this theory?
11:30
You had a large amount of donations to politicians.
11:32
Uh obviously you have coming from Stanford, MIT, etc.
11:36
And then uh you have these investments gifts {slash} advertising {slash} donations to ProPublica, Vox, uh this new publication Semafor, The Intercept that have all been uncovered now.
11:52
Did he do this paying off of all the elites you know, splashy cashy giving money to everybody because he knew he was doing a fraud and that this is evidence of this is a premeditated fraud or do you think this is a deranged individual who just was seeking status? I don't know.
12:10
The the motivate There's a video you can watch of this guy for some reason FTX has left up all of their videos on YouTube from 3 years ago called quantitative trading 7 and 1/2 thousand sell wall of Bitcoin on Binance.
12:24
It's a 17 and 1/2 minute YouTube video of SBF trading arbitrage across markets.
12:32
I think it provides probably the best like natural non-scripted insight into this guy's behavior that you could see you know, because it's not like him being interviewed.
12:43
It's just him living in his world.
12:45
And he's just you know, a mouse trying to get a piece of cheese.
12:49
Like he's you know, he's like out there.
12:52
He's you know, scrambling around in the markets. He's finding edges.
12:54
He's finding advantages and he's and he's clearly just taking advantage of them all day every day. That's who he is.
13:03
Now you put a person like that in an unregulated environment and there was this clustering demand for an unregulated environment because of a lot of what you guys are saying which is people have this disdain for the elitism and the and the institutional rot and all these things.
13:20
So Bitcoin emerged as a solution out of 2008 to the you know, the the what felt like institutional rot that that governments have a key role in but when you have no regulation and you have no trusted central authority involved mice that are trying to find cheese will rule the day.
13:41
And I think that's what happened here.
13:42
If it wasn't this guy it was someone else.
13:44
It was going to be someone else.
13:46
And then all of a sudden everyone's clamoring and saying hey we needed the government to protect us. No one protected us.
13:51
Someone's got to save us.
13:51
Where are the regulators?
13:52
Where are the people that are supposed to keep an eye on this stuff?
13:54
When the whole premise of so much of what was being sold was non-regulatory regimes, was openness, was peer-to-peer trust protocols and it turns out that in that sort of an environment the mouse that is hungriest for the cheese will get the cheese.
14:08
And that's exactly what happened.
14:09
And I don't I don't know how much of it was I don't agree with this.
14:12
him saying I'm creating intentional fraud which certainly seems to be the case versus him saying I'm going to pay these guys I don't know how much of it was even that intelligent.
14:20
But the guy was clearly like trying to get a piece of cheese.
14:23
Okay, so this cheese eater, this rat is a League of Legends uh you know, expert playing on eight different monitors at a time the cryptocurrency game while shopped up on speed. Yeah.
14:36
I'm not saying that to be cruel.
14:38
I'm saying that because they admitted it.
14:40
They talked about it in their staff meetings instructing their traders and team members of how to take speed.
14:48
He admitted to it in an interview this week.
14:50
He said that it was all legal prescription drugs but they were taking them. Yes.
14:54
Yes and literally in the same videos you're referencing people see speed patches or I don't even understand this but there are patches you can put on your body to deliver speed to you at some you know, dose or whatever.
15:06
Sacks you buy this theory? No.
15:06
Here that this isn't about the elite side of it.
15:10
It's about the non-elite the anarchy side of it. No.
15:14
And this cheese eating rat was just wanted to eat more cheese.
15:19
I don't buy this narrative because I see too much design intentionality behind what happened.
15:25
So in other words it wasn't just a series of individual decisions that didn't add up.
15:29
Many of those individual decisions by themselves were totally unjustifiable and moreover there were too many there's too much evidence of sophisticated behavior here.
15:40
Again he overnight went from portraying himself as a smart guy in the room to the to the babe in the woods.
15:45
And so for example when you look at the construction of all these entities and the corporate org chart, you know, of all the related entities.
15:54
It's a very sophisticated attempt to obscure and construct certain you know, protections.
16:03
When you look at the way that Alameda was exempted from the normal margin requirements in FTX.
16:08
There was the so-called backdoors.
16:10
There was intentionality there.
16:11
There was intentionality in terms of who was hired to staff these organizations.
16:15
Again, they hired wasn't hired. No board, no CFO. Yeah, exactly.
16:20
And and the guy who was in charge of compliance what like Chamath talked about in previous episode was the guy who was involved in the ultimate bet uh poker cheating scandal.
16:30
You know, not not exactly the super mode. Yeah, exactly. Right.
16:34
Or look at this goofy goofball Caroline Ellison who was put in charge of Alameda, right?
16:40
It's it's being done for a reason, right?
16:42
He's setting it up in a certain way and you know, the one time I interacted with him at this um tech conference he was sitting there holding court and he had all of his minions around him who were following his orders.
16:52
This was a guy who was controlling his business.
16:54
He was making the decisions at I think a task level and um he knew exactly what was going on here.
17:03
So look, I just don't buy I do not buy this idea that that he was like a blind mouse who's just stimulus response you know, in the moment.
17:15
By the way that that wasn't that wasn't my point, Sacks.
17:17
My point was whether it was him or it was going to be someone else it was bound to happen. No, I don't buy it.
17:25
that a criminal would have emerged like that the idea that we want to have completely free unregulated Bahamian-based trading you know, environments that we can supposedly trust because someone puts on a good face when there is no real regulatory body and regulatory authority overseeing it at some point it was going to happen. Well no, hold on.
17:43
It is a Look, Coinbase is a fully regulated institution They're not set up in the Bahamas and they're not unregulated.
17:48
And he set up and he had no oversight. He had no board.
17:51
There was no regulatory regime. There was nothing.
17:54
he able but Okay, so first of all look, I don't think this had to happen.
17:55
I think that again I think that excuses too much because it implies that if it wasn't SBF it'd be somebody else.
18:01
I actually think that this was a highly concerted effort.
18:04
Listen, he courted regulators.
18:06
He donated to politicians.
18:09
He courted the media and donated to them.
18:11
He was Yeah, he was really good at it. Yeah.
18:13
He was really good at it.
18:13
He was unusually good at it.
18:15
No and I totally agree on that.
18:16
Yeah, super smart, super connected, really thoughtful design on how he committed this on how he did this.
18:22
think an insider could have pulled off something at this scale.
18:24
I think See this is where Chamath you're exactly right.
18:27
I think you needed to be an insider.
18:30
here is he knew the playbook. And he admitted it.
18:32
You pointed this out with the chats that were released where he said he he. He he.
18:35
Yeah, look I mean look just like the he he chat.
18:38
Look how like convoluted and intertwined all these people are like Gensler's intertwined with the parents. Yeah.
18:44
Parents apparently bundled a bunch of money to Elizabeth Warren.
18:48
You know, he was dating the CEO of the business that he owned 90% in.
18:52
There are all these other random shell companies that he owned 100% of where they were lending back and forth hundreds of millions to billions of dollars. Yeah.
19:00
To David's point that is a sophisticated con that you have to architect.
19:06
And the way that he was able to get away with it is that not a single reporter or regulator thought to dig in.
19:13
And the reason I think is because he said all of the right things that wanted them to embrace him.
19:22
And the reason is he admitted it.
19:23
is the this is the dumb game that we woke westerners have to play.
19:25
We say the right [ __ ] bless and then everyone thinks we're a good person. Exactly.
19:29
Imagine Pull that quote up on what he said cuz it's actually a good quote. what I just said.
19:32
It is that the It is actually what he said.
19:34
He said it's it's the it's the game that we woke westerners have to play.
19:37
We say the right [ __ ] bless and everyone likes us.
19:40
said the most uncomfortable thing out loud which is look, by having gone to Crystal Springs High School, by having professors my parents that went to Stanford, by having gone to MIT I can pull this off.
19:52
That's what he said because he can go with those things because I'm a I'm a champion of effective altruism that I can justify any of these decisions how amoral or immoral that they be because I'm trying to help, you know, my brother stand up a multi-million dollar pandemic response business. I'm trying to do this. I'm trying to do that.
20:09
And all of these regulators and all of these reporters said, "Okay, you get the hall pass."
20:16
Now, imagine if you replaced him with some random kid in some developing country or even from the United States who did went to public high school who went to some random state school.
20:29
Do you think that they could have pulled any of this stuff off? No.
20:33
You need that patina of the privileged class.
20:35
The New York Times That's what he had.
20:38
The patina of the privileged class.
20:41
Even after this fraud, the New York Times wrote more of a puff piece on him than the hit piece they wrote on Brian Armstrong last year when Brian Armstrong wouldn't tow the line on allowing politics at work. Remember that? Yeah. Unbelievable.
20:54
Yeah, so the big enabler I think what Chamath is kind of saying here is that the big enabler here is not crypto per se.
20:58
It's all these institutional biases and elite biases that he was able to play into partly because he was a big insider.
21:05
I mean, in a way he monetized his parents' life work.
21:08
The problem that I think this allows us to put a fine point on is the following.
21:15
You know, in society we've confused a lot of people to think that the opposite of liberal is conservative or Republican.
21:25
And I think that's the cycle that drives the mind virus inside the mainstream media.
21:30
The problem is the opposite of liberal is illiberal. Okay?
21:32
And what illiberal means is to be narrow-minded and unenlightened.
21:37
It means to be puritanical.
21:39
It means to be fundamentalist.
21:43
And this is really what it allows us to see now.
21:45
We have now had 6 years of data.
21:48
Case after case after case where if you are woke if you are a social justice warrior, if you have the right credentials that justify your upbringing, if you have institutional bonafides that come from your parents you get to create the narrative and you get a hall pass.
22:11
And everybody else basically is at the subject and the mercy of the mainstream media.
22:15
And so, if you don't kiss the ring and bow down to them, they will try to destroy you or run you out of town.
22:22
But if you are one of them, they will give you a hall pass.
22:26
And when it's time for them to change their mind in order to tell the truth they won't do it.
22:31
And so, these types of grifts will continue as Friedberg said because there is no check and balance without a healthy independent media.
22:41
There is no way for all of us to actually know what's really going on.
22:44
Guys some person in the media could have asked the question and dug in deeper around the connections between Alameda and FTX for the last 24 months.
22:55
At no point at a venture firm.
22:57
At no point could any person have asked these questions and found ex-employees and said you know, are there any unseemly connections here between FTX and Alameda?
23:07
There was no disgruntled employee.
23:08
I mean, every company has disgruntled employee whistleblowers.
23:13
But here where there was billions of dollars being made by tens of people, not a single person who felt on the outs said anything?
23:19
Well, he was also giving millions of dollars to press outlets in donations.
23:23
the questions weren't asked and then this kid paid hush money to the mainstream media.
23:28
Let me ask you a question of you guys.
23:29
Do you think that it's the media's responsibility in this context or do you think that there should have been a regulatory authority that had oversight of this business like there is for every bank and every trading operation in the United States?
23:39
And every one of those businesses has a compliance officer and has regulators up the wazoo making sure the customers are kept safe and protected.
23:48
Or do we think that that should should offshore vehicles be allowed like this that allow people to operate?
23:54
It's a reasonable question, but there was a chicken and egg question.
23:55
We were all standing around holding our hands while the CFTC and the SEC were fighting.
24:01
That's not something that consumers can be expected to adjudicate.
24:06
So, yes, we should have legislation that clearly defines all of this, but there were enough parameters that created regulatory frameworks where a bunch of good actors did operate in them and are continuing to do so like Coinbase.
24:17
So, I don't think this is a regulatory issue.
24:22
I think that if you believe there are people who are supposed to forensically examine things and get to the bottom of things and ask hard questions those people did none of that here.
24:30
And And what's What's even more worrisome is what they're showing is now with an massive amount of data that shows that you could ask hard questions they don't care to because it makes them look bad.
24:44
I disagree with this, Chamath.
24:44
I think regulators failed here because they have been reactive to crypto.
24:48
They have not been proactive and they have not been clear with the crypto community that what they were doing was illegal and they should have put the regulations in quicker and they're playing catch-up, but all three groups failed. The media failed.
24:59
The regulators failed and VCs failed.
25:02
Capital allocators failed I apparently to do diligence here and install proper governance.
25:07
You cannot put a company like this you know, in business with billions of dollars and have no board of directors or he or he lied to them.
25:16
And I agree with you, my My point is that while regulators are are basically fighting a territorial turf war Yes. Okay?
25:25
The media could have still done their job. They chose not to.
25:26
Guys, it was worse than that because not only it's not just a case where the SEC failed to exercise any oversight of him or dig into any of these questions.
25:35
He was in the room with them crafting the next set of regulations.
25:38
He was working on the regulations that you're talking about that are supposedly needed. them.
25:44
They let the fox in the hen house.
25:47
He was going to craft a a new type of regulatory license for these types of exchanges with the result that he was going to get one and some of his competitors weren't.
25:57
This is one of the things that triggered CZ to basically, you know, do what he did. He rug pulled them.
26:03
Basically, that SBF was trying to get Binance and and competitors like that banned while SBF would be one of the sole people to get the license.
26:13
By the way, Jason, he was he was working with the regulators.
26:17
Yeah, Jason, you just said something derogatory towards CZ. He rug pulled them.
26:21
Did he do that or did he actually expose the fraud?
26:24
That's what I mean by rug pulling. Yeah.
26:26
He was partners with him.
26:26
Okay, well, he was partners with him and then he realized he was weak and that he was doing some stuff that was shady and he decided he would eliminate a partner who was creating regulation as Sacks said.
26:37
Whatever you want to say. He knifed him.
26:40
did was indicate a desire to liquidate his position in a token that was supposedly perfectly liquid.
26:45
And that's basically that caused the everything to unravel.
26:49
Yeah, but these were partners, right?
26:51
I mean, they were these were deep business partners. And they were partners.
26:54
They were collaborating on these tokens together.
26:56
So, they No, they were No, Jason.
26:58
Like part of the big loan that that initiated all of this stuff like a year and a half ago was to buy FTX off the cap table.
27:08
So, you know, this All I'm saying is like you used words in your framing of a guy, you know, who's built a business is like he is this nefarious bad actor.
27:19
Binance gave FTX its first investor. But But David's right.
27:22
All the guy did as far as we can tell right now is tweet, "I'm selling this token because I don't believe in the capital structure of this entity."
27:30
And he got those tokens by being bought out of the cap table. I know.
27:35
He was partners with them.
27:35
So, if if you If you don't like rug pulled, how about backstab?
27:38
No, that's They're not partners.
27:40
No, you still don't understand, Jason.
27:42
I Of course I understand.
27:44
They had He was the first investor.
27:45
So, to do this to a company you were bought out.
27:48
No, no, he He was bought out.
27:50
It was part of the consideration. Hold on.
27:52
Part of the consideration were these tokens, these FTT tokens.
27:54
Jason, when you when Uber went public Yes.
27:57
and you got distributed stock Yeah.
27:59
did you distribute and sell Uber at any point? Previous to that I had.
28:05
I sold some to Masayoshi. No, no, no, no, no.
28:07
I'm asking you a question.
28:08
When Uber went public and you got distributed from Uber your stock.
28:13
So, you've never sold a single share of Uber after it was public.
28:16
No, I sold it before in private markets.
28:18
Okay, but I'm sorry, but that doesn't make Jason a partner of Uber.
28:19
You're just a stock holder. Yeah.
28:21
This is slightly different, I think. But okay, fine.
28:25
You guys If you guys want to consider No, he was the first investor in the company.
28:28
That might have been a historical situation, but their status at the time that CZ tweeted was that they were competitors. Okay, fine.
28:35
I mean, they became competitors. I agree.
28:39
And by the way, to Chamath's point about this rug pulling language, I think we're getting kind of down a rabbit hole here, but listen.
28:45
What CZ CZ did perform a service in this sense, okay?
28:48
SBF claims that 4 billion more was about to come in.
28:51
I personally don't believe that.
28:53
Sounds like [ __ ] to me, but if it is true, that would have been a bad thing.
28:56
The more money that came in to that operation, SBF proved that he was a very poor custodian of customer funds. For sure. For sure.
29:05
I'm I'm not defending SBF.
29:05
The longer this went No, no, I I'm just highlighting the language that you use is sort of like again part of that establishment elite narrative and I'm just questioning you should maybe steel man.
29:16
Take a second to just steel man a more dispassionate view, which is here's a counterparty, okay?
29:21
Who when he left the cap table was given half cash, half tokens. Okay.
29:29
And he decided to sell his tokens.
29:33
Yeah, and tweeted publicly and caused a run on the bank.
29:35
It was not a run on the bank. There was no bank. Where's the bank?
29:39
This run on the bank language, okay, is something This was in the semaphore coverage, okay? it publicly. That's my point.
29:46
So, I'll still maintain that no.
29:49
hey, these tokens are worthless.
29:49
I need to liquidate them as fast as possible.
29:52
But, why would you do that publicly?
29:54
Why would Why would you do it privately? to move the market.
29:57
He wanted to move the market.
29:57
He said No, he's not about to move the market.
30:00
He's letting people know why to kill his competitor as Chamath was just saying.
30:03
CZ wanted to kill his competitor who was trying to kill him.
30:07
he was trying to do that.
30:08
We got We got to back up here because I think we've done a lot of like 30,000 foot like lessons and like takeaways from this whole thing, but we haven't really established what it is that SBF did wrong.
30:18
So, I think we need to sort of take a second to un-muddy the waters, okay? Okay.
30:22
And part of that, I think we should start with this idea of a run on the bank because the press who've been writing puff pieces about SBF, I'd say mainly Semafor, which he was a big donor to, Yeah.
30:32
they've been trying to frame it as a run on the bank.
30:34
And then that implies that it's not really his fault, it could happen to anybody.
30:38
Lots of banks have had this problem, okay?
30:39
First of all, they're not a bank.
30:41
Banks actually have the legal right under certain conditions to take customer deposits and loan them out, okay? Yes. They did not.
30:48
Their terms of use did not allow that as Stephanopoulos pointed out.
30:52
SBF's answer to that was well, we had this like margin account program.
30:56
There were other provisions in other terms of use.
30:57
But, most of the customers who lost money, the vast majority, did not opt into that program.
31:03
They never agreed to that.
31:03
So, that's that's point number one.
31:05
Point number two is I think we need to to look at this language of margin account, okay?
31:12
SBF's explanation of how customer money was siphoned off for his own personal use, i. e.
31:18
to Alameda, is that Alameda had a margin account.
31:21
So, I think we can perform a service here by explaining why it wasn't a margin account.
31:25
And you know, Chamath and you guys understand this really well.
31:29
The way that a margin account works is the following, okay?
31:35
Cuz I think some of us have them set up with investment banks.
31:37
You go to an investment bank, say Morgan Stanley, and you over you post collateral.
31:42
You actually over-collateralize.
31:44
So, for example, you might take a $100 million of stock, post it at the investment bank, and then they will let you loan a certain percentage, nowhere near 100%. Maybe 50% loan to value.
31:58
if you have a very, very liquid security, you may get 50% coverage, which means if you posted a $100 million you could get a $50 million loan. Okay.
32:06
So, you had a $100 million in Amazon stock, you're some Amazon VP, you can get $50 million loaned.
32:12
But, and if and if it's a private asset, it's anywhere as high as 30%, 35%, but typically it's about 25%.
32:21
My expectation is an illiquid token like this would have basically gotten 5% or 10% coverage ratio at the best of it.
32:28
And then what happens is you have these maintenance values.
32:30
So, if all of a sudden the value of these entities multiplied by that percentage that you're allowed to loan falls below, you have to post money.
32:38
That's how a margin account works.
32:40
It's just there is no free lunch in that. Yes, exactly.
32:44
There's Let me just say say quite simply, very simply what I it appears this guy did.
32:49
He took customer deposits in US dollars.
32:53
He then converted those dollars into some other asset.
32:56
And he had a mark on that asset.
32:59
Let's call it a dollar, a token.
33:02
And then those dollars were moved to somewhere else.
33:05
No, this is simpler than that.
33:06
transferred in some other token.
33:06
Listen, we need to We need to We need to finish the No, we need to finish the explainer around the margin account, okay?
33:13
Because what SBF did is this.
33:16
He took customer deposits, gave them to himself, i. e. Alameda.
33:21
In In No, he gave a He took customer deposits in US dollars. They were wired in. Correct.
33:28
those dollars out and he put a fake token in. And he called it that. right. He So, so that's right.
33:34
said all he therefore he said the balance sheet is good.
33:35
But, the value of that token it turns out isn't a dollar, it's 10 cents. That's right.
33:40
It was his It was his sort of It was sort of his made-up token that FTT is the name of that token.
33:45
that he tightly controlled the trading of and and artificially popped up the price.
33:50
way, it wasn't just FTT, but yeah.
33:50
Yeah, but but but here's the thing, it wasn't just the fact that his collateral was no good.
33:55
It was also the fact that and this is from the bankruptcy uh filing by the the new the Enron trustee guy.
34:05
He specifically said that Alameda, unlike every other margin account on the platform, had the auto-liquidation provisions turned off.
34:13
So, wait, we have to finish the thought around how a margin works.
34:14
So, like Chamath said, you over-post collateral.
34:19
And if the value of that collateral goes down or the the the the the position your trading account, the value of that goes down, you either have to post more collateral or they will actually liquidate your collateral to pay off the loan.
34:33
So, Morgan Stanley will never lose money on a margin account, never.
34:35
And like the whole point is because they don't make money on it.
34:39
They loan you the money at like, you know, a few percent, like very cheap loan.
34:43
LIBOR plus Yeah, exactly.
34:45
The The that is not a risk account to them.
34:48
And so, again, the in the example, let's use an example of the In the In the case of the VP at Amazon who's got a $100 million in Amazon, they have a $50 million loan.
34:55
If Amazon loses half its value, then that triggers the automatic selling of Amazon shares to get it back down to 50% coverage.
35:05
So, now you have to sell 25 million of Amazon shares if the full 50 million was pulled down to get back down to 25 for 50% leverage.
35:13
Yes, and they don't wait until like Amazon stock is at the exact level where now the collateral equals 100% of the loan.
35:20
They will keep that 50% loan to value.
35:22
And they will liquidate you, you know, and and by the way, you can lose your entire amount, right? So, Yes.
35:28
you know, this is why I'm trading on margin is so risky is that get wiped out completely.
35:33
wiped out very, very quickly with a small move down.
35:34
Because they are the custodians of that Amazon stock.
35:35
They are holding it for you now and they have the right to sell it to cover your margin.
35:42
You're saying that governor, that basic tenant, that basic safety control was turned off by Alameda.
35:47
And it's even more sinister, Alameda controlled like 90 or 95% of these FTT tokens and was owned by Sam Bankman-Fried.
35:59
So, he owned that company.
35:59
Then he claims he had no operating The The Listen, what should have happened is with that collateral is that as the value of their position was going down and/or as the value of the collateral was going down, it should have been liquidated to pay off the margin loan. And that did not happen.
36:16
And the reason it didn't happen is that Alameda got a special exception on the platform to turn off auto-liquidation.
36:21
Therefore, it was never a margin account.
36:23
If even if it was a margin account, okay?
36:25
And and FTX somehow misadministered the margin account, it should never have taken other customers' deposits and used them to pay back that money.
36:35
What should have happened is if FTX was going to lose money on a margin account, that would hit the FTX corporate treasury, okay?
36:45
And when the FTX corporate treasury ran out, the company files for bankruptcy then.
36:49
And then all the other customers, hold on, their account is still there.
36:53
Their money is there in segregated accounts.
36:55
And in bankruptcy, they get their money back.
36:57
The idea that a margin account could ever cause another customer to lose money, that like whatever that is, that's not a margin account.
37:06
There was a great article. There's a great article.
37:07
This one was a journalist that did his job properly. His name is David Z. Morris.
37:13
He wrote an article in CoinDesk that summed up, for anybody that's interested, all of the actual fraud and all of the crimes that were committed in excruciating detail.
37:22
And what's so sad about all these interviews and this press tour is if anybody would just read this article, you can construct the right questions to ask this guy just based on this one article.
37:31
But, the the point I wanted to make is that one of the most interesting insights was these guys had lost an enormous amount of money already in calendar year '21.
37:45
And so, this is what's so crazy, Jason, about, you know, you using language like rug pulling and like, you know, nobody is actually trying to like be clear.
37:55
Like you guys are giving this guy a hall pass.
37:57
I'm not giving him a hall pass.
37:59
Any industrious reporter could have found an employee who said, "Wait a minute, we just blew a $3 billion hole in our balance sheet in calendar year '21 '20 and now we're sitting here at the end of '22."
38:09
Chamath, there's a great article that says Hold on a second. Hold on. I need to respond.
38:15
I am not giving him a pass.
38:15
And for you to blame journalists who are reflecting the crime and not putting any light on VCs and the capital allocators who made this investment and who did no diligence and did not put governance in it is the height of arrogance, Chamath.
38:29
This is not the press's fault. This is the VCs' fault.
38:33
This is the capital allocators' fault.
38:35
You're blaming the people who are telling the story after the crime is done.
38:38
No, they're they're covering No, YOU CAN'T HANDLE THE TRUTH.
38:43
JASON, they're covering the story.
38:43
No, they're not telling the truth. Can I get in here? Can I get in here?
38:46
All right, listen, Jason.
38:47
I will defend you against Chamath saying that somehow you're you know, that you're letting SBF off the hook.
38:54
I know you don't want to let SBF off the hook.
38:55
However, you are letting the press off the hook.
38:57
And the reason why Hold on a second.
38:59
The reason why you're using this inaccurate language like rug pulling and run on the bank when there was no run and there was no bank is because you've been infected by this language that the media has inserted into the discourse.
39:11
The media Listen, and hold on a second.
39:14
Investors may have got it wrong last year.
39:18
Investors may have got it wrong when they did that last round.
39:19
But, I think investors now understand what's happening.
39:22
But, the media is still covering for SBF by mis-explaining what happened.
39:27
Okay, give me a percentage, Sachs, of who's to blame here.
39:30
VCs who invested and didn't set up any governance, regulators who did not set rules around crypto, and then three, the media.
39:39
What percentage out of a 100% is the investors, the regulators, and the press? Go. Three numbers.
39:47
say that before the fraud got exposed, 1/3, 1/3, 1/3.
39:49
1/3 each before the fraud got exposed, but they were all jointly and severally liable.
39:55
But, after the fraud's been exposed, no investor is still defending SBF.
40:00
But, I I I think that the investors who were swindled by him, they feel bad about it. So, 30 30 30 is absurd.
40:07
The press had no way to know the fraud was going on, just like the VCs who put the money in. about?
40:13
What are you talking about?
40:14
Jason, are you being stupid?
40:14
Like, wasn't wasn't Wasn't the journalist that exposed the fraud at Theranos?
40:19
He's the guy that went and did all the work.
40:21
John Carreyrou should be celebrated. Hold on a second.
40:25
John Carreyrou went and found this thing when no when everybody else was like, "This is perfect.
40:29
It meets all of our priors." Let me finish, please.
40:33
It meets all of our priors. This is great.
40:34
And then John Carreyrou, hold on.
40:36
And so, John Carreyrou was like, "This doesn't pass the smell test to me. Let me go do some work."
40:39
And he pulled one little string, and over the course of 18 months, he exposed the whole bloody thing. So, hold on a second.
40:45
So, what is incredible to me is that it was possible to expose this thing before. Nobody did. I agree with David.
40:51
It's about equal responsibility before, but afterwards, the bulk of the responsibilities now sits with regulators to clean it up and journalists to tell the truth. Okay.
41:02
And now, may I respond to that since you called me stupid? You are delusional.
41:05
Number one, every one of those investors in Theranos could have taken a [ __ ] blood test at two different places, like Jean-Louis Gassée did, and write a blog post, and prove that Theranos didn't work.
41:18
And they withheld disbelief.
41:21
Investors putting in a hundred million dollars, including Rupert Murdoch, didn't even take a [ __ ] blood test or tell one of their diligence teams to do it.
41:29
The same thing happened here with the investors in FTX. They did zero diligence.
41:34
They set up zero governance.
41:36
This was a failure of the investors and the governance for 99% of the problem.
41:42
And then, regulators should have caught it, and the regulators in fact did catch Theranos.
41:47
So, you're completely wrong, Chamath, again.
41:50
The journalists come in after the fraud is happening.
41:53
The investors and governance is responsible for stopping these things.
41:57
FTX was a failure of governance and investors, and so was Theranos. The end. You're completely wrong.
42:04
The question is post post exposure.
42:07
Why are you guys obsessed with post?
42:07
How about avoiding these things?
42:09
You guys are blaming journalists is ongoing.
42:12
Because the story is ongoing.
42:14
for something that is capital allocators' responsibility.
42:16
It is our responsibility to do diligence.
42:17
It is our responsibility to create a board of directors that checks on Elizabeth Holmes I didn't I didn't disagree.
42:24
I Nobody's disagreeing with that.
42:26
You just called me stupid.
42:28
You just called me stupid for pointing out something that you refused to accept.
42:31
I'm the I'M THE BIGGEST BOOSTER OF CAPITAL ALLOCATORS. I'M NOT STUPID.
42:39
YOU CAN DEBATE WHATEVER YOU WANT.
42:41
YOU GUYS ARE giving a pass to the investors You're doing your Fredo impression again. I'm not doing Fredo. I'm not doing Fredo.
42:48
You guys are being absurd.
42:48
This is why people say that we're delusional on this podcast. Don't call him dumb.
42:55
The reason people say we're delusional is that we won't take acceptance of this issue. Totally. He loses it.
43:03
No, I'm not going berserk.
43:03
You guys have a blind spot.
43:05
Capital allocators are responsible. Don't call me dumb.
43:09
I'm not going to defend I'm not going to defend any single one of those investors.
43:19
But, the reality is it just went up.
43:20
But, the reality is I think that if you think that you can If you It's your decision to defend the mainstream media.
43:26
I think that that's fine. I'm not defending them. No, you are.
43:28
You said they have no responsibility. blaming the VCs. It's different.
43:30
The culpability is with the investor class that has not had proper governance and diligence.
43:38
How many articles have been written excoriating them? Uh yeah, some. A lot. They look like fools.
43:46
Yeah, I mean, the Show me Show me Show me the Washington Post No, show me the Washington Post, New York Times that's like digging in to that malfeasance or that lack of oversight and holding them accountable in a way that you feel exposes this problem to create change.
43:59
Well, if we look at Theranos, those people who invested, including Draper, and um Just show me the examples Rupert Murdoch, they they really went after them, for sure. Yeah. What about here?
44:15
Uh Wall Street Journal, one day ago.
44:17
Sequoia Capital apologizes to its fund investors for FTX loss.
44:19
Venture capital firm tells fund investors that it will improve due diligence on future investments after a hundred million dollar loss. Yeah.
44:25
Let me Let me Let me read you I'm going to I'm going to read you I'm going to read you a sentence from the New York Times coverage of SBF.
44:34
And Sam Bankman-Fried Hold on a second.
44:36
Sam Bankman-Fried is neither a visionary nor a criminal mastermind.
44:38
He is a human who made the same poor choice that generations of money managers have made before him.
44:44
Are you effing kidding me?
44:47
That was the New York Times coverage.
44:48
They also said I just called him Andrew Sorkin, a suit.
44:53
And then, Semaphore, who was on the take, who received millions of dollars of money, said it was a run on the bank. Okay.
45:00
Now, where is their apology? Hold on a second. Where is their apology? Sequoia has apologized. Where is their apology? Oh, it has to come.
45:07
I'm not defending the press. I'm saying No, you are. You are. You are. NO, I AM NOT.
45:10
I AM LITERALLY telling you that the New York Times has been asleep at the wheel.
45:16
The press is always demanding an apology from everybody else.
45:20
The press is so incompetent on this.
45:20
The Twitter Spaces yesterday did a better job of trying to ask questions and getting to the truth than a single journalist has done, or the collective body of all of journalists. Absolutely.
45:31
Literally, randos on Twitter Spaces did a better job than Sorkin.
45:34
Let me tell you why the why no one trusts the press, Jason.
45:37
First of all, they have an agenda.
45:39
And secondly of all, when they make a mistake, they never admit it.
45:42
When's the last time they did an apology or retraction?
45:44
When's the last time they did what Sequoia did? I don't know.
45:48
And they need to apologize in a timely manner.
45:49
I am in agreement with you on that.
45:51
But, I think we have to first say, and this is where you guys have a blind spot, is what is the responsibility of capital allocators and governance and regulators.
46:01
I think it's one, two, three.
46:04
Our industry is responsible for setting up proper governance.
46:06
The regulators are responsible for making sure that scientific claims are backed up.
46:10
And then And then, press is a distant third.
46:14
You know who I think is responsible? SBF. One, two, and three.
46:16
And then, we can talk about four, five, and six. Okay. Four, five, and six.
46:22
Capital allocators, regulators, the press a distant sixth. I agree. Let's go on to China.
46:26
God, it's so spicy today in here. God, it's so hot.
46:31
I mean, I do think I do think when we attack the mainstream media, Jason feels a little twin tinge of like uh insecurity and illegitimacy cuz he were a journalist.
46:39
My My personal perception is I think that No.
46:42
I I have I think that you have a an incredibly romantic view of the craft as you practiced it back then, which I think was full of integrity. Yes, I know. That's true.
46:54
I thought I think that you you don't adequately realize how massively the industry has changed in the last 20 years since you became an investor. it more than you do.
47:01
I fully realize that the media has absolutely become biased, and they have lost Uh in some cases, yes.
47:08
I mean, if you're taking from SBF and then giving him They are corrupt if they're taking money from SBF and then giving him kid glove coverage. Absolutely.
47:17
That is the definition of corruption in my mind.
47:20
What is What is it called when you don't take money necessarily, like the New York Times, and still treat them with kid gloves? What is that? It is extreme bias.
47:25
And the New York Times became incredibly biased.
47:28
think the bias Why do you think that bias exists?
47:31
They were always left-leaning, but I I can tell you why.
47:34
They when Trump came in, a generation of new journalists became activist journalists.
47:40
They didn't want to tell stories and take it straight down the middle, and let the facts tell the story, and let the audience make their own decision.
47:45
They felt an existential risk when Trump came into office. They got syndrome.
47:51
They picked a side like MSNBC and Fox did, and the business model became, for the New York Times, pick a side and get the subscribers.
47:58
It was a deliberate cynical choice on the New York Times' part to go full MSNBC or full Fox, the two extremes in mainstream media, in order to get the subs.
48:09
And they literally rallied the troops there to do anti-tech, anti-Trump uh coverage, and they became activists.
48:16
And when journalists become activists, they are no longer journalists.
48:20
They're activists or commentators. And that's the problem.
48:24
It's being presented as journalism when in fact, it's activism.
48:26
So, you're you're you're right.
48:28
So, So, shout out to Matt Taibbi, who just did a Monk Debate on on this Yeah, on this very topic.
48:33
And he has a great great Substack, basically saying what you're saying, Jason.
48:38
And the the best quote is, "The story is no longer the boss.
48:42
Instead, we sell narrative."
48:44
He's a lifelong journalist whose father was a lifelong journalist.
48:48
And he understands the way the business has changed.
48:49
And it's like what you're saying.
48:50
And And this is why independent media, whether it's Substack, whether it's call-in shows, whether it's All-In podcast or other podcasts, Joe Rogan, Sam Harris, whoever it is, independent voices are now what consumers are seeking out because they can sense the bias.
49:06
They know Rachel Maddow and Tucker have a an axe to grind, and they're left and right.
49:10
They didn't expect the New York Times, Washington Post, and Wall Street Journal to, you know, they they knew they were leaning.
49:17
They didn't expect them to pick a side.
49:19
Do you think we should cancel If folks Do you think folks are better off keeping their New York Times subscription or replacing that New York Times subscription with a basket of Substack?
49:31
Uh yeah, you answered your own question. It's the latter.
49:32
I I think you're on your own as a consumer now.
49:34
You're going to have to And I think this podcast and the nuance we have to shout out to Friedberg for nuance.
49:40
What we've done on this podcast is funny sacks?
49:44
to explain to people Friedberg's not the only one with nuance J Cal.
49:49
Nobody would describe David Sacks with the word nuance.
49:51
nuance department on this podcast? What am I?
49:53
100% he is and but you wouldn't know that cuz you leave when science corner starts. department, okay? Sometimes nuance. All right.
49:59
I'm the truth department.
49:59
You can take the both sides department.
50:03
consumers need to become extremely literate and they have to do their own search for truth in today's age.
50:07
They don't they shouldn't trust New York Times, they shouldn't trust us, they should trust themselves.
50:11
They shouldn't trust necessarily the CDC or, you know, the World Health Organization.
50:16
They should trust themselves and come up with their own process for figuring out the truth in the middle of this mess.
50:24
By the way, this is a good reflection on what's happened with the rest of media with respect to the creator class.
50:31
Where right, it used to be the movie studios and, you know, a handful of kind of aggregated creators that made all of the content of the record labels.
50:37
And now, you know, independent artists, independent producers, independent creators and now independent journalists are going to become the bulk of volume that's going to be consumed.
50:48
It's just a different consumption model.
50:49
But we've already seen it happen with music, we saw it happen with movies and we've seen this disruption happen across all of these other media classes.
50:55
Journalism and what we call the press is very likely going to be kind of that next layer of disruption.
51:04
I would trust having a conversation with you about science topics over reading a science article 100%.
51:09
you know, in the New York Times or Wall Street Journal if I'm being honest.
51:12
I would much prefer to talk to you about it and if it was markets, I'd rather talk to Chamath and if it was SaaS I would talk to Sacks or operating a company.
51:19
Speaking of operating a company or politics. Uh we'll we'll see. We'll see.
51:29
Uh I think it's about having unique insight insight, right?
51:32
Like that wasn't the case.
51:34
And what's interesting is that the people who are the professionals that have the knowledge and the touch points are also becoming journalists in the sense that they're also becoming speakers of their truth. Right?
51:44
And I think Twitter's a good enabling platform for this.
51:46
We see it on YouTube where like scientists are putting out their own videos or market actors like people that are traders in the market go out and they put out their own videos and they put out their own podcast and I think we're probably a good reflection of that. Yeah.
52:01
Uh in the sense that like we are the actors in the market and we're not just the independent observer that has kind of a surface level view.
52:07
We have the depth to be able to talk about the things that we choose to talk about.
52:13
And I think that's where consumers find value and will continue to find value in terms of who the journalist or speaker is that they're going to start to trust for their information.
52:21
I had a chance Yeah, I saw I saw that interview you did with newcomer. Oh, yeah. Yeah, yeah, I saw that.
52:27
Yeah, I saw that that coverage.
52:29
I mean, he he speaks a lot about the this phenomenon of going direct.
52:33
Of course, he's against it.
52:34
Now, he interprets going direct as an attempt by um newsmakers to avoid answering tough questions or take tough questions.
52:43
I think that's ridiculous because for example, I go on CNBC all the time, I go on Emily Chang and Bloomberg all the time.
52:52
I submit to like really tough questions.
52:54
I actually like those sort of sparring sessions. Yeah.
52:55
I did hard talk this week. Have you ever done that?
53:01
So so that's not what's going on here.
53:01
I think what's going on is we have expertise, we want to communicate them and we do feel like the media has become a very unreliable narrator.
53:10
There is too much bias and sloppiness.
53:12
Not all of it is agenda, some of it's just pure sloppiness and there's no reason why we shouldn't go direct and people want to hear from us.
53:21
Their audience wants to hear from us.
53:23
Same for look look at Draymond.
53:24
Look at Draymond and the success he's had with his pod.
53:25
I mean, no basketball player has ever gone direct and created content like Draymond's created and it's totally changed the game.
53:33
And he was so clear, he's like we are I am the media now.
53:34
You need JJ Redick, you know, old man in the three, amazing amazing podcast.
53:39
I was going to tell you guys a story.
53:42
So I was in the Middle East last week or this week, sorry.
53:46
And I had this crazy experience where I was trying to understand what was going on in China.
53:52
And so I started on CNN and the whole thing was the propaganda machine around a democratic revolt, you know, pushing for democracy and trying to depose G.
54:04
Then I moved to Al Arabiya.
54:04
So one channel up, I went from channel 10 to channel 11.
54:10
And instead what they were actually doing was interviewing people on the ground and what they were talking about was literally how these PCR tests have become far too burdensome and they just wanted it to end and more reasonable restrictions to get in and out of quarantine.
54:25
Then I went from there to BBC.
54:25
And in BBC, they had a China scholar who was talking about how for decades actually the Communist Party supports local level protests and demonstrations because they've realized that it is a part of their political system to make sure that people feel like they have a say.
54:45
And I was like taking a step back and I'm like, if you listen to the US narrative and even Jason like in our group chat, people fomenting for like revolution and this is Tiananmen 2.
54:52
0 and I'm like, well, I'm reading two other channels that tell us a completely different set of things and I just thought, man, people just really fit the data Such a good point, yeah. their bias.
55:05
Yeah, I mean, we are projecting.
55:07
We want to see a revolution in China.
55:09
The people in China want to have their lives back.
55:11
Well, I would love to see more democracy in the world. Yes, guilty as charged.
55:17
I would like to see people be more free in the world. Dictator.
55:20
I think most people just want to improve their condition.
55:22
And I don't think people are as tied up on the philosophy of the government as they are about improving their condition and as long as their condition is improving, they're willing to put up with any form of government. And history shows that.
55:33
By the way, the conditions in China have improved better than everybody's in the last than any anyone 500 million people out of abject poverty and that's the great success of engagement.
55:46
Isolationism would not have created that amazing outcome.
55:50
500 million people going out of abject to Apple building factories.
55:54
Apple building factories is what I'm referring to. Oh, okay, fine.
55:58
If if they if they want to build something over there, I guess that's better than us throwing open our markets and giving China MFN status to destroy American manufacturing and build up their economy so they can become a pure competitor to the United States.
56:13
Yeah, I mean, this is the balance of engagement, isn't it?
56:14
If you engage too much, you give everything up.
56:16
You know, Sacks, which of the current Republican agenda do you disagree with most strongly? Just as an aside.
56:26
Well, most Republicans are in favor of our Ukraine policy, that sort of unlimited appropriation of weapons and aid to them.
56:34
Don't you disagree with immigration policy of Republicans and Democrats?
56:37
Well, I I have a more nuanced position on immigration which is I think we need to have a border and it can't be just like an open border which is the de facto policy we have now.
56:46
But at the same time, I do think that we should have H1B visas and we want to like Chamath said, we want to be an all-star team for the world.
56:52
We want to have the best people want to come here. So there's a balance. It's a balance.
56:56
And then, you know, look, I think that I was happy to see the marriage equality bill finally pass the Senate.
57:02
Yes, they did get about a dozen Republicans voted for it. That's great. Yeah.
57:07
But that's not the majority, unfortunately.
57:11
You know, look, on on what I would categorize as the old social issues uh you know, like gay marriage, like cannabis legalization, I was on the liberal side.
57:21
Yeah, I don't think, you know, banning abortion entirely, a total abolition is going to work for this country.
57:27
I think Republicans will lose elections if they insist on that and I think they're getting that message.
57:32
So um so yeah, I mean, look, I I think that um I've always considered myself to be pretty centrist.
57:39
And so you're not a globalist.
57:41
You don't believe in open global markets with the US.
57:45
In general, I understand the benefits of free trade and I don't think we should be isolationist with respect to trade.
57:49
I don't think that we can be a successful country if we are we isolate our economy. So I do want to trade.
57:56
However, with China in particular, I think we made a mistake in throwing open our markets to their products, giving them MFN status while Most favored nation.
58:09
Yes, enriching them to the point where they became a pure competitor to the United States.
58:13
Now, look, I understand why we made that mistake 20 years ago because everybody thought the theory was that if we help China become rich, that China would inevitably become more democratic and they'd be filled with gratitude towards the United States and they'd actually become more more hospitable towards us, more westernized.
58:31
And I think that theory has just proven to be wrong. I mean, they have not.
58:35
Or it's going very slowly. One or the other.
58:37
Let's own our clips here.
58:39
Here is Chamath's prediction from episode 61 and we'll see you on the other side of this quick clip.
58:46
My worldwide biggest political winner for 20 2022 is Xi Jinping.
58:51
I think this guy is uh he's firing on all cylinders and he is basically ascendant.
59:00
So 2022 marks the first year where he's essentially really ruler for life.
59:04
And so I don't think we really know what he's capable of and what he's going to do and so that's just going to play out.
59:10
You think he's the biggest political winner, really?
59:12
Oh my god, I think I think it's going to be a he's going to run roughshod not just domestically, but also internationally because you have to remember he controls so much of the critical supply chain that the western world needs to be successful.
59:25
You're I think you're completely wrong.
59:28
I think he's losing his power, he's scared.
59:29
That's why he took out all these CEOs.
59:31
He's consolidating power because he fears that they're going to win too big and then displace him and he has massive real estate problems over there that could blow up at any moment in time.
59:43
He could face a civil war there.
59:44
I think he's totally isolated himself. Civil war?
59:46
They don't even have guns.
59:48
Every major country is removing their factories and removing their dependencies. there.
59:52
What are you talking about?
59:52
What are they going to What are they going to riot with?
59:56
Uh did you not see Tiananmen Square?
59:56
Did you not see the riots in Hong Kong?
59:58
Are you not paying attention, Chamath?
1:00:01
There's been many riots in China.
1:00:01
They just happened to kill all the people. souls were crushed.
1:00:04
I'm not saying they were crushed, but he still will have massive amounts of uh I believe protests and yeah, he'll have to kill people.
1:00:13
I think the the the bigger risk is is that China gets better for Xi Jinping but worse for everybody else in China.
1:00:19
It's already worse for all the billionaires uh over there.
1:00:20
It's worse for the tech industry.
1:00:23
You've now got Evergrande, that whole uh you know, gigantic debt implosion.
1:00:27
I think there could be contagion from China next year.
1:00:32
I don't think Xi's going to lose his grip in any way, but I'm not sure China's going to have a good year next year. Wow, nailed it.
1:00:39
I think all three of us kind of got this right. talking about?
1:00:43
You got none of it right.
1:00:44
Well, I I said there were going to be there were going to be riots and they were going to they're going to have uh a recession.
1:00:49
I mean and Jason, let's be honest.
1:00:50
Look, that they would be squashed.
1:00:51
That's exactly what happened. Both things happened.
1:00:53
I actually think I have a pretty decent ability to steel man pretty concretely the details.
1:00:56
I think that at best when it comes to things like democracy and your belief in US exceptionalism in a specific political worldview, at best you straw man.
1:01:07
And I think that you get very biased without seeing the forest from the trees.
1:01:11
The reason I said that is not because I'm like, you know, some huge Xi supporter.
1:01:15
I'm just trying to steel man what happens when one individual person Mhm.
1:01:20
gets anointed leader for life of 1.
1:01:23
3 billion people that then controls 20% of the world's GDP.
1:01:26
There is no other single human being as powerful as him as of this month.
1:01:36
Can I just say that this show this show is going to become insufferable if every time you sort of said something in the past that was sort of correct, you're going to have to replay it?
1:01:45
I don't know how to replay it.
1:01:47
I was actually I was actually playing that one for you, Sacks.
1:01:49
I think you were the one who nailed it.
1:01:50
I was giving That was a softball to you, actually, not me. week, J Cal. All right.
1:01:54
This show is going to slow down if we play every clip that I got right.
1:01:58
Wait, you guys are asking to pull clips all the time.
1:02:00
I just pulled one clip about China. Well, you nailed it. No, no, no, look.
1:02:03
I I think like and also look what he did. the king of pull a clip.
1:02:05
The the Evergrande thing, look what he did this week.
1:02:08
They said, "Okay, you know what?
1:02:10
The real estate industry can now issue secondary stock sales, raise equity and equitize themselves."
1:02:14
So, they're going to find a soft landing for the equity part of for the real estate industry in China.
1:02:21
And now they're reopening. So, I don't know.
1:02:23
I mean like I'm not sure what we're supposed to comment.
1:02:25
What I'll What I will stand by is what I said, which is I don't think we have a very clear view about what's going on, what the substance of these protests are, and what people actually want if you're only consuming US media.
1:02:39
And so, if you find a way to get a diet from a bunch of different sources all around the world, you may be get a better sense.
1:02:46
I had an accidental window into that by being in a completely different part of the world this past week.
1:02:50
Freeberg, any final thoughts on China and what's going on there before we move on to chat GPT, your favorite story?
1:02:57
I think one of the things we often miss is that China the CCP does have their hand on the throttle.
1:03:08
Like they throttle up and down.
1:03:11
We always think that it's a linear line and that it's super dogmatic and fixed, but there's certainly responsiveness in the release of the lockdowns in Guangzhou and Beijing this week seems to have been a pretty good indication that when things do get when the when things when the tides do do change, leadership there seems to respond.
1:03:29
Not always, but enough to kind of keep things going. So, Should they reopen?
1:03:33
I think it's 60% of the population or so is vaccinated with obviously vaccines that maybe aren't as um don't have the same efficacy as the ones uh here in the United States.
1:03:44
Do you think they should open up and just let it rip uh or do you think they should keep still try to maintain this zero COVID policy?
1:03:50
Cuz that is the debate right now.
1:03:54
From what you're saying, what's what's what's what's the objective?
1:03:55
Cuz that's what I'm asking That's what I'm asking you, yeah.
1:03:57
Well, from the objective of economic growth, they need to open and they need to keep their economy working and they need to keep their labor force engaged or else they're going to continue to suffer.
1:04:06
So, if economic growth is the objective, they they need to open up, right?
1:04:11
If the long-term health cost of the nation balanced again against that is the calculus that they're kind of weighing, there's probably some more nuance to that.
1:04:23
And certainly um my understanding is there may be a precedent setting which is, "Hey, we've said that it's a zero COVID policy.
1:04:33
Therefore, we have to hold strict to it, hold to the line."
1:04:37
Um else it looks like we're weak.
1:04:37
And so, there's also this, you know, maintaining the authority of the CCP objective.
1:04:43
Uh so, there's a lot of maybe competing objectives right now.
1:04:47
Uh certainly don't have a sense of how they're they're weighing them all.
1:04:49
But But I think that once once all those videos came out this week, Yeah.
1:04:52
you guys saw them, but people were screaming.
1:04:55
There was an apartment on fire where the doors were locked with steel beams on the on the base of the building.
1:05:00
At least that's what the the video said.
1:05:01
I don't know how much truth there is to that, but that's what was said.
1:05:05
And clearly people are extremely distraught and unhappy with the conditions of the lockdown.
1:05:12
At some point enough peace people with enough loud voices, you know, something's going to change.
1:05:17
I mean, let's just remember like the bargain that's struck in that country and with all countries is that, you know, the citizenry to some extent are willing to tolerate their government so long as their conditions continue to improve. And there's a bargain.
1:05:34
There's some bargain that's struck.
1:05:36
And as soon as that bargain starts to go south for for for the citizenry, then that that that governing entity is at risk.
1:05:45
And I think that that's what we sort of started to see this week was the conditions are getting far far worse and far less lovable for so many people in that country that the government had to shift.
1:05:57
Do you think, Chamath, this uh COVID strategy and then we'll move on was basically Xi Jinping wanting to get to that Congress, his coronation, and now that that's over, maybe he can change gears and then Like I said, I my belief is that I have a very poor access to enough data to have a to steel man um what is actually going on there.
1:06:18
But one explanation could actually be that in the absence of enough hospital infrastructure and ventilators and a bunch of these other things, they had to take a pretty severe approach to this um disease.
1:06:36
I don't know what they know or didn't know.
1:06:38
Maybe they understood, you know, the virulence of it.
1:06:41
Maybe that they have uh slightly different aging characteristics of their population.
1:06:45
Maybe they genetically responded to the SARS-CoV-2 virus.
1:06:47
I I don't know any of these things enough to tell you, Jason. Yeah.
1:06:53
But the reality is what Freeberg says is right, which is that you cannot grow an economy if people are inside locked in their apartments.
1:06:59
And it looks like they have decided that that's coming to an end and they're going to, you know, deconstruct all of these things.
1:07:08
So, you know, the the Chinese growth engine is coming back.
1:07:13
And I think that that's going to be an important factor economically that we're going to have to figure out.
1:07:20
Because it's going to have a huge implication to American growth and American inflation.
1:07:26
Sacks, if uh in fact the Chi- if the lab leak theory is correct, China might have some insights into this uh disease that uh maybe the West didn't.
1:07:36
Maybe that plays into their policy a bit. Any final thoughts here?
1:07:40
J Cal, I'm just surprised to hear that you have a problem with their lockdown uh policy over there.
1:07:46
Because aren't they just implementing Democratic Party orthodoxy?
1:07:50
I mean, isn't this the policy that Tony Fauci and Barbara Ferrer, you know, that all the health officials were I was not in favor of lockdowns.
1:07:56
What are you talking about?
1:07:57
You're the one who had all your mask and had ventilators day one.
1:07:59
Isn't this uh the the lockdown policy am I getting us astray?
1:08:05
Because you're the same Democratic Party.
1:08:06
I'm just asking a question as the moderator.
1:08:08
Isn't this basically I getting us astray?
1:08:10
Isn't this what Gavin Newsom and Cal. Answer the question. Yes. Hold on.
1:08:14
Let me Let me finish the question.
1:08:18
I was in favor of if people wanted to stay home, stay home.
1:08:20
And then if people wanted to go out and take the risk, take the risk.
1:08:23
I was always in personal choice.
1:08:25
what Gretchen Whitmer in Michigan and Gavin Newsom in California subscribed to, the idea that the way to fight COVID was through lockdowns?
1:08:33
Now, yes, Newsom had 10 pages of exceptions for his political donors.
1:08:37
And he didn't use the police to lock people in apartment buildings.
1:08:41
He may have wanted to, but they didn't actually do that.
1:08:43
But can you really tell me that this lockdown policy has been disavowed by people like Fauci or by the the health authorities like the Barbara Ferrers of this world?
1:08:55
They still subscribe to this view. Do they? Really?
1:08:57
Is anybody doing lockdowns right now? Show me.
1:08:59
not able to do it because no one agrees anymore.
1:09:02
But tell me tell me where anybody Tell me where any of the health experts who said that lockdowns were the correct response have repented and disavowed that view. Yeah, I don't know.
1:09:12
I haven't been I haven't been tracking their uh mea culpas.
1:09:15
But you you yourself were in favor of lockdowns for a period of time. Yes, you were. No. You absolutely were.
1:09:21
I'll pull the tape for the next episode. mandate. No, it's not true.
1:09:23
I I in favor of a mass mandate.
1:09:25
And I said the mass mandate was the alternative to lockdowns.
1:09:29
I was saying that by May of 2020. All right, listen.
1:09:31
Let's move on to the next thing.
1:09:34
I I don't represent I'm an independent.
1:09:35
I don't represent the Democratic Party. I don't represent Fauci.
1:09:40
No, you represent mainstream media.
1:09:40
I do not represent mainstream media.
1:09:42
I you just said I was an old school journalist who's mortified with where it is today.
1:09:47
I represent the old school mainstream media.
1:09:49
I thought your explanation was fabulous. Oh, thank you. Thank you. Obviously, J.
1:09:52
Cal, I'm giving you a hard time, too.
1:09:53
I I know that you were not a big lockdown proponent, but you understand the point I was making which is your point.
1:09:59
But you're just Let's stop Let's stop making me the Democratic spokesperson. I am not.
1:10:03
I'm feeling a lot better after my beer.
1:10:05
Let's talk about Open AI. Come on. Let's go. All right, so listen.
1:10:09
Open Let's talk about Open AI. All right.
1:10:11
Open AI is a a company that builds artificial intelligence software and platforms.
1:10:20
They have one platform called GPT.
1:10:22
It is on its third version.
1:10:22
As part of GPT-3, they created is a chat interface where you can ask questions to AI.
1:10:32
The results are nothing short of stunning when they hit.
1:10:38
Some of them are a little bit mixed, but Freeberg has spent the last 48 hours drinking white Russians with Oatly milk and playing with ChatGPT including his question to the ChatGPT which was write a script of Chamath, Sachs, and J.
1:10:55
Cal talking about the future of AI in the style of a Quentin Tarantino movie and I have to say it was pretty great, the result. Go ahead, Freeberg.
1:11:05
Tell us what you discovered.
1:11:07
Wait, I think you guys should read this real quick.
1:11:08
We can cut it out if it's stupid, but Okay, Chamath, You want to actually read the script?
1:11:12
Yeah, read read the script.
1:11:12
I don't think you got the Quentin Tarantino part right.
1:11:14
I mean, I think it was like impressive, but it there's no F bombs.
1:11:17
There's no like Tarantino-esque dialogue.
1:11:19
And by the way, they they excluded all of that from Open AI.
1:11:20
They took They took out all cursing and violent cursing and violence from Open AI.
1:11:25
That explains it because if this was a Tarantino script, there would be some choice language in here. But it's interesting.
1:11:29
It's like the PG Tarantino script cuz with no cursing, no violence, no sex.
1:11:33
There's no pop culture references.
1:11:35
Yeah, you do the narrative. Go ahead.
1:11:43
The scene opens on Chamath Palihapitiya, David Sacks, and Jason Calacanis sitting at a dimly lit table in a smoky bar.
1:11:47
A bottle of whiskey sits in the center of the table half empty.
1:11:51
Chamath leaning forward, his eyes intense.
1:11:53
So, what do you guys think about this ChatGPT thing?
1:11:59
David Sacks takes a sip of his drink.
1:12:01
It's impressive, no doubt about it.
1:12:01
A large language model trained by Open AI that can generate human-like text in a variety of styles and formats is a game-changer. Jason nods. Definitely.
1:12:09
It's one of the most advanced AI systems we've seen to date.
1:12:14
It's going to have a huge impact on the way we use and interact with AI. Chamath frowns.
1:12:21
That's what I'm worried about.
1:12:21
With a technology like this, the possibilities are endless, but so are the potential dangers.
1:12:27
Sacks leans back in his chair. Oh, come on, Chamath.
1:12:30
You're not still stuck on that whole AI is going to kill us all thing, are you?
1:12:35
Chamath leans forward again, his voice low and serious.
1:12:36
Look, I'm not saying that ChatGPT is going to turn into Skynet overnight, but we need to be careful about the way we develop and deploy AI.
1:12:49
We need to make sure that we're using it for good and not just for our own gain. J. Cal nods. I agree.
1:12:54
We need to make sure that we're responsible in the way we use AI and that includes making sure it's accessible and affordable to everyone, not just a select few.
1:13:05
Chamath with a dirty smirk.
1:13:08
And let's not forget about the importance of regulation.
1:13:11
We need to make sure that we have the right laws in place Okay. Okay.
1:13:14
This is This is about misuse of AI. Okay.
1:13:18
The director's calling cut. Cut. That's it. Scene. Exit scene.
1:13:20
We're We're not going to finish it. IT'S NOT THAT FAR OFF.
1:13:24
I MEAN, it's 60% of the way there.
1:13:27
I actually there there There was an even better line. in a Biden for Sachs.
1:13:29
If you blame Biden, it would have been perfect.
1:13:31
Let me tell you guys something stunning about this this platform. So, this is GPT-3.
1:13:35
5 which is an interim model to the what people are saying is the long-awaited GPT-4.
1:13:43
0 model which I think they announced in 2020 and has been in development for some time.
1:13:49
So, the model this GPT-3.
1:13:49
5 model was trained in three steps.
1:13:53
They do a great job explaining it on the Open AI blog site where they collect some some data and then there's a supervised model meaning that there are humans that are involved in tagging and then the model kind of, you know, learns from from that system.
1:14:09
Then you ask the model questions, you get output.
1:14:10
And then humans rank the output and so the model learns through that ranking system.
1:14:14
And then there's kind of this third optimization thing and then it's fine-tuned.
1:14:18
So, the the model itself has several steps of kind of human involvement and, you know, kind of it sources its own data and and and builds it.
1:14:27
You know what's incredible about this model?
1:14:29
The total size of the software package that runs the model is about 100 GB. Isn't that amazing?
1:14:33
Like you could fit this model on probably what?
1:14:38
20% of the storage space on your iPhone and you could run this thing and you could probably just talk to it for the rest of your life.
1:14:44
And it it's really kind of an incredible milestone, but I think what was so stunning to me about this I know you guys are probably expecting something to be said like this, but you could see so many human knowledge worker roles and and and functions being replaced by this extraordinary interface.
1:15:05
So, kids can do homework. That's easy.
1:15:07
Software engineers can get their code optimized and can get their code written for them.
1:15:13
There's great examples of how software code has been written by this interface.
1:15:17
You could see real estate, insurance salespeople being replaced by some sort of software-like interface like this.
1:15:25
Copywriters, you know, make me 100 versions of a commercial or an ad Customer support completely replaced. Right?
1:15:34
If you guys remember, there were these automated customer support companies that started, you know, two decades ago and there was this great flurry.
1:15:41
All BPO businesses were all about lower cost human labor.
1:15:44
Now the cost of human labor goes to zero.
1:15:46
My prediction which is So, everyone's got the obvious prediction which is there's going to be 100,000 startups that are going to emerge.
1:15:52
I mean, this is kind of like this moment where the internet came along and everyone's like this changes everything.
1:15:58
I do think everyone thinks and feels that.
1:15:59
So, the obvious next step is a bubble will form.
1:16:01
And so, Can I ask a technical question though, Freeberg?
1:16:04
And then Chamath is probably thinking the same thing.
1:16:08
Let me just finish my market prediction and then we'll do the But I think because everyone's so hyped about this and and we all know this, all of the VC attention, all the investor attention is shifting to this capability and how do you apply this sort of capability across all of these different industries and all these different applications.
1:16:23
And as a result, my guess is the next hype cycle, the next bubble cycle in Silicon Valley will absolutely be this generative AI business. Okay.
1:16:32
But this is a little technical, but how would it know the difference between like y o u r and u r when it is processing natural language if you were to do like your anus or Uranus?
1:16:46
How would that Freeberg, how would it know the difference between Uranus and your space anus?
1:16:51
It'll It'll learn that, you know. It's a joke. Yeah.
1:16:55
It was a joke about your anus. It didn't land.
1:16:56
I think I think the AI The AI probably would have made a better joke than that.
1:17:00
It would have made a better joke, for sure.
1:17:01
Somebody did it in our group chat and said, "Do me intros like J. Cal." And they were terrible.
1:17:06
So, I think they have a job for another year.
1:17:07
I asked the AI to pretend you're the All-In pod besties telling Uranus jokes.
1:17:13
That would be pretty funny.
1:17:13
Sorry, let me just say one more thing about this Open AI thing.
1:17:16
I do think that the biggest and most interesting thing to think about is how this will disrupt the search box.
1:17:25
The The search you know, the way search works at Google, you know, and internet search is there are these kind of servers, these web crawlers that go out and gather data.
1:17:34
Some of them are structured data feeds and some of them are just crawlers.
1:17:37
And then that data is indexed or in the in the structured way it's kind of made available for for serving directly on the search page.
1:17:47
And so much of that is is indexing.
1:17:47
So, I search for a bunch of keywords.
1:17:50
Those keywords and and perhaps with some natural language context are matched to a result page and I click on that and it and it's linked out.
1:17:58
Years ago, Google started a product called the one box where they could take structured data like what is the weather in San Francisco today?
1:18:03
And that the top of the search result page just presented that data cuz it knows with high certainty the question you're asking and it knows with high certainty the answer it can give you.
1:18:12
Yeah, clipped it from somebody's website, right?
1:18:15
So, if that starts to become everything, then that one box interface and it's not just Google's ability to access all this data and index it and serve it and store it, there could be a lot of competitors to the one box and a lot of competitors ultimately to search.
1:18:30
And ultimately, you know, Google's core product, their their their search engine, could be radically disrupted by an alternative set system or set of systems that have more of a natural language chat interface.
1:18:43
Which Literally which is literally why Google bought DeepMind and there were a collection of human-powered search engines Mahalo included, ChaCha, answers.
1:18:54
com who were trying to do the human-based version of this. It just didn't scale.
1:18:57
We We don't want to get ahead of ourselves because one of the things we don't know is how much is going on in DeepMind.
1:19:01
They're they're they're not very open like Open AI is.
1:19:05
They They talk about some of the advanced frontier stuff like AlphaFold and so on and they've been public about that, but a lot of that is really to generate interest and hype in what's next.
1:19:13
But my understanding is DeepMind's been applied to everything from ads ad ad optimization, but also the ranking on YouTube videos to get people more engagement on YouTube, etc. etc.
1:19:23
So there's all these ways that DeepMind's been applied within Google services that we don't see.
1:19:29
and certainly within search. Yeah.
1:19:29
But the question is is there an entirely new interface for search Yes.
1:19:35
that risks Google's core search business.
1:19:38
And I think that there certainly will be a lot of money thrown at this and if anyone has any interesting ideas, send me an email. Sacks and then Chamath. Yeah.
1:19:45
Yeah, I think I think that's a really interesting point.
1:19:46
I saw a thread on this where somebody was asking GPT, you know, a bunch of questions like they were like generally like coding questions and they were actually comparing the result in Google versus GPT and Google would just give you a reference to like a link to some page, whereas GPT-3 would actually construct the answer, like a multi-paragraph answer that was far more detailed and in a way user-friendly. Yeah.
1:20:16
And whereas like the Google page would kick you over to a reference where it was like this 1 2 3 sort of maybe someone had created a checklist but it just wasn't that detailed.
1:20:23
It really is pretty interesting.
1:20:25
I thought Andreesen tweeted a really interesting example as well where he asked GPT to create a scene from a play starring a New York Times journalist and a Silicon Valley tech entrepreneur.
1:20:41
They're arguing about free speech and each person asserts the view associated with his profession and social circle.
1:20:48
We don't need to read the whole thing but I thought this was like spot on where I was actually like both sides are making their best arguments and it's like to each other in a conversation that seems intelligible.
1:20:59
Like they're making their points at the right time in the conversation.
1:21:02
It's like they're playing off each other.
1:21:04
In other words, it actually reads like a conversation.
1:21:07
I actually thought this one was more impressive than the one with the bestie impersonation because I agree.
1:21:11
I actually thought that the one about all in didn't really capture our personalities per se but this one actually does a pretty good job capturing the arguments in this debate. So Chamath? pretty impressive. Any thoughts here? Yeah, a lot.
1:21:25
I mean, I've been spending a lot of time learning about this area.
1:21:31
Six years ago, a team that I partnered with who was at Google that built TPU, we've been building silicon for this space.
1:21:38
So we've been kind of going from the ground up for the last 6 years.
1:21:41
Couple things that I'll say.
1:21:43
The first is that I think we're going to replace SAS with what I call MASS, which is models as a service.
1:21:52
And so, you know, a lot of what software will be particularly in the enterprise will get replaced with a single use model that allows you to solve a function.
1:22:00
So, these chat examples are one and you can name a bunch of SAS companies that were purveyors of SAS that'll get replaced by essentially GPT-3 or some other language model.
1:22:12
And then there'll be a whole bunch of other things like that.
1:22:14
If it's a, you know, expense management company, they'll have a model that'll allow them to actually do expense management or blah blah blah forecasting better.
1:22:22
So, I think SAS will get replaced over time with these models incrementally. That's phase one.
1:22:31
But the problem with all of these models, in my opinion, is that they're still largely brittle.
1:22:33
They are good at one thing.
1:22:37
They are a single mode way of interfacing with data.
1:22:44
The next big leap, and I think it will come from one of the big tech companies or from OpenAI, is and we talked about this, I talked about this a few episodes ago, a multimodal model which then allows you to actually bring together and join video, voice data in a unique way to answer real substantive problems.
1:23:01
So, if I had to steel-man the opposite side reaction, so I think there's a lot of people gushing over the novelty of GPT-3, if I had to or ChatGPT, if I had to if I had to steel-man the opposite, what I would say is it's going to get somewhere between 95% of all of these very simple questions right because they're kind of cute and simple.
1:23:25
There there is no consequence of saying write a play because there is no wrong answer. Right?
1:23:30
You either kind of it it it it tickles your fancy or it doesn't.
1:23:32
It kind of entertains you or it doesn't.
1:23:37
When this stuff becomes very valuable is that when you really need a precise answer and you can guarantee that to be overwhelmingly right.
1:23:44
That's the last 1 to 2% that is exceptionally hard.
1:23:49
And I don't think that we're at a place yet where these models can do that.
1:23:51
But when we get there, all of these models as a service will be very much commoditized.
1:23:58
And I think the real value is finding non-obvious sources of data that feed it.
1:24:07
So, it's all about training.
1:24:07
So, meaning you can break down machine learning and AI into two simple things.
1:24:13
There's training, which is what you do asynchronously, and then there's inference, which is what you're doing in real time.
1:24:17
So, when you're typing something into chat API or a ChatGPT, that's an inference that's running and then you're generating an output.
1:24:25
But the real key is where do you find proprietary sources of data that you can learn on top of?
1:24:33
That's the real arms race.
1:24:36
So, one example would be, let's say you build a model to detect tumors. Right?
1:24:41
There's a lot of people doing that.
1:24:42
Well, the company that will win may be the company that actually then vertically integrates, buys a hospital system, and get access to patient data that is completely proprietary to them and covers the most number of women of all age groups and of all ethnic, you know, ethnic categories.
1:25:01
Those are the kinds of moves in business that we will see in the next 5 to 10 years that I find much more exciting in trying to figure out how to play in that space.
1:25:10
But I do think that ChatGPT is a wonderful example to point us in that direction.
1:25:14
But I'm sort of more of that case, which is it's a cute toy, but we haven't yet cracked the 1 to 2% of use cases that makes it super useful.
1:25:24
But I think the first step but just I just but the first step will be the transformation of SAS to MASS and then from there we think we can try to figure this out.
1:25:32
It reminds me of in a way when you when you give that description of like, "Hey, this is really interesting but it's not complete."
1:25:38
Is remember when GPS came out and like people were like doing turn-by-turn navigation, they'd drive off the road because they were trusting it too much.
1:25:45
And then, you know, over 20 years of GPS, we're kind of like, "Yeah, it's pretty bulletproof but keep your eyes on the road."
1:25:50
Same thing that's happening.
1:25:51
Uh and and these changes tend to be slow.
1:25:55
These last 100 or 200 basis points literally takes decades. Exactly.
1:25:57
So, the last 15% of self-driving is like the decade-long, you know, slog. a century. 15% may take a century.
1:26:03
But the last 2% will take a few decades.
1:26:06
It's like the change happens very slowly and then all at once.
1:26:08
For people who don't know what a a TPU is, that's a tensor processing unit.
1:26:11
This is Google's application-specific circuits, right?
1:26:17
And custom silicon that they invented for TensorFlow at the time.
1:26:19
Yeah, so if you want to try now the modality of AI, we've changed that as well.
1:26:23
So, now we're totally in the world of transformers.
1:26:24
So, we're not even using you know, you're not letting the tensors flow the way they used to.
1:26:30
All right, there's been a slowdown in SAS.
1:26:31
Sacks, what is happening in the software as a service world?
1:26:35
There was a good update by Jamin Ball who works for Altimeter, our friend Brad Gerstner.
1:26:40
He does these really great updates on what's happening in the SAS world.
1:26:43
The big thing this week is that Salesforce had its quarter and I I would consider Salesforce to be the bellwether for the whole SAS category.
1:26:51
I think they're the largest pure SAS company.
1:26:53
They were the first multi-tenant SAS like company at scale.
1:26:58
And what they've shown is a huge slowdown.
1:27:01
Basically, their net new ARR that they just added in the previous quarter dropped 2/3 compared to the previous quarter.
1:27:09
But because their sales and marketing spend was the same as the previous quarter, it exploded their CAC payback, which means the amount of time it takes to pay back your customer acquisition cost for a given customer.
1:27:23
So, you see they're 155 months it would take now to pay back the customer acquisition costs. That's over 10 years. That doesn't work.
1:27:32
I mean, I think before this quarter, it was more like 2 and 1/2 years.
1:27:36
That's something that you can afford.
1:27:38
A company can't, you know, if you're spending 10 years of you know, gross profit on a customer to acquire them, the business doesn't make sense.
1:27:46
So now, I'm not saying any of this to pick on Salesforce.
1:27:48
It's an exceptionally run company, you know, one of the absolute best.
1:27:53
Uh Marc Benioff, fantastic CEO, founder, great human being.
1:27:58
But I think the point here is that what you're seeing is the whole SAS industry is really slowing down here.
1:28:06
In the first half of the year, you saw SAS valuations correct.
1:28:11
Now we're actually seeing SAS top line correct.
1:28:13
And, you know, there's an interesting question here.
1:28:16
If your CAC payback goes from 2 and 1/2 years to 10 years, you have to bring your CAC down. How do you do that?
1:28:22
You can either reduce marketing or you can reduce sales.
1:28:25
So, in other words, you can reduce cut the sales team.
1:28:28
You can either cut people and head count from your own team or you can cut spending you do on advertising or events or money that you spend on other companies.
1:28:38
Either way, there's going to be a big contraction in jobs basically around this industry.
1:28:43
And I think that what that could do is cause a vicious cycle where Death spiral.
1:28:52
seeing I wouldn't say death spiral.
1:28:52
I think this vicious cycle for the next year or so where seat contraction becomes the norm instead of seat expansion.
1:29:00
So, if you go back over the last 10 years, a major tailwind at the backs of SAS startups has been that every year year, you start with 120%, 130%, 150% of last year's revenue just from your existing customers. Why?
1:29:15
Because they were hiring more and more people and they needed to buy more and more seats.
1:29:18
But now head count growth is frozen, and in fact companies are doing major layoffs.
1:29:23
So, the baseline for next year could be seat contraction.
1:29:25
So, instead of starting with 120% of last year's revenue, you might start with 80 or 90% because there's going to be so much churn.
1:29:33
So, I think that, you know, SaaS companies need to take this into account.
1:29:37
This idea that like growth is on autopilot, that could start to go in reverse.
1:29:41
I don't think permanently, but I think for the next year or so.
1:29:43
This is why I also I tweeted, you know, 2x is the new 3x.
1:29:47
If you can grow 2x year-over-year in this environment, that is as good as or better than growing 3x last year. Oh, it's clearly better. Clearly better.
1:29:55
Like a lot of companies that weren't that great could grow 3x last year because it was so times were so frothy.
1:30:01
Everyone was buying everything.
1:30:03
But now it is going to be really hard to even double year-over-year.
1:30:06
Companies need to take that into account into their financial planning.
1:30:11
Uh you need to restrain your burn because a lot of the revenue that you predict is going to be there may not be there. All right.
1:30:17
Uh thanks so much to the secretary of SaaS.
1:30:19
I think you got a board meeting. I do. I got to run.
1:30:21
One of the interesting things I saw in terms of use cases is somebody used the chat GPT to describe rooms.
1:30:27
Then they took the descriptions of those rooms, and then they put them into like Dolly or stable diffusion, one of those, and it created the visual.
1:30:35
I'm curious if you think, you know, the self-driving APIs uh and machine learning that's going on, and then you got images, then you got chat, maybe you have proteins going on with the AlphaFold stuff.
1:30:48
When these things start talking to each other, is that going to be the emergent behavior that we see of general AI, and that's how we'll interpret it in our world is these hundred different vertical AIs uh hitting some level of reasonableness to Chamath's point on
1:31:06
data sets, and then all of a sudden the self-driving AI is talking to the one that's looking at cancer and tumor diagnosis and the chat and the image ones, and maybe stable diffusion, the protein AI, and the one that's looking at cancer cells start talking to each other. Yeah, I'm not sure that's as likely as
1:31:20
Yeah, I'm not sure that's as likely as the there's a lot of solutions that will emerge within verticals, and I think you can distinguish them.
1:31:32
So, I kind of gave this example a few months ago.
1:31:37
If you remember Kai's Power Tools was a plug-in for Adobe Photoshop.
1:31:39
It came out in 1993, I believe. Of course.
1:31:41
And Kai's Power Tools completely transformed the potential of Adobe Photoshop cuz Photoshop had all the basic brushing and editing capabilities within it.
1:31:50
Kai's Power Tools was statistical models that basically took the matrix of the pixels, and, you know, created some evolution of them into some visual output like a blur.
1:32:03
And so, you could blur motion blur something, and you could change the parameters, and now your photo looked like it was going through a motion blur.
1:32:10
Ultimately, Photoshop bought and implemented those tools, but those were similar.
1:32:14
They were statistical models that made some representation of the input, which was the image, and then created an output, which was an adjusted image.
1:32:22
I would argue that that is very similar, although the models behind it are very different uh in terms of the contextual application um of statistical models in software.
1:32:34
And you could see stuff like, for example, a chatbot that replaces help me figure out whether my credit card charges are correct or not.
1:32:41
Instead of having a customer service agent, an offshore service agent helping you resolve that.
1:32:48
Or help me return my item.
1:32:51
Um or there are very specific kind of verticalized applications that can plug in that ultimately replace what was manual and human driven before cuz humans used to manually make the motion blur in Photoshop, and then it was automated with these software packages.
1:33:04
And I think you can kind of think about it in that same way that these these are known knowns.
1:33:09
They don't require necessarily human physical labor or some you know, human responsiveness.
1:33:15
So, if 95% of the work can be handled, it will get handled by some verticalized solution.
1:33:19
So, I think the physical labor versus the non-physical labor is one way to think about the distinction.
1:33:25
Meaning, is there some change in the physical world?
1:33:29
Driving is absolutely a change in the physical world.
1:33:31
You have to move physically through space.
1:33:32
So, that one is a very distinct class.
1:33:34
All the stuff that's like communication, imagery, static imagery, audio, and then visual, video, there's some stacking that happens there.
1:33:45
And some of those will be kind of siloed, and then some of them will merge, and you'll have these kind of unique kind of combo models.
1:33:52
And so, look, as they start to work together, I think we'll see them, you know, completely rewrite some of these verticals like movie production or music production, right?
1:34:01
Or advertising, or we're seeing it now with with video and and creative arts uh with um you know, some of the the the visual stuff on open AI.
1:34:11
a lot of journalism, a lot of creative arts have become the wisdom of the crowds over the last two decades where, you know, artists were looking at the collective works of the internet, interpreting it, and then coming up with content, which is kind of what these AIs are doing.
1:34:26
Uh and then who legally owns the collective content is going to be a big question.
1:34:31
Chamath, you talked about data sets, and Microsoft is being sued right now in GitHub because they uh used open source to create tools in AI to help augment programmers.
1:34:41
Like, while they're programming writing code, it gives them suggestions.
1:34:45
And now the open source community is suing them for using their data set.
1:34:49
So, what what do you think about the legality of data sets, Chamath?
1:34:52
And and should they get some kind of protection if you make a GPT-3 based on Quora or based on Wikipedia, should you have to get approval to use that data?
1:35:01
What's Wikipedia example?
1:35:03
Yeah, it's it's the exact opposite.
1:35:06
They say that this is actually your work.
1:35:09
Um and I think that that's the right legal framework.
1:35:10
But the the answer to your other question is this is why I think the hunt for proprietary data actually becomes the hunt that matters.
1:35:19
All of this other stuff I think is a lot less important because I think you have to assume that all of these models will eventually just get commoditized.
1:35:24
So, they'll be they'll be a, you know, like you see like Jasper AI, and you see a bunch of these generative AI companies. It's really interesting.
1:35:33
But the problem is when you sit it on top of the same substrate, you'll have a convergence. Right?
1:35:38
Everybody's chat model will eventually look and sound and feel like the same thing unless you're giving it a few special ingredients that other people are not.
1:35:48
And so, it's the hunt for those ingredients that will make this next generation of of models really valuable.
1:35:54
So, to give it an example, you'd have Wikipedia, which is creative commons, anybody can use.
1:35:58
But Quora as a data set, not everybody can use.
1:36:00
That's owned by a company.
1:36:01
So, Quora would have an advantage. Take an extreme example.
1:36:04
If Quora didn't allow themselves to be crawled, right? Okay? Which they don't. Yeah.
1:36:13
And but then the and then they developed their own language model which used the best of the internet, so call it, you know, GPT, and Quora, Mhm.
1:36:21
maybe they are slightly better in certain domains than others.
1:36:26
The The other extreme example is the one that I used in health care, which is you know, if you have access to patient data that you will not license to anybody else, you know, it stands to reason that that model actually then has much better chances of uh highly effective clinical outcomes versus any other model.
1:36:44
Apple Watch comes to mind, right?
1:36:46
Apple has all that watch data.
1:36:48
If they could pair that with Epic's, you know, data set, you know, like what could they do together?
1:36:53
So, this is going to be like this is the the new oil is going to be data.
1:36:58
And and by the way, like to to to to talk about Apple for a second, the smart thing is they've gotten so methodically.
1:37:03
They've never touted the AI.
1:37:03
You know, they introduce one or two distinguishing features every year, right?
1:37:09
So, like the the the ECG, which was introduced many, many years ago, is has only gotten slightly more usable like five or six years later, but in the meantime, there's, you know, tens of millions of watches collecting this kind of data.
1:37:23
So, to your point, it's it's using these devices as Trojan horses to collect training data. That is the oil.
1:37:31
Uber and Tesla have all this data of the data being collected by, you know, the Well, hold on.
1:37:38
uh phones or the cameras in the cars.
1:37:40
The other difference though is that you have to be in a realm where you don't need regulators to go with the last mile.
1:37:47
So, the problem with ADAS, I think, or level five autonomy is that eventually you get to a point where even if the model becomes, quote unquote, perfect, you still need regulatory approval.
1:37:58
And what I'm saying is I think you have to focus on areas of the economy that are not subject to that or where the regulatory pathway is already defined.
1:38:08
So, for example, if you use a health care example, let's say that you had the largest corpus of breast cancer image data, and you could actually build an AI that was a much better classifier for tumors versus other things, the FDA actually has a pathway to get that approved very quickly.
1:38:24
The problem with, you know, level five autonomy is that there is no clear pathway.
1:38:26
It's not Again, we go back to almost the a crypto example.
1:38:30
We don't really know who will govern that decision, and we don't know how that will be governed.
1:38:34
So, I think the the thing that investors have to do, and entrepreneurs, entrepreneurs have to pick their end market very carefully, and investors have to realize that this dynamic exists as well.
1:38:44
If you're going to do this right, make your money.
1:38:46
Imagine the Robinhood trading you know, trader data set watching people sell in shares, and then predicting markets with it with AI.
1:38:52
I mean, it could be crazy.
1:38:55
that or payment for order flow that's used by Citadel and the other big But not AI, right? So, then But who knows? Maybe they are using AI.
1:39:00
No, no, no, no, no, no, they are.
1:39:01
I I I I can tell you as somebody who sells We sell a lot of machine learning hardware into this market.
1:39:09
The biggest buyers are the US government and these ultra high frequency trading organizations.
1:39:15
Friedberg, any final thoughts here?
1:39:15
I'll give you the final word.
1:39:17
How could this affect astronomy?
1:39:17
How could this affect, you know, our search of the galaxies, you know, going out past Pluto, Saturn?
1:39:26
Breaching Uranus, any any of those things, how could it impact?
1:39:31
Any I'm trying to get a Uranus joke to land.
1:39:35
Help me out there, Chamath.
1:39:37
You got to know your the I think you need to have more space-related um Yeah, let's workshop this one with me.
1:39:44
Or gut or gut biome-related in you know Or gut biome, yeah.
1:39:47
So, how would this affect Supergut?
1:39:48
Use the promo code twist.
1:39:51
have to trick Friedberg into thinking we're asking a serious question.
1:39:53
Get him down the science path and then rug pull him.
1:39:56
Now, that's the right use of a rug pull. rug pull, exactly. Okay, let's do it. Here we go. Let's workshop it.
1:40:00
the wrong So, uh tell us, you know, when you're doing like Supergut, use promo code twist to get 25% off.
1:40:06
When you're doing Supergut, you're analyzing people's guts.
1:40:09
How would you then have machine learning in this, you know, API this chat API and GPT-3?
1:40:14
How could that help with processing all of that, especially when it passes through Uranus? Friedberg.
1:40:23
Okay, over there, Friedberg. You are hungover.
1:40:26
I'm hungover, but I also had like a 7:00 a. m.
1:40:29
board meeting, so I'm also just a little beat up.
1:40:33
Were you grumpy on the board meeting?
1:40:34
Did you get a little cantankerous with the You were fine.
1:40:37
You just kept You kept the rage in control.
1:40:39
have my caffeine fuel and then I kind of cranked down afterwards. All right, everybody.
1:40:44
We will see you next time for the secretary of sass, the dictator, and the sultan of hungover.
1:40:51
We will see you next time. Bye-bye. Love you guys. Bye-bye. Let your winners ride. Rainman David Sacks.
1:41:07
And then Sacks we open sourced it to the fans and they just got crazy with it. Love you guys. Besties are gone.
1:41:30
We should all just get a room and just have a one big huge orgy cuz they're all just useless.
1:41:33
It's like this like sexual tension that they just need to release somehow. Wet your bed. Wet your feet. Wet your feet.
1:41:44
We need to get merchants out back.