Sam Bankman-Fried testifies, says he "skimmed over" FTX terms of service
arstechnica.com
arstechnica.com
This line encapsulates everything I've thought about this trial.
Feels like watching the man scream at the steam roller in Austin Powers.[0] He just can't keep his feet out of his own mouth. At this point I can't tell if it's hubris, or just straight ineptitude.
This line captures the recent wave of Silicon Valley bullshitters so perfectly.
No. Rambling idiots were historically judged.
The quick-talking stream-of-consciousness bullshitter relies on rapid technological change to suspend listeners' disbelief. Because ex ante, you can't know if you've got someone brilliant but nervous (or simply neurodivergent), or one of these.
SBF is the same classic huckster, but wrapped up in a hoodie.
Who were widespread in the industrial era. Less so before. (In part because peasants didn't have the disposable income to scam.)
Two years later.
"How it's going?" "Making progress, not quite there yet. I'm going to need more funding."
And so on.
FTX founder gets by on 4 hours' sleep and multitasks on 6 screens https://news.ycombinator.com/item?id=29580668
In 2021, FTX had a funding round of $420.69M https://news.ycombinator.com/item?id=33530674
Sam Bankman-Fried and Matt Levine on How the Crypto Market Works https://news.ycombinator.com/item?id=28075024
[1] https://www.forbes.com/sites/willskipworth/2023/10/16/sbfs-l...
Secondly though he seems to have quickly latched on to a specific sort of messaging that's built around "honesty" and admitting his own failings, not in a real way it seems though. I think he's seen a bunch of people get "cancelled" in recent years and seen that this sort of admission of guilt tends to work well for people. What he doesn't realise though is that this isn't him being accused in the court of public opinion about something the public don't agree fully on, it's a legal accusation in court about specifics that the law is clear on. I'd say this was just bad advice he was being given, but it seemed to happen so quickly that I think it's just his own bad take on the best way out.
I mean his publicity tour immediately after FTX went under worked pretty well for a while. He got Kevin O’ Leary to blame CZ in front of congress. And Michael Lewis painted a pretty positive picture. He talked to everyone he could and the vast majority of the media were not sure if he did or did not commit a crime.
All he has to do is convince the jury to have a reasonable doubt.
The best explanation I’ve seen is that the book was mostly done when it all came crashing down, and SBF was already the befuddled hero and herald of a new age, and Lewis just didn’t want to rewrite the whole damn thing.
But I know Lewis has gone through some personal tragedy, and I think it has softened his gaze, at least upon young people who act like a foundling in Brave New World.
If he's convicted it won't be on the details, but on credibility. The "I don't recall" strategy won't fool a judge, but it might fool enough jurors.
I think the prosecution is going to have to work unexpectedly hard to nail him to the wall.
But I also suspect there's more to the story than we're being told.
The fact that so much of the money seems to have gone on bribes and political donations, the fact that his parents are lawyers but apparently were shocked - shocked! - to discover he was doing something illegal, the fact that he seems so casual about the whole thing - it's all a little strange.
His father "is considered a leading scholar in the discipline of tax law" and is a Stanford Professor of Law and Business.
So with no obvious warning signs this must have come as a complete surprise to him.
His only plausible strategy is confusion and delay to buy time. A shame that it ends up burning millions more dollars of prosecutor resources.
slaps hood “These gears turn slowly but watch how fine they grind!” Money well spent imho.
I think he's clearly very smart, but not nearly as smart as he thinks he is.
It's mean spirited, but the phrase "she's what stupid people think smart people sound like" struck me at the time. I think the same may be true here. This kind of behavior has been reinforced so much in their lives that in some cases they start to also get high on their own supply. But they to are profoundly stupid people, and start to think that how they interact with the world is what smart people "sound like" and it perpetuates into these spectacular fraudulent failures.
I bet it's because she spoke with confidence. Stupid people follow confidence disregarding everything else.
There are a subset of people, and Elizabeth herself who, to this day, believe we and the world were robbed of Theranos' revolutionary advances, because they "really, truly, were this close and would have done it by now", and think that that is the miscarriage of justice.
https://www.nytimes.com/2022/11/29/technology/sam-bankman-fr...
Earlier comment:
The failure was in trying to be the everything box and committing fraud to get there. I actually do think it would have been revolutionary and the failure of Theranos has set the field back.
With their own equipment? I don't believe so. Theranos as a company could reliably do this... they just did it with Siemens equipment with the labels removed.
> I actually do think it would have been revolutionary
It would be, were it possible. Not that we can't find things to be possible that were previously thought not to be, but this is pretty well established biology, and has been steadily researched for decades by both academia and industry (all of the big lab players would love it if they could do droplet analysis)... if there is an advance there, it's a pretty safe bet to say it was never coming from a 19 yo freshman chem eng dropout.
The phrase is a 'poetic' comeback. It's successful in the sense that it's exactly what everyone wishes they could come up with when put on the spot, so it's easy to remember. Like you say, it is unfortunately rather mean spirited, and I say unfortunate because I suspect that it contains truth and usefulness.
My running theory is that the structure of natural language and jargon is encoded such that it carries with it useful information. And this sort of makes sense. The problems we encounter in reality have a certain structure to them. And the natural progression of encoding is to make frequent things short and infrequent things long (thanks information theory). Over time communities shape language and jargon into the structural shape that is most useful to solve the problems they encounter.
Primarily, I've started down this road to try and get my head wrapped around LLMs. Why are they useful at all? And how can something like prompt engineering even exist? The answer being (maybe) language structure sometimes matches problem structure. And prompts from people who "know" how to speak "right" carry additional structure that helps the LLM to provide better results.
However, I think the same can be applied to "what stupid people think smart people sound like". If I'm right then there's a certain sense in which (for extremely well known problems) talking "correctly" helps you to be successful (although all things being equal actually knowing is going to be a better scenario).
And if you abstract this, then there's a benefit to not knowing what the truth is (it frees up your time) but knowing what the sentence structure of "smart" approximately is (it allows you to pick winners to invest in).
So perhaps we can de-meanify the phrase: "She's what non-experts think experts sound like." Or to maybe up the poetry: "Bach and pop use the same notes, but as the children mature so does their taste."
It is what people without any real depth of thought think that geniuses sound like.
Stringing together a lot of occasionally insightful sentences without having any overarching and consistent point to the whole thing.
And if you're conditioned by social media to have a low attention span and you have difficulty noticing how frequently the argument skips between tracks then it all sounds brilliant to you.
> Defense lawyer Mark Cohen did his best. Unfortunately for him, the cross-examination was conducted by Sassoon, who looks like someone who uses “summer” as a verb, and often appears deceptively timid, with her hands held close to her chest. In her cross, she simply unhinged her jaw and ate Bankman-Fried.
She's here on Mastodon https://hachyderm.io/@molly0xfff and also did live stream on her YouTube channel https://www.youtube.com/@molly0xfff/streams.
I feel like he'd need a game plan, either waiting for something additional his team can bring in, or seriously thinking that the practice run would result in a higher likelihood of a not guilty verdict. I don't think either of these is likely, and think the prosecution would likely do better out of the practice run.
> Another $413 million had gone directly to Bankman-Fried himself, through his wholly-owned company Paper Bird.
This article says $7.3 Billion have been recovered: https://www.reuters.com/technology/bankrupt-crypto-exchange-...
If I steal your entire wealth and your family ends up on the street, it doesn’t matter that I wasn’t violent.
Hope he gets what he deserves: a ton of prison time.
Once upon a time a man was accused of a crime. [..] The king of that land sentenced the man to death, and he pleaded for his life, offering to teach the king's horse to talk if he would be spared. The king agreed; if the accused could teach the horse to talk in two years, he would be pardoned. Otherwise, the original sentence would be carried out.
Sometime after that, the man's best friend asked him why he made a promise he couldn't keep, and why he was acting so cheerful about the whole business. The would-be horse teacher said, "A lot can happen in two years. The king could die. I could die. Or the horse could talk."
A hung jury buys you, I dunno, another year of freedom/jail instead of prison.
Perhaps he's planning to tamper with the jury. Perhaps he thinks he can bribe jurors; I imagine he still has access to plenty of moolah, and he certainly has some powerful friends.
He's valuing the entertainment of millions above years of his own freedom.
More substantively I wonder if he will try to argue his use of the money towards effective altruism made fraud OK during sentencing - surely the money was better spent on charity rather than remaining with the credulous.
Why he’s a modern day Robin Hood! And who could be more deserving of charity than himself?
Ngl, I feel terrible for all the people that lost FTX money, but it is kinda amazing to see the world see what a terrible mess and how irreducibly dumb most of the effective altruists are.
Kinda like when gwern went off on his eugenics rant and people went: are all of these "rationalists" libertarian lunatics?
Check
> Bankman-Fried’s defense, in the direct testimony, was trying to put blame on the lawyers: FTX chief regulatory officer Dan Friedberg... A significant thrust of questions was about a bank account controlled by Alameda Research that did not bear Alameda’s name; it was instead called North Dimension, it came into existence around 2020 — that is, while Bankman-Fried was still Alameda’s CEO — and it was where FTX customers were told to wire their funds. Bankman-Fried said that the North Dimension bank account was all Friedberg’s idea. Sassoon asked if Bankman-Fried, as Friedberg’s boss, had given him any direction, or if Friedberg just popped ideas across Bankman-Fried’s desk. The word-salad began again, interrupted only by the judge saying things like “So I take it the answer is you don’t remember; is that about it?” or “Listen to the question and answer the question directly.” ... Then Sassoon did what I had been waiting for: pointed out that Bankman-Fried had been the one who hired Friedberg in 2020. She asked if he’d been hesitant to hire a general counsel. He said he had been hesitant to hire the wrong general counsel — “I did want a general counsel who was comfortable with reasonable risks.” Sassoon then asked if Bankman-Fried was aware of Friedberg’s history of working at a company that had an insider trading scandal? (The company in question, by the way, is UltimateBet.) That there had been a criminal scandal? “Were you aware that Dan Friedberg used illegal narcotics with your employees?”
Checkmate
He's clearly "smart" in the sense that he knows how to read a book and operate a computer, but we need to stop treating him as anything other than a middling dude who stumbled into a grift.
As a result my "we don't like this guy so we'll talk up any minor negative thing we find" detectors start to go off. I think he just took money from the wrong people, kinda like Madoff.
2) TOS in financial sectors are generally very well thought out and would not surprise me at the startup level that executive leadership should have a solid understanding of what it does. In this case the TOS in question would allow/disallow FTX the exchange from lending money to sister org Alameda. This seems like a pretty crucial thing that the CEO, SBF, should know.
1. A CEO obviously should have a basic understanding of the rules on using customer assets regardless of whether he read the TOS. It's like Sam claiming he didn't know he had to pay taxes since he never read the tax code.
2. More importantly, when issues arise, a non-criminal organization would have called up legal and said "can we do this?", and waited for them to review the TOS, respond with an answer, and follow that. Sam seems to have purposefully avoided oversight, building out a Board, building out legal, sharing information, etc.
Same thing happened with that one woman that ran “Theranos”. Just know the right people. Although I think she was a slightly better public speaker. The yoda voice shit was weird though.
The business idea of venture capital is not to follow through with the investment all the way to a mature company with long term profits. It is to offload to later investors. It's hard to fault anyone as long as they go in with their eyes open.
One of the most effective sales techniques is to sell the customer's own dream right back to the customer.
To qualify the prospects you need a marketing strategy that builds a pipeline of potential customers likely to possess the type of dream that you can work with.
Here there was no persuasive salesmanship needed to convince "investors", someone else had already done that in advance, or the customers had built interest on their own.
Sam was just there to use their money, and if a certain amount of losses were inevitable that was supposed to be overcome by greater amounts of gains in the long run. Skimming just a little bit off the top to where it's obscured by the noise of the losses is what lots of financial outfits have done for centuries. The higher-integrity firms have decades of experience carefully disclosing their terms for this activity to their customers, which is one of the reasons why more persuasive sales efforts might be a lot more common.
As soon as Sam started skimming without proper disclosure, that's what made it ordinary theft.
Testifying opens SBF to cross-examination. He is so arrogant that he thinks he will ace it, but evidence points towards him fumbling it.
“advice of counsel” defense waives the right to attorney-client privilege.
I think it's a hail Mary. All his buddies were offered deals to flip on him because he was the big fish. There's no bigger fish for him to flip on, so prosecutors will throw the book at him, no reason to go easy. A conviction is pretty certain and would be pretty bad, so any strategy that could make things better but can't make things worse is ideal.
They all belong in jail.
Sam is just a massive sociopath, born out of this.
What happened there ??? The guy who calls out a lot of fraud, now can't see it?
I'm disappointed because I found his books enjoyable ... although maybe in retrospect I can't remember that much from them. Maybe he was never really calling them out, but just liked the story.
I know there have been a bunch of articles about this -- but what's the tl;dr ?
I guess he was always up close and personal with everyone, because that's how he got access to tell these interesting tales. And then you sort of compromise your own viewpoint
It seems that he likes SBF on a personal level, which I think is reasonable, but then it seems that he falls short of making strong statements about the fraud. I think people have interpreted this combination as an implicit support of SBF, where it could in fact just be that, publishing before the trial, Lewis didn't want to "call it" the wrong way and limit the relevance of his book.
If anything, when the trial is over they re-release with an epilogue evaluating the trial.
It would probably be more than a chapter of material too.
So any Don Quixote will look like Robin Hood to him.
Flash Boys was just taking sides in a dispute with no real good guys, but the upstart/little guy to him was default "good guy". He took some more explicitly political positions in some of his other books. And now his SBF/FTX fanboyism falls under the same category that "the little guy" has to be the good guy.
The Big Short is probably his last great book / and film.
Definitely present in Moneyball where the A's are the good guys - they play a similar role as Burry, exploiting inefficiencies in the market created by the big, bad guys (e.g. the Yankees, or housing lenders), for profit.
Through line of his books seem to be "little guy is good".
In many of the stories the little guy he valorizes is just another neutral actor in a zero sum game.
The Big Short it just so happens that the "little guy" who was right got to make money off fairly bad actors! I wouldn't say they were themselves "good", rather more neutral players in a financial market.
He now seems to have chosen a little guy (SBF/FTX) who were explicitly bad actors. Also interesting questions in the whole Blind Side story now.
I also recommend his interview of Matt Levine (5 episodes ago) to get a good TLDR of the book. Here's my TLDR on the book after finishing it, which mirrors Levine's closely: Michael Lewis sees that Sam thinks he's above the rules and acted with wild disregard for risk because he thought he was smarter than everyone else. It was all illegal and will land him in jail for a long time. But, he did it because he thinks he's the smartest man who ever lived and has the best intentions, not because he sees himself as a crook.
It's a minor nuance, but an important one. It speaks more to the ego of founders and children of privilege, than to a criminal conspiracy. The articles you've read don't like that nuance, and call Lewis soft.
>What happened there ??? The guy who calls out a lot of fraud, now can't see it?
He calls it Fraud many times in his book and in his podcast. He just doesn't call it theft.
Hasn't Lewis said as much?
I haven't read his books, but I think he writes about people that he basically admires. He doesn't set out to destroy his subjects.
He's decided to face cross-examination, so the judge held this session to check that his defence doesn't include saying stuff that's inadmissible. But from this account, it doesn't sound as if he's got anything interesting to say at all; and I find it surprising that someone supposedly so bright talks in "word salad", i.e. can't construct a coherent sentence.
I suspect the story about him being bright comes from people who believe wealth is proof of intelligence.
This court case stuff is a different skill set though and also I think it would be beyond almost anyone to talk themselves out of his situation.
Testifying opens SBF to cross-examination. He is so arrogant that he thinks he will ace it, but evidence points towards him fumbling it.
“advice of counsel” defense waives the right to attorney-client privilege.
Does it also remove his ability to appeal on a "competence of counsel" basis or is that only if he goes full-blown self-representation?
If you weren't aware of this consider it an awakening. One of the best ways to beat somebody in court is not to have a stronger argument but to prevent them from presenting evidence.
I'm curious whether or not this kind of thing was what the Constitution's authors had on mind.
(I've also heard that the meaning of "peers" has changed, but no idea if that's really true.)
>> One decision for Kaplan is whether Bankman-Fried will be allowed to blame FTX lawyers when the jury is back in the courtroom. As we previously wrote, this is called an "advice-of-counsel defense" in which SBF could argue that he sought advice from company lawyers, received advice that his conduct was legal, and "relied on that advice in good faith."
[0] https://arstechnica.com/tech-policy/2023/10/sam-bankman-frie...
Once that's on the the table, then you need to have asked the lawyers: can we do this? And then you need a clear response from the lawyers (i.e. a memo) saying: "Yep!"
SBF's testimony, peeling away some of the word haze, amounts to: "I think the lawyers were loosely aware of at least some of this stuff and I don't exactly remember them telling me we couldn't do it." That's miles short of the standard (Kaplan-summarized) test.
https://storage.courtlistener.com/recap/gov.uscourts.nysd.59...
At the end of the hearing, the defense attorney decided they weren't going to proceed with that defense, and instead go with the general defense of, uh, SBF thought the lawyers were telling him this was all kosher, so it can't be fraud.
In any case, he is actually testifying in his defense in front of the jury today, and Monday, and likely Tuesday. Based on this hearing, when he goes into cross-examination, the prosecutor is going to open him up and completely eviscerate him, with SBF happily handing the prosecutor the surgical tools to do so.
Wouldn't it be better if these journalists stopped coaching the accused?
The only other effect I see is prejudicial rubbernecking entertainment.
I imagine he's had good legal advice from good, smart lawyers. But I doubt SBF thinks they're as smart as he is - oh, no!
Which is why it's so bizarre that he expected to get away with blatantly "borrowing" billions of dollars without permission. Even if he didn't lost one penny he would still have gone to prison for many years.
People keep saying this. I’ve seen literally no evidence of it. Open to having my mind changed. (Working for Jane Street for three years is insufficient without knowing what he accomplished.)
Yes. Very deeply, yes.
The day we can define it is the day we can create it. Until then, genius and insanity will be indistinguishable until after the fact. (That said, there are clues.)
I suppose you can't; not definitively. I find it very hard to tell whether someone brighter than me is only a little brighter than me, or dramatically brighter than me.
My heuristics are:
* Nobody is dramatically brighter than me; I don't believe in geniuses.
* Bullshit is a giveaway. If I detect bullshit (e.g. "word salad"), then he's not brighter than me.
[Edit] In my experience, barristers/courtroom attorneys tend to be very bright, and also very good with language. Language is "what they do". Word salad doesn't seem to me to be a good strategy, when facing a good lawyer.
Social media has dragged us out of that infancy. But the bias, particularly among the pre-digital generations, remains. Bankman-Fried's word salads played that bias, which generated social proof. That, in turn, played into others' bias towards judging character by social proof. That generated wealth, which played into still others' quick filters.
So he takes the concept of having a paper trail of communications can put people in jail to heart - smart.
He then sets up their instant messaging to auto-delete such that there is literally no paper trail - dumb.
A smarter fraud version would have been to have clean paper trail of communications that were legal & a secondary channel where all the sketchier stuff goes on.
The even smarter version is to not discuss anything illegal in electronic form.
The smartest version being to just not do illegal stuff and run a legal business that makes money.
He should have read far more Matt Levine.
Well, yeah. But he belongs to a social circle that lurves technology; his generation does everything via mobile devices, AND all his buddies are into cryptocurrencies and fintech. They discuss everything electronically.
I wouldn’t say SBF is dumb, just that he was intelligent enough to get a crypto startup out of the ground, then he was sociopathic enough that the business ended up being a giant fraud machine.
He’s not like dumb dumb. It’s that his sociopathic tendencies prevent him of doing or continuing in any serious endeavour without taking supreme advantage of it. He was smart enough to make a business, but since he’s a sociopath he couldn’t deal with the fact that that business had to be reasonably honest, because to him people and money are just tools for him to get what he wants.
Im convinced by the trial and other info he is actually an EA, he just has a radically different understanding of what that means than you or me. For him, killing 500 people to save 500,000 is a no brainier, the same way stealing from millions of customers to save a company that can help tens of millions is just obviously correct.
He doesn’t care because he can’t care. It’s not in his DNA.
Really? It seems to me all evidence points to the opposite.
Clearly, it is possible for the same person to be both quite smart and deeply, profoundly stupid.
To use a relatable metaphor, there's a reason D&D splits mental stats into separate components: intelligence, wisdom, charisma and alignment aren't the same! Except in real life there's probably way more variables even than those and they're often correlated or anti-correlated in various ways.
This is the primary reason conversations on HN involving SBF, Elon Musk, Peter Thiel, Steve Jobs, Eliezer Yudkowsky etc can be so tedious.
It is certainly something everyone needs to be reminded of. Being (or sounding) smart on one subject doesn't make you smart on others by definition.
As someone who has surfed the Dunning-Kruger wave myself many times, I'm super cautious of looking out for other people doing the same.
It's totally possible for someone who is exceptional in one way (e.g. rich and famous) to be quite pedestrian, boring and mundane otherwise.
Depending on your views you might see Thiel as an exceptionally good tech exec but an exceptionally bad political kingmaker (unleashing forces he regrets & can't control).
But yes, also terribly mundane. Musk for example seems to fall for every midtwit conspiracy and agitprop post on the social media platform he owns, like your random poorly educated uncle.
And I think I’ve identified the cause: lack of continual contact with reality. I’ve yelled on a conference call with a bunch of Honeywell engineers “STOP FUCKING HYPOTHESING AND CHECK THE LOGS!” Smart people are prone to forming their own little bubbles; groups of smart people even moreso. They suffer from a lack of continually having their bubbles popped by abrasive facts, or dumb people who point out the obvious.
I’m having work done on my house, and my contractor is an endless source of simple, “this is how it works” data. I conceive of interesting structures for my wife’s greenhouse, and then he explains why it won’t work in single syllable words.
This is the primary reason conversations on HN involving SBF, Elon Musk, Peter Thiel, Steve Jobs, Eliezer Yudkowsky etc can be so tedious
Exactly.
Really smart people sometimes get lost in the structure of narrative and forget to do basic queries. I also learned that you gotta be careful who you criticize even if you are totally right. People won't change their opinion on your statement when the evidence vindicates you.
https://www.urbandictionary.com/define.php?term=failson#1748...
Anything related to executive functions, like like impulse control, irritability, attention problems, mental disorders, personality disorders, and many other things–including bad attitude and too much self confidence–can make smart people act foolishly.