content warning: this is basically a largely-unedited dissertation i wrote a while back about sam bankman-fried/the fall of ftx based on this book and a number of other sources

devoted readers will recall that the only thing i’ve written about crypto was centered around a cube made of solid gold. this is why my jaw dropped when somehow another crypto-related metallic cube makes an appearance in this book:

The third and final item on the list of Sam’s desires was the tungsten cube. The tungsten cube was a bit of a head-scratcher. The imaginations of crypto people everywhere, it turned out, had been gripped by tungsten cubes. Tungsten, the architects now learned, was the planet’s trendiest dense metal. Crypto people were then memeing about “the intensity of the density.” A company in the Midwest had supposedly created the world’s biggest tungsten cube. A mere fourteen inches, it had cost a quarter of a million dollars and weighed two thousand pounds. Sam had apparently ordered up such a cube, flown it into the Bahamas, and wanted it displayed on a plinth in his mini-city. The architects never were shown Sam’s dense and precious cube, but they nevertheless incorporated it into their drawings. “We designed a space for it,” said Ian. That space was the big atrium in the city’s main building. The tungsten cube would be the first thing a visitor to the global crypto empire would see. Rising from the sea of abstraction, the earth’s most concrete object.

this is almost certainly the most popular out of the three crypto books i read, likely to be found right now in a display near the front of your local (non-used) bookstore or in the non-fiction section of airport newsstands. that’s because the author is michael lewis, arguably the biggest name in finance writing, responsible for blockbuster books that also become actual blockbusters like the big short, moneyball, and the blind side. usually this is the sort of thing i would avoid at all costs (even free), but the story-behind-the-story has an irresistible twist: michael lewis originally set out to write a hagiography of (at the time) crypto golden boy sam bankman-fried (aka sbf), who falls right into lewis’ pet subject of “outsider whiz kid nobody believes in who uses math or something to become super successful and revolutionize some industry”. however, turns out sbf f’d up big time and so lewis accidentally ended up with a front-row seat to his spectacular fall. the question on everyone’s mind: how does lewis handle it?

well, people are saying that this might be the book that torpedoes his reputation and discredits him for all eternity. why? lewis not-so-subtly comes out in support of SBF near the end of the book, which is not a good look in light of the recent revelations from the currently-ongoing trial of sbf. the most damning testimony and evidence so far seems to indicate sbf explicitly ordered his devs to program a backdoor that allowed his personal crypto trading firm alameda research to go into nearly-unlimited debt trading on FTX (and thus eating into the deposits put down by legitimate paying customers) and that he had alameda co-CEO (and unrequited lover) caroline ellison cook the books and make seven alternative doctored statements of alameda’s financial position to show to potential investors and lenders when they were in dire straits (sam and caroline have this wonderful relationship which included passing each other bullet-pointed memos of pros and cons of continuing their relationship. lewis doesn’t hold back and recreates many of them in all their lurid details, a highlight is one where sam of course puts “I really like fucking you” under the list “ARGUMENTS IN FAVOR”). you have to wonder why they didn’t wait until the trial was over to release the book, i mean the final outcome of the trial could have made a fitting final chapter. maybe the publisher pushed it out now in order to capitalize on the media hype from the trial. or maybe lewis, in spending weeks shadowing his subject and swallowing more of sam’s honeyed words than anyone besides his inner circle, grew so enamored with him that he not only wrote a book continuing to extoll him and defend him in the light of a catastrophic fall from grace, wanted the book to be released right now because he believed it could assist in exonerating sbf or at the very least could shield him from public scorn. from what i’ve heard, even in the light of damning evidence being revealed at sbf’s trial, lewis has doubled down on defending him in the media during his book tour for the book.

so what exactly does lewis say in the book? for starters, there’s very little ominous foreshadowing of future collapse in the early parts of the book besides a few offhand mentions in footnotes, and the whole thing could probably stand as it is pre-collapse. maybe lewis was just lazy, and he wrote all of that early on and didn’t want to change it. or, it could’ve been on purpose. however, despite no explicit foreshadowing, astute readers can still spot some warning signs in lewis’ descriptions of sam’s reckless behavior early in his career. the most telling story is from the early days of alameda research, which he quit a safe and renumerative career at a major wall street firm to pursue. millions of dollars of a cryptocurrency called ripple went missing, and the management team wanted to pause trading until they found out what had happened. they were concerned that there may have been bugs in the automated trading software, a flaw in the strategy, or perhaps it had simply been stolen by hackers or an insider. but through the entire incident, sam was very nonchalant and sounded almost careless about the missing money. this was because he was convinced that it would turn up eventually (he gave it an “80% chance”), and thought it was silly to stop trading and lose time exploiting the temporary market opportunities they had identified. then, the crypto market soured and regular trading started losing money, there was a huge argument among management and sam wouldn’t relent, and in the end his entire management team and half his investor money pulled out. it sounds very similar to what would happen later down the line with alameda research and FTX, but in this case there was a “happy” ending: without anyone to hold him back, sam’s gets to fully unleash his automated high-frequency trading system and starts making tons of money, and they even recover the missing crypto (it was a mixup with the exchange they had been sending some to). “He’d been right all along!” triumphantly declares the narration. moral of the story: his actions may look reckless and messy, but if you just give him time and Let Him Cook, sam comes through because he’s a certified Lewis Whiz Kid.

an episode that suspiciously isn’t mentioned in going infinite is lewis’ fawning onstage interview of sbf at a big crypto conference, which is covered in one of the other books i read, zeke faux’s number go up. faux writes:

Lewis looked like a prep school headmaster, wearing a blue blazer with peak lapels and a white button-down with blue accents, his floppy hair parted perfectly to the side. And as he started lavishing Bankman-Fried with praise, he sounded as if he was presenting a prize to his star pupil.

“Three years ago, nobody knew who you were. And now you’re sitting on the cover of magazines. And you’re a gazillionaire. And your business is like one of the fastest growing businesses in the history of the planet,” Lewis said, to applause from the crowd and giggles from his subject. “You’re breaking land-speed records. And I don’t think people are really noticing what’s happened, just how dramatic the revolution has become.”

As Lewis went on, Bankman-Fried tapped the toes of his silver New Balance sneakers, sometimes pressing his legs with his elbows as if to hold them still. It seemed like Lewis saw him as another one of the truth-telling, sytem-disrupting outsiders he liked to write about. But the author’s questions were so fawning, they seemed inappropriate for a journalist. Listening from the packed auditorium, I started to question whether Lewis was really writing a book, or if FTX had paid him to appear (Lewis later told me he had in fact come to report for his book and that he was not compensated.)

Lewis said he knew next to nothing about cryptocurrency. But he seemed quite confident that it was great. The writer said that, contrary to popular opinion, crypto was not well suited for crime. He posited that U.S. regulators were hostile to the industry because they’d been brainwashed or bought off by established Wall Street banks. I wondered if he simply hadn’t heard about the countless crypto scams, but the thought seemed preposterous. “You look at the existing financial system, then you look at what’s been built outside the existing financial system by crypto, and the crypto version is better,” Lewis said.

Not A Good Look Mr. Lewis, appearing in public visibly intoxicated on the Sam Juice.

as far as i can tell, lewis’ thesis is that ftx was never insolvent and that all the ftx money is still there somewhere. you might even call lewis an “FTX truther”. the money is just misplaced and waiting to be found in sam’s disorganized mess of random investments, shell corporations, bank accounts, and so on, but sam could find it if you just give him time. unfortunately, sam was pressured into declaring bankruptcy and signing away control against his will by his lily-livered coworkers and conniving lawyers looking for a huge bankruptcy management payday. at one point lewis does some basic napkin math and puts together a list of “Sam World” (combination of ftx, alameda, and all sam-affiliated companies) assets and liabilities to the best of his knowledge, and much like caroline’s cooked books, they come out 4 billion ahead. it’s important to remember that lewis is basing nearly all of his conclusions based off of what sam personally communicated to him.

lewis introduces john ray (the interim CEO brought in to clean up the bankruptcy mess) fairly favorably, he comes off as the hard-boiled detective coaxed out of retirement for one last job. he’s crude and pragmatic, with no time for bullshit or wiseguys. but almost right away, lewis turns on him and smears him as an out-of-touch boomer who doesn’t understand anything about how FTX works and who pigheadedly refuses to ask for help or communicate whatsoever with the only guy who does (sam). apparently ray took one look at a picture of sam and concluded “that boy just ain’t right”, deciding that there was no point working with him because he was a “crook”. in retrospect, his intuition seems to have been spot-on. he also has choice things to say about the other members of the inner circle:

on nishad singh -

He’s narrow, it’s tech, tech, tech. There’s never a problem he can’t solve. He’s not going to steal money. He’s not going to do anything wrong. But he has no idea what’s going on around him. You ask him for a steak and he puts his head up the bull’s ass.

on caroline ellison -

She’s as cold as ice. You have to buy words by the vowel. An obvious complete fucking weirdo.

as ray’s investigation proceeds, lewis characterizes it like so:

Ray was inching toward an answer to the question I’d been asking from the day of the collapse: Where did all the money go? The answer was: nowhere. It was still there.

then, there’s the final paragraph of the book, the note that lewis chooses to end the book on:

I’d come back after they’d all gone, to look for something. After searching the jungle huts I was prepared to agree with others that it, too, had been no more than a figment of the imagination. But there was one last place to check, an old storage unit that no one had bothered to go through. The facility was just off the road that Sam drove every day between the Albany resort and the jungle huts. To the naked eye, it wasn’t a place you’d store anything of value, just lines carved in the jungle by tired-looking buildings with corrugated metal fronts. But it was there that I found it. It was just inside one of the ten unmarked FTX sheds. The wooden create had been addressed to Ryan Salame. It must have been far too heavy to move farther into the unit, so they’d just dumped it right at the front. The tungsten cube.

the implication is clear: just like the missing cube, just like early alameda’s lost ripple, FTX’s missing customer funds are still there, waiting to be found.

the thing is, the facts are actually all there in the book, it’s just that lewis has completely misinterpreted them. if you read between the lines carefully and consult a few other sources, you can still extract the “real” story of SBF and FTX from going infinite, which goes a little something like this:

SBF was born to two berkeley professors (one of them a professor of ethics, you can’t make this stuff up), as a child he was a loner with a melancholic disposition (literary term for “depression”). this seemed to be mainly because he was so precocious, already an atheist when all his peers were still worshipping at the altar of santa. in some ways he believes himself to be ahead of many adults as well, one major epiphany around the tender age of ten is when he discovers that many of them unironically believe in God. school is extremely boring for him but he gets perfect grades anyway, enough to go off to MIT. really the only thing he’s interested in is games or puzzles. he gets really into playing magic: the gathering and through it makes his only childhood friend, with whom he doesn’t do much except play magic (they end up getting good enough to go to nationals). it’s never mentioned explicitly but to me “autism” seems written between every line.

at MIT, sam is just another aimless and dispassionate physics student until he happens to fall in with effective altruism. effective altruism is one of the modern replacements for the Grand Narrative and/or God tailored towards sensitive nerds. it is an evolved form of utilitarianism (already one of the most autistic ethical theories) that encourages its adherents to earn as much money as possible so that they can donate as much as they can to charity. the “effective” part stresses that that money should be used as efficiently as possible, saving the most lives at the least cost. although it positively reeks of optimalism, i can’t find much fault with it in theory (most charities are notoriously inefficient), however in practice it seems to be a great way to rationalize or justify runaway greed. also, of late the nerds seemed to have corrupted things by redirecting the vast majority of effective altruism money away from simple and concrete (but unsexy) initiatives like malaria nets and into nebulous organizations doing “research” and “advocacy” to help “prevent” various hypothetical sci-fi apocalypses (mostly AI). anyways, i think the appeal of effective altruism to sam and friends is that it’s a way to frame life as a sort of feel-good game with a measurable goal, scored in “dollars donated” or “lives saved”.

as a result of becoming an effective altruist, when he’s getting ready to graduate sam decides to interview at several wall street trading firms, who are always eager to snap up super smart math/physics grads for their complicated algorithmic trading strategies. jane street capital, where he eventually ends up working, seems to be a perfect fit for him from the start, since their interview process consists entirely of a whirlwind assortment of games and puzzles. he is their top recruit that year and ends up doing very well in the firm, since it turns out that quantitative trading is pretty much just a game or puzzle you play with real money and a rulebook consisting of US financial law and contract terms. the only trouble sam has is getting along with his coworkers, a situation he partially improves over time by agonizingly teaching himself to project appropriate facial expressions and avoid misunderstandings (remember what i said about autism?).

after his third year, he’s doing so well that he’s about to pull down a bonus of $1 million for the year, and the head honchos tell him in his annual review that in ten more years he could be looking at between $15 million to $75 million a year. but he’s still unhappy, tormented by the idea that this may not be the optimal path for him, that he could be doing way more good (making more money) doing something else. finally, he spots the opportunity he’s been looking for: the nascent crypto scene. as an MIT-educated wall street trading genius, he’s convinced that he can effectively pubstomp the emerging crypto scene, filled with midwits and grifters and still way too sketchy for respectable professionals like jane street to touch. with the modest conviction that it was going to make him a billionaire, he leaves jane street and new york to found a crypto trading firm in san francisco with a bunch of fellow effective altruists. this is where the story really begins, a third of the way through the book.

flash-forward a couple years later and sam’s on top of the world, running FTX/alameda out of the bahamas with a bunch of hand-picked effective altruist nerd friends. sam does so well because he acts the part and presents a good story to respectable investors. he’s not like all the other conmen and grifters in crypto, he’s the MIT whiz kid who worked on wall street and is now bring his smarts to crypto, working to make it legit and even courting US regulators. to many, sbf looks like the long-awaited Adult in the Room for crypto, and they finally drop their guard enough to get in on it. turns out, on the inside it was all bullshit, complete disorganization and disarray behind the scenes at FTX.

like many villains, i don’t think sam ever thought he was the bad guy. unlike many crypto guys, i don’t think sam every intended to commit any sort of fraud or theft. alameda research got into a bad spot somehow, maybe it was due to a big bet that went bad (like the luna crash of early 2022 that vaporized billions of dollars and tanked other crypto firms), or maybe they got hacked big time. either way, they just needed a little bit of money (several billion dollars) to make it through a couple of months, by which time they could have easily made it all back because they’re very smart people. luckily, sam had easy access to a huge pile of money that people had trusted him with, the customer deposits at ftx. as long as everybody didn’t ask for their money back all at once, there’d be no harm no foul, it was strictly a temporary situation until they got back on their feet. of course, crypto, like banking, is a business that’s mostly built on confidence, so this also requires projecting an image of strength and not letting anybody outside the inner circle know that anything is wrong, which may involve a tiny bit of lying to investors, lenders, and customers, an act known to law enforcement as “fraud”. the “fake it till you make it” approach.

maybe if they had enough time it would have all worked out, and sam would have come out ahead like he had during the crisis in the early days of alameda. unfortunately, during this delicate time of temporarily-borrowing-customer-funds-without-permission, reckless sam trying to give off an aura of strength poked the wrong bear: rival exchange operator CZ. he made a snarky tweet about CZ’s reluctance (or perhaps inability) to step foot on US soil due to the legally-dubious nature of CZ’s exchange. but sam seemed to have forgotten that CZ held a major card in his pocket: approximately $500 million worth of FTT, a crypto “token” that was functionally equivalent to stock in FTX. CZ fired back where it hurt, and announced to his millions of followers that after “recent revelations” (a leaked document, purportedly listing alameda’s assets and liabilities that revealed it was heavily reliant on its own pile of FTT), he was going to be liquidating his position in FTT. as the price of FTT dipped, so did confidence in FTX, and the worst-case scenario occurred; everyone came asking for their money back at once. but they didn’t have enough of it left to pay everyone back because it had been loaned to alameda, who seemed to have (temporarily?) lost it continuing their underperforming trading strategies. and so FTX was forced to halt withdrawals and declare bankruptcy. oops! sam tried to cook but it was too late, all he had to work with was a slapped-together informal balance sheet full of, as matt levine characterized it, “howling ghosts”.

epilogue: sam bankman-fried was sentenced to 25 years in prison on march 28, 2024, after a disastrous trial performance, though eventually 100% of the money lost in FTX was recovered thanks to a rebound in the crypto market and sbf’s lucky early investment in AI firm anthropic.