how ea is portrayed in "going infinite"

Michael Lewis's book about Sam Bankman-Fried is annoyingly patronizing about effective altruism

Oct 06, 2023

Pinocchio is narrated by Jiminy Cricket; the Great Gatsby is narrated by Nick; Sam Bankman-Fried’s story is narrated by Michael Lewis. The book isn’t written from Sam’s perspective, exactly. Michael Lewis is a sympathetic fly on the wall documenting a tragically flawed underdog being destroyed by his own infinite ambition.

But like the Great Gatsby, where Nick speaks so fondly about Daisy (who Gatsby loves) and so unkindly about Tom (who Gatsby hates), you get the sense that Sam’s love for the effective altruism movement colours a lot of what ended up in the book.

Spectacle is another thing that colours what went into the book. Michael Lewis doesn’t take the tabloid road and make this a book about drugs and shrimp rings and “the polycule”, thankfully. But I can’t shake the feeling that most of the characters in the book are basically zoo animals on display for the public’s amusement.

He’s constantly saying, implicitly or explicitly, that effective altruists are high iq children who don’t trust emotions or adults. It feels condescending: “Of course you failed. You’re children! Idiot savants, like something out of a Dostoevsky novel.”

But he also seems to really believe EAs are genuine, and that’s not nothing.

The first mention of effective altruism is in the preface. It doesn’t say effective altruism, exactly, but

He [Sam] needed infinity dollars because he planned to address the biggest existential risks to life on earth: nuclear war, pandemics far more deadly than Covid, artificial intelligence that turned on mankind and wiped us out, and so on.

And that really sets the tone for the rest of the book. At least according to his book, Michael Lewis believes that Sam, Caroline, Gary and Nishad were motivated to earn a lot of money because they wanted to give it to charity.

Chapters 1 and 2 introduce the character of Sam, including the way he thinks about expected value and the way he taught himself to make facial expressions, and Chapter 3 properly introduces effective altruism.

One other oddly big thing happened to Sam at the beginning of his junior year. Completely out of the blue, a twenty-five-year-old lecturer in philosophy at Oxford University named Will Crouch [Will MacAskill’s unmarried name] reached out and asked to meet with him. Sam never learned how the guy had found him—probably from the writing Sam had been doing on various utilitarian message boards.

He devotes several pages to talking about Peter Singer, Toby Ord and Will MacAskill, and the early version of 80,000 Hours Will was promoting on his visit to Harvard. As far as I can tell, Lewis basically gets those things right. And then he included a quote from Will that completely blew my mind:

“The demographics of who this appeals to are the demographics of a physics PhD program,” he said. “The levels of autism ten times the average. Lots of people on the spectrum.”

…yeah.

Moving on: the end of Chapter 3 mentions allegedly the first earning-to-giver on Wall Street, Matt Wage (nominative determinism strikes again). Chapter 4 says that Sam became a vegan and an effective altruism because he thought through the arguments and concluded they were correct, not because of feelings of guilt or empathy. And Chapter 5 gets into the Alameda origin story, which is where EA really gets involved.

The business hadn’t even really been Sam’s idea but Tara’s. Tara had been running the Centre for Effective Altruism, in Berkeley, and Sam, while at Jane Street, had become one of her biggest donors. … Her success [trading crypto] led Sam to his secret belief that he might make a billion dollars by creating a hedge fund to trade crypto the way Jane Street traded everything.
But he couldn’t do it by himself. Crypto trading never closed. Just to have two people awake twenty-four hours a day, seven days a week, he’d need to hire at least five other traders. He’d also need programmers to turn the traders’ insights into code, so that their trading could be automated and speeded up. Tara had been making a handful of trades a week on her laptop; what Sam had in mind was an army of bots making a million trades a day. He’d need to hire some lower-IQ people to do the boring stuff, like finding office space and getting food for the traders and paying utility bills and probably lots of other things he hadn’t thought of.
His access to a pool of willing effective altruists was his secret weapon. Sam knew next to nothing about crypto, but he did know how easy it was to steal it. Anyone who started a crypto trading firm would need to trust his employees deeply, as any employee could hit a button and wire the crypto to a personal account without anyone else ever having the first idea what had happened. Wall Street firms were not capable of generating that level of trust, but EA was.

Lewis touches on how Alameda started out just sending money to random EAs in specific countries and getting them to transfer it back as a way of exploiting arbitrage opportunities. It’s absolutely wild.

Caroline, Gary and Nishad join Alameda in this chapter - Lewis briefly touches on each of their EA origin stories - and wealthy EAs including Jaan Tallinn invest in the young hedge fund. (No friends and family discount on the interest rate though - “investors were charging them a rate of interest of 50 percent” according to Lewis.)

Lewis does cover half of the EA founding team leaving Alameda as well, and to his credit I think he did interview some of the people who left, although it still comes off as very Sam-friendly compared to the accounts I’ve heard.

“The prospect of losing a couple of hundred million dollars that would have otherwise gone into solving the world’s problems felt pretty high stakes,” said Ben West. Under the circumstances, they thought it was insane to continue trading, but Sam insisted on trading. Crypto markets would not remain inefficient for long. They needed to make hay while the sun shone.

Lewis says the other managers were trying to figure out how to get rid of Sam; “aimed to bankrupt Sam, almost as a service to humanity”; and “told our investors he was faking being an EA, because it was the meanest thing they could think to say”.

In the end, for Sam to leave he had to want to leave, and Sam did not really want to leave. And so, on April 9, 2018, his entire management team, along with half of his employees, walked out the door, with somewhere between one and two million dollars in severance.

People in the wider EA community started hearing things, but…

“There was no smoking gun.” No one thing Sam had done for which they could easily condemn him. It was, as Tara said, “one hundred small things.”

And after a while when Alameda and then FTX started doing so well, some of them even started to think that maybe they were wrong, maybe Sam was just rude and stubborn and untrustworthy in a way that makes you good at business, not in a way that means you should cut all ties.

In Chapter 6, Lewis shares Sam’s writing about the situation:

“I did damage to the EA community,” he wrote. “I made people hate each other a little more and trust each other a little less … and I severely curtailed my own future ability to do good. I’m pretty sure my net impact on the world has, so far, been negative and that is why.”

At that point, Sam stopped hiring so many EAs.

Chapter 7 focuses on George Lerner, a psychiatrist who “was treating maybe twenty EAs” when he was invited to move to the Bahamas and join the FTX staff. He says really mean things about Caroline particularly. And he comments a lot on the trends in his EA patients:

“A lot of EAs chose not to have kids,” said George. “It’s because of the impact on their own lives. They believe that having kids takes over from their ability to have impact on the world.”
“There are two parts of being EA,” said George. “Part one is the focus on consequences. Part two is the personal sacrifice.”

And trends in the EA staff in the Bahamas:

“Everyone is complaining about the lack of dating opportunities,” said George. “Except the EAs. The EAs didn’t care.”
The non-EAs thought the EAs thought they were smarter than everybody else.

*shrug*

Chapter 8 introduces a new EA character - Igor Kurganov. “Kurganov was a Russian-born former professional poker player to whom Musk had entrusted the task, it was reported, of giving away more than $ 5 billion worth of his fortune. He was also a self-described effective altruist”.

Lewis describes Sam as spending more of his money now:

considering buying some of Twitter with Elon Musk, in a bid to convince him to donate more of his money towards EA-style causes
investing $500 million in Anthropic
going to EA leaders’ retreats where they discuss how to spend his money
and trying to buy political power

In a very short time, Sam’s money had bankrolled some of the most spectacular failures in the history of political manipulation. Carrick Flynn, for example. When Sam stumbled upon him, Carrick Flynn was a newcomer to elective politics. He was the quintessential Washington, DC, policy wonk—one of the faceless minions in blue suits who sit along the wall behind the more important people and occasionally rise and whisper something in their ears. Carrick Flynn’s most important trait, in Sam’s view, was his total command of and commitment to pandemic prevention. His second-most important trait was that he was an effective altruist.
[…]
Flynn asked some fellow EAs what they thought about him running for Congress. As a political candidate he had obvious weaknesses: in addition to being a Washington insider and a bit of a carpetbagger, he was terrified of public speaking and sensitive to criticism. He described himself as “very introverted.” And yet none of the EAs could see any good reason for him not to go for it—and so he’d thrown his hat into the ring. Somewhere along the EA trail he’d become known to Sam.
[…]
The people of Oregon not only did not appreciate the effort; a lot of them started to kind of hate Carrick Flynn. And Carrick Flynn was not designed to ignore their feelings. Attacked by other candidates during a debate, he simply walked out in the middle.
[…]
Anyway, to Sam, the money he’d spent on Carrick Flynn had been a drop in his second bucket. Other congressional races had worked out better.

Chapter 8 also introduces the FTX Future Fund:

From the moneymaking division of FTX, it was just her, Sam, Gary, and Nishad; from the money-giving side were the four employees who worked for FTX’s philanthropic wing. They shared their employer’s habit of turning life decisions into expected value calculations, and their inner math yielded similarly surprising results. In 2020, Avital Balwit had won a Rhodes Scholarship and turned it down, first to run Carrick Flynn’s congressional campaign and then to give away FTX’s money. Leopold Aschenbrenner, who had entered Columbia University at the age of fifteen and graduated four years later as class valedictorian, had just declined a spot at Yale Law School to work for this new philanthropy. Their boss, a former Oxford philosopher named Nick Beckstead, was also present, as was their spiritual guru, Will MacAskill—who was of course in some way responsible for everyone, including Sam, even being there.

Lewis introduces the idea of longtermism here, and tries to shoehorn it into his theme of high iq kids with no use for feelings who follow the numbers.

One day some historian of effective altruism will marvel at how easily it transformed itself. It turned its back on living people without bloodshed or even, really, much shouting. You might think that people who had sacrificed fame and fortune to save poor children in Africa would rebel at the idea of moving on from poor children in Africa to future children in another galaxy. They didn’t, not really—which tells you something about the role of ordinary human feeling in the movement. It didn’t matter. What mattered was the math. Effective altruism never got its emotional charge from the places that charged ordinary philanthropy. It was always fueled by a cool lust for the most logical way to lead a good life.

And for this one sentence, I remember how exciting things were very briefly:

After handing out $ 30 million in 2021, they were on pace to hand out $ 300 million in 2022, and then $ 1 billion in 2023. As Nishad had put it to me not long before, “We’re finally going to stop talking about doing good and start doing it.”

And then the rest of the book is about the collapse of FTX. I won’t get into it; I expect the trial over the next six weeks will cover this much better than the book could.

It’s weird - whenever Lewis wants to talk about Sam, Caroline, Nishad and Gary, he calls them “the effective altruists”. And I guess that is how they met and a way to distinguish them from other key players in FTX! But I found it jarring every time.

I’ve spent so much time thinking about how those people were naive utilitarians or possibly just bad actors, but either way criminals who make upstanding people like us look bad, that I forgot how relatable Nishad and Gary and Caroline and Sam are. I’m glad the book helped me to remember.