Part IThe Story
The coffee was supposed to be a quiet one. It was the last week of January 2026, the World Economic Forum was grinding through its annual ritual of cooperation in Davos, and Brian Armstrong, forty-three years old, bald, dressed in the black that has served as his uniform for a decade, was sitting with Tony Blair, a former British prime minister. Then Jamie Dimon walked over. The chairman of JPMorgan Chase, the largest bank in the United States and the man who once called bitcoin a fraud, did not take a seat. He pointed. "You're talking nonsense," he said, according to people the Wall Street Journal later spoke with. The message, as those people described it, was that Armstrong should stop going on television and saying untrue things about the banks.
Earlier that week Armstrong had made a circuit of business-news programs accusing the banking industry of sabotaging a bill in the U.S. Senate, the Digital Asset Market Clarity Act, which was meant to settle which federal regulator oversees which crypto assets. The specific fight was narrow and almost comically technical: whether an exchange like Coinbase could pay holders of dollar-pegged stablecoins a reward of around 3.5 percent while the average checking account paid less than a tenth of one percent. The banks called it deposit flight. Armstrong called it competition. A few months earlier, in front of the Senate Banking Committee, he had said it more plainly: "Many Americans are earning about 0.14% in their savings account, but in Treasurys, you can earn 4.5% right now. That seems unfair to me."
The scene is worth holding still for a moment, because everything about it is inverted. The man doing the scolding ran the institution that was supposed to be permanent. The man being scolded had founded his company fourteen years earlier with a $150,000 check from Y Combinator, a two-bedroom apartment, and a cryptocurrency then trading, by his co-founder's later recollection, at around $6. He had written in his twenties that he wanted to "start breaking free." He had spent a year as a young man in a country where the banks locked their doors. And now he was at Davos, the most establishment room on earth, being treated by the head of the biggest bank in America not as a curiosity but as a rival. Dimon does not cross a room for curiosities.
Armstrong did not, by any account, raise his voice. He is not a voice-raiser. He is, in his own description, "a little bit of an introvert," a "conflict-averse person by nature," a man whose favorite place is a room with engineers and a whiteboard. He had once confessed to Tim Ferriss that the chief-executive job was "just not that fun, to be honest." Nine months after Davos, the bill he had alternately blocked and championed would die on the Senate floor by a single vote, and he would find himself on X, preemptively defending himself against a newspaper story that had not yet been published.
So: an introvert who became crypto's chief lobbyist. A libertarian who built the most compliant exchange in his industry. A man who preaches exit and has never once left. The coffee at Davos is where all of that becomes visible at once.
By the Numbers
The Coinbase Ledger
$7.18BCoinbase revenue in 2025, with $1.26B in net income
19%Armstrong's ownership stake in the company he co-founded in June 2012
$516BDigital assets held on the platform as of October 2025
~12%Share of all bitcoin in existence held in Coinbase custody
108MCustomers on the largest U.S.-based crypto exchange
$328.28Closing share price on April 14, 2021, the first day of trading
$49MCoinbase's 2024 contribution to the pro-crypto super PAC Fairshake
A Country Where the Doors Were Chained
He was twenty-seven or twenty-eight—he isn't sure which—when he flew to Buenos Aires with no place to stay. The plan was a year-long South American circuit, one city a month, "a vision quest to figure out what I wanted to do with my life." He spoke high-school Spanish. He found a hostel, then a rented room, figured out the ATM, bought a local phone. After a month of that he could not face doing it again in Brazil or Chile, and so he simply stayed. About a year, with a few trips across borders.
"I can tell you at the time I felt very lost," he told Ferriss in October 2022. It is one of the more honest things a billionaire has said on a podcast. He had a tutoring company, UniversityTutor.com, that he had co-founded in 2003 as a Rice undergraduate and that was by then "kind of limping along, barely successful." He had tried consulting in the enterprise risk management division of Deloitte & Touche and did not feel successful there either. He had done an internship at IBM in San Jose, near where he grew up, the son of two engineers, a self-described shy kid who had gone through Bellarmine College Preparatory, a Jesuit school whose motto he later recalled as "men for others."
What he noticed in Argentina was not, at first, the inflation. It was the mood. "There wasn't optimism about the future," he said. "People kind of had this vibe, keep your head down, don't do anything too crazy because everything could get taken away from you tomorrow." Restaurants operated in a gray zone; the fire inspector came by not for the fire code but for the envelope. Argentina, he had read, had been something like the eighth-largest economy in the world around 1900, and then came "basically a hundred years of just mismanagement, corruption, bad government, endless protests." He drew from it a lesson that would later sound like ideology but at the time felt more like an observation about weather: "the people misplaced their faith in the government to do good things."
Ferriss, who had lived there in 2004 and 2005, supplied the image Armstrong did not. Walking around Buenos Aires three years after the 2001 meltdown, Ferriss said, he saw graffiti all over the banks—"vultures," "thieves"—and some of them "still had chains locking the doors shut from when people were prevented from withdrawing their cash savings." The doors had been chained so that depositors could not get their own money out.
Keep the chained doors in mind.
The other artifact from this period is stranger and more revealing. Right out of college, Armstrong wrote a book. It was called Start Breaking Free, self-published, and since he had no publisher he read a book called How to Market Your Book, which told him to start a blog, so he started StartBreakingFree.com. When Ferriss dug it up, Armstrong was genuinely startled—"That's impressive you found that"—and then explained, with an engineer's deadpan, that he had done things in the wrong order. "Most people would probably go do that first and then write the book. But I somehow got it backwards and was like, 'You know what? Let me write the book to figure out how to do it.'"
Writing to figure out how to do it. This is the oldest habit he has, and he has never abandoned it. When he later had to decide whether to tell his company it would no longer engage in political activism, he wrote "two or three or four drafts" and threw them away. "When I'm trying to formulate my thinking," he said, "I try to write." Every morning, he has told Forbes Australia, he writes down three things that, if they were the only things he accomplished, would move the company's goals forward, and he does them before he opens his email.
A man who thinks on paper. A man who, at the age when his peers were building résumés, wrote a book about escape before he had escaped anything.
The Man Who Didn't Have Any Bitcoin
In December 2010, he read a nine-page paper by a pseudonymous author named Satoshi Nakamoto. He was, by then, back in California and looking for a job in Silicon Valley; within months he would join Airbnb, where he worked on fraud prevention for a company moving money into 190 countries and, by his own count, was employee number forty or so. "It definitely captured my imagination when I read it," he recalled at a Bloomberg event in August 2018, "because it talked about how the whole world could have this universal currency that ran on the internet... I felt like this is the most important thing I have read in like five years."
He started going to bitcoin meetups in San Francisco, and his description of them is one of the few genuinely funny lines in the Armstrong archive: "the room was half brilliant computer scientist people and half completely crazy people." He also remembers thinking he was already too late. There were exchanges. There was Mt. Gox. It was 2011.
The first thing he built was wrong, and he knew it the minute he launched it. With a friend, on nights and weekends, he wrote an Android app—"just called Bitcoin Wallet or something"—that ran an entire bitcoin node on the phone. Wired wrote a small item about it, which thrilled him. Then he watched people use it. Every time you opened the app it said Syncing and sat there for three or four minutes catching up to the blockchain. "I knew it instantly we built the wrong thing," he said. What the world needed, he understood in that moment, was a company that ran all that infrastructure in the cloud, with backups and security. "And I remember thinking at that time, 'Well, that's not going to be me because I have a day job.'"
He would later file this under one of his favorite Paul Graham–isms: action produces information. He loves that phrase. He has said he would put it on a billboard. The point of launching the wrong app was not that it was wrong; it was that being wrong in public was faster than being right in private.
The hosted wallet got him into Y Combinator's summer 2012 batch. One early, almost fatal complication: Ben Reeves, a British programmer who had co-founded Blockchain.info and was supposed to be part of the founding team, parted ways with him just before the YC money came through, over how the wallet should work—whether users or the company should hold the keys, the argument that has shadowed Coinbase ever since. Reeves went his own way and built a wallet in which users controlled their own coins. Armstrong went to YC alone, with the slogan he would pitch on demo day: "Coinbase, the easiest way to get started with Bitcoin."
Fred Ehrsam came in through Reddit. He was a trader at Goldman Sachs who had noticed Armstrong's posts and reached out—the rare Wall Street person in those rooms who was neither crazy nor a computer scientist, a man who understood order books and settlement and the dull machinery of moving money. He would co-found Coinbase, leave it six years later, and in 2018 walk into Sequoia Capital to pitch a partner named Matt Huang on starting Paradigm, now one of crypto's most respected investment firms. Ehrsam is the first of many people in this story who leave. He left well.
The company had launched, and nothing was happening. "We had no traction, really. Things were flat." A couple hundred people signed up from Reddit posts and never came back. So Armstrong did the thing YC drills into every founder: he emailed five of the users and gave them his cell number. On the third or fourth call, a man said the app seemed cool, "but I don't really have any Bitcoin. So I just signed up, and then I didn't have any Bitcoin, and I didn't come back."
Armstrong asked whether he would have used a Buy button, right there in the app.
"Yeah, probably," the man said. Buying bitcoin otherwise meant wiring money to Japan, to Mt. Gox. "It was too much of a hassle, so I just didn't do it."
The button was simple. Building it was not. Armstrong had to find a bank willing to let a bitcoin startup touch the ACH network, convince that bank of an anti-money-laundering policy he had written, code the integration, and figure out where the bitcoin would actually come from once someone clicked. In October 2012 Coinbase launched the ability to buy and sell bitcoin by bank transfer—the first time, as he tells it, that an American could connect a bank account and simply buy. By February 2013 VentureBeat was reporting more than $1 million in bitcoin sales a month. "That's the moment we had product market fit," he said. "It started to grow organically every week after that. And it was like a boulder we were chasing downhill."
Notice where the founding insight came from: a customer explaining that the product was useless because he was missing the one thing it required. The entire company—everything that followed, the custody, the licenses, the lobbying, the Davos coffee—sits on top of that phone call. A man who did not have any bitcoin, and a button.
— Brian Armstrong, The Tim Ferriss Show, October 2022If you just take some steps into the unknown, you might not be doing the right thing, but oftentimes by doing that action, you'll realize what you should have been doing instead.
Half Brilliant, Half Crazy
Hiring was nearly impossible. "Somebody has to be a little crazy to join something so early stage where it's like, okay, it's two people with a laptop and a dream," he told Guy Raz on How I Built This. His approach was not charm—he has none of the performed kind—but volume. "If I went to a networking event in Silicon Valley, I'd walk up to every single person. Are you interested in crypto? I'm doing this company... Can I get your email? You know, no, no, no, no. I'd get one yes after 10. I'd go do 30 more."
The first employee found them instead. Olaf Carlson-Wee, a Vassar graduate, cold-emailed the founders in 2013 and was paid in bitcoin for three years—a choice that, depending on which year you checked, was either a hardship or the best compensation package in the history of Silicon Valley. He, too, would leave, to start his own crypto fund. The pattern was setting: Coinbase as a launch pad, a place people came to learn the strange new trade and then exited from, carrying what they had learned into the wider industry. A 2021 Bloomberg piece called them the "Coinbase mafia."
The money came from people who had to be argued into it. In May 2013, Fred Wilson of Union Square Ventures led a $5 million Series A—at the time, the Wall Street Journal reported, the biggest funding round for any bitcoin startup. Wilson would sit on the board for years and, by Armstrong's account, guide him through a kind of Socratic method rather than instruction—asking the question until Armstrong arrived at the answer himself. In December 2013, Andreessen Horowitz led a $25 million round alongside USV and Ribbit Capital. In January 2015, $75 million came in, led by Draper Fisher Jurvetson, with the New York Stock Exchange, USAA, and several banks participating. Banks. The thing bitcoin was supposedly built to replace was buying a stake in the company that made bitcoin easy to buy.
This was the cleft that Armstrong never pretended to straddle comfortably, and it opened early. The meetups were half crazy people, and a good share of those crazy people were cypherpunks for whom the point of bitcoin was that no one needed permission—no bank, no government, no know-your-customer form. Coinbase, from its first bank partnership, was a know-your-customer company. Its users had to provide taxpayer identification. It would never list Monero or other privacy coins. In November 2017 the IRS ordered it to hand over records on customers who had transacted at least $20,000 in a year; in February 2018, Coinbase told roughly 13,000 of them that their names, birth dates, addresses, and transaction histories from 2013 to 2015 would go to the government within 21 days.
To the purists, this was apostasy. To Armstrong it was the price of the thing he actually wanted, which was not a revolution but adoption. He did not approach bitcoin, as one later analysis put it, as a political project but as a technology that needed institutions to thrive. And yet he was not exactly a pragmatist either. He read Milton Friedman, recommended Friedman's PBS series Free to Choose, called it "kind of Libertarian, so some people will like that, some people won't." He listed Atlas Shrugged and The Sovereign Individual among his recommended books. He carried Buenos Aires around in his head. He believed in the escape. He just thought you got there by filling out the forms.
That contradiction—the escape artist who files every form in triplicate—never resolved. It simply grew larger.
A Bank With No Sign on the Door
In 2018, a Fortune reporter went looking for Coinbase's headquarters and had trouble finding it. Three floors high above Market Street in San Francisco, sweeping views of the bay, jars of chocolate coins at reception beside a sign reading "Initial Chocolate Offering." But no logo outside the building. None in the lobby. In the hallway outside reception: "fortified metal doors and an intercom." The reason, staffers explained, was that most of them owned cryptocurrency, some in amounts that made them paper multimillionaires, and a kidnapper could capture someone and "pull out their fingernails" to learn where the keys were.
The bank in Buenos Aires had chained its doors to keep its customers out. The bank that wanted to replace it had armored its doors to keep its employees alive.
The boom of 2017 had made the company enormous almost overnight. That August it raised $100 million at a $1.6 billion valuation, becoming crypto's first unicorn; Armstrong, Fortune noted, then owned more ether than bitcoin. By 2018 Coinbase claimed 25 million customer accounts, five times the number two years earlier, putting it on a par with Charles Schwab. It had obtained New York's notoriously demanding BitLicense in January 2017. And then it had broken under the load.
On December 19, 2017, Coinbase listed Bitcoin Cash, and prices on the platform went haywire, prompting an insider-trading investigation. In January 2018, consumer complaints filed against the company with the Consumer Financial Protection Bureau more than doubled to 889 in a single month, more than 400 of them categorized as "money was not available when promised." From 2018 to 2023 the Better Business Bureau rated Coinbase an F. The common complaint was that customers got locked out of their accounts and found it "nearly impossible" to reach a human being. The friction-free button now had, behind it, a backlog. In a later telling to the podcaster Shawn Ryan, Armstrong recalled the early version of this: 25,000 sign-ups in a single day, 20,000 support tickets piling up, and a hacker who nearly drained the company's funds while the founders were out at lunch.
He began, in this period, to hire adults. In March 2018, Emilie Choi arrived from LinkedIn, where she had run mergers and acquisitions, to lead corporate development. She was promoted to president and chief operating officer in May 2019 and became, in Armstrong's description, one of the very few people licensed to speak to him without deference. "She can just tell me, 'Brian, you're being an idiot. Don't do that,'" he said. He put her in the same category as his family, his board, and Ehrsam—people "that have your best interests at heart and can tell you when you're being an idiot." It is a short list. He guards it.
The other big hire of that year was more exotic. In April 2018, Coinbase acquired Earn.com and made its chief executive, Balaji Srinivasan, its first chief technology officer. Srinivasan, a Stanford-trained engineer and former Andreessen Horowitz partner with an almost Talmudic appetite for theorizing, would leave after a year and go on to write The Network State, a book arguing that online communities could crowdfund territory and eventually become countries. Armstrong would later recommend it, appear at its conferences, and quote its central idea back as his own. Srinivasan, too, left. His ideas stayed.
The mistakes of this era had a characteristic shape: a fast move, a public stumble, a fast admission. In February 2019 Coinbase bought Neutrino, an Italian blockchain-intelligence firm, and it emerged that its founders had been connected to Hacking Team, which had sold surveillance tools to governments with abysmal human-rights records. Within two weeks, Armstrong wrote that the company had "did not properly evaluate" the deal and that the ex-Hacking Team staff "will transition out of Coinbase." No elaborate defense. A post, a reversal, a door opened for the people who had to leave.
That October, on stage at Vanity Fair's New Establishment Summit, he was asked about the early deals he had cut in cafés. "I won't deny that I have been in coffee shops," he said. He confirmed Coinbase had been profitable in 2017, 2018, and 2019 and had generated more operating profit than it had raised in venture capital. "I sort of think of us as the anti-unicorn unicorn," he said, and when asked whether an $8 billion valuation nonetheless made him a unicorn: "I leave that up to investors... I just like to build products with technology."
That last line is the one he believes. It is the one he keeps returning to whenever the job asks him to be something else.
Mission First, and the Exit
The walkout happened in June 2020. The company had sent everyone home in May—Coinbase was among the first large tech firms to announce it was going fully remote and would no longer recognize a formal headquarters, a decision that, the San Francisco Standard would later report, cost it some $25 million to unwind its San Francisco footprint. George Floyd had been murdered in Minneapolis. Armstrong, describing it later, said the biweekly all-hands meetings had already been drifting for a year or two toward questions that had nothing to do with crypto, "almost... this thing where who can make the exec team squirm the most on stage."
At one of those meetings, someone asked whether Coinbase would publicly support Black Lives Matter. Armstrong deferred. "That wasn't good enough. And they kind of held the mic and they pressed the issue," he said, "and this is where a bunch of people walked out of the company." In a later telling he put the number at around 300.
His reaction, in the executive meeting afterward, was bafflement—"What does Coinbase have to do with police brutality?"—and the bafflement is the most revealing thing about the episode. He did not, at first, understand the question as a question. He asked colleagues what BLM actually stood for. Equality, they told him. "I was like, great. Okay, if that's what people want to hear, I don't understand it, but let's go out there and say it." He tweeted support. People came back to work.
And then, over the following months, he felt "increasingly uneasy." He learned the movement was associated with positions, like defunding the police, on which he did not believe his company should have a view. He concluded that the company was split between people who thought it should engage broader social issues and people who thought crypto and economic freedom were already mission enough—and that he, as the leader, had failed by not saying which camp he was in.
So he wrote. Two, three, four drafts, discarded. Some colleagues told him, "do not share this under any circumstances." Someone told him that an underrepresented person would never work at Coinbase again. He went to the company's employee resource groups and asked what mattered most to them at work. "We just want to be respected at work, learn things, be able to build cool stuff," he says they told him. "It was basically what everybody else wanted."
What tipped him, in his own account, was not principle but exhaustion: "God, the CEO job is just not that fun, to be honest. If I have to come in every day and be put in front of the hot mic and have to answer all these crazy things that I don't have an answer to, that's not what I signed up for." He framed the choice with brutal symmetry. "Either I have to go... or they've got to go."
In September 2020, he published "Coinbase is a mission-focused company." It said the company would not engage in social activism unrelated to its mission. It cited the ways activism had consumed Google and Facebook. And it offered an exit: a generous severance package for anyone who disagreed. About 5 percent of the company took it.
— Brian Armstrong, The Tim Ferriss Show, October 2022It created a bunch of drama for a few months, a couple journalists wrote hit pieces on us and things like that, but afterwards it was better. It was one of the best things I've ever done for the company to be honest.
The fallout was real and did not stay confined to op-eds. In November and December 2020, Nathaniel Popper of the New York Times reported on Black employees' experiences at the company and on internal data, two years old by publication, showing women had been paid about 8 percent less than men in comparable roles and Black employees about 7 percent less. (The writer Aaron Renn later noted that the analysis made no adjustment for education or experience.) In May 2021, Coinbase's chief people officer announced the company was eliminating salary and equity negotiation entirely: "all employees in the same position, in the same location, receive the same salary and equity offer." Armstrong had refused to make his company a vehicle for politics and then quietly rebuilt its pay system around the grievance.
He says other CEOs called afterward to ask how to do what he had done. Whatever one thinks of the memo, it was an act of definition, and a particular kind—not here is what we stand for but here is what we will refuse to carry, and here is the door if you need it. The severance offer was the heart of it. Exit, offered politely, with a check.
The Anti-Unicorn Goes Public
On April 14, 2021, Coinbase listed on the Nasdaq. It did not do an initial public offering; it did a direct listing, the less-trodden route in which no new shares are sold and no bankers set a price over dinner. Nasdaq set a reference price of $250, valuing the company at around $47 billion. The stock closed the first day at $328.28. Bitcoin had been rising all winter; in the first quarter of 2021, Coinbase's revenue rose ninefold to $1.8 billion from $90.6 million a year earlier. For full-year 2021 it would report $7.8 billion in revenue and $3.6 billion in profit.
"I think over a thousand millionaires were created in that IPO," Armstrong said later, "just from various employees and investors who had worked on Coinbase over the last ten years. So that was incredibly fulfilling." He was thirty-eight. Bloomberg had estimated before the listing that he could be worth between $7 billion and $15 billion.
Then the boulder he had been chasing downhill turned around.
The year 2022 was an anthology of setbacks. In April, Coinbase entered India with great fanfare and relied on the country's Unified Payments Interface for rupee deposits; within days, the National Payments Corporation of India said it was "not aware of any crypto exchange using UPI," and Coinbase suspended most of its Indian business. An NFT marketplace, teased for months as a challenger to OpenSea, launched in May to a shrug. The company released a film trilogy about NFTs that one critic on Twitter summarized as "Cringebase." In June, days after posting a Twitter thread in which he told dissatisfied employees who were criticizing executives that they could quit, Armstrong cut 18 percent of the workforce, about 1,100 people. In the second quarter, Coinbase lost $1.1 billion, a company record. For the year, it lost $2.6 billion. In July, federal prosecutors charged Ishan Wahi, a former Coinbase product manager, in the first-ever crypto insider-trading case; after being summoned to a meeting at Coinbase's Seattle office, Wahi had bought a one-way ticket to India and was stopped before he could board his May 16 flight. He would be sentenced to two years in prison. His brother got ten months.
Fortune that October opened its profile on an image that turns out to be the key to Armstrong's public life. He was standing on an empty stage in front of a green screen, a makeup sponge patting his face, reading off a teleprompter for an investor video. "Coinbase is a company with an ambitious vision: to create more economic freedom," he read, and flubbed the last part. "We'll try that one again." The analyst Mark Palmer was quoted wondering aloud whether the founder who establishes the vision is the right person to execute it.
And then, in November 2022, FTX collapsed.
The shape of Armstrong's career makes the most sense when viewed against Sam Bankman-Fried's. Bankman-Fried had been everything Armstrong was not: a media darling, a ubiquitous presence in Washington, a man who talked fluently about effective altruism while running his exchange out of the Bahamas. Armstrong had spent ten years being called slow, square, and over-lawyered. When FTX imploded under what prosecutors would prove was fraud, Armstrong wrote an op-ed for CNBC arguing that the problem was not too little regulation but too little clarity—that U.S. regulators' "regulation by enforcement" had pushed more than 95 percent of crypto trading offshore, to places like the Bahamas, where a firm "could take dangerous risks with no repercussions."
He also, for the first time, started talking. "More recently I've been getting out there and doing lots more media," he told Ben Thompson of Stratechery that December, "just because I think in times of crisis for the industry, it's important for me to be more visible." Coinbase's stock had fallen 85 percent. "We've taken a very different approach over the last 10 years in terms of just leaning into compliance and security and trust, and it's been frustrating sometimes because it feels like we have to move a little bit slower. And following the rules is not always fast, but it's been a good validation of our strategy."
In January 2023 he cut another 950 jobs, citing the risk of "further contagion" and promising to shut down projects "where we have a lower probability of success." The same month, New York's financial regulator fined Coinbase $50 million and required another $50 million in compliance spending after finding the company had accumulated a backlog of more than 100,000 unreviewed alerts from its transaction-monitoring system. The careful company had not, it turned out, been careful enough. It was simply more careful than the ones that blew up.
By December 2023, CoinDesk was calling him "crypto's last big man standing." Bankman-Fried was headed to prison. Changpeng Zhao of Binance had stepped down in a plea deal. Armstrong had outlasted them all. The man who had never left was the only one left.
Suing the Referee
On March 22, 2023, Coinbase received a Wells notice from the Securities and Exchange Commission, the formal warning that enforcement action was coming. The company's response was not a settlement posture. Its chief legal officer, Paul Grewal, published a post titled "We asked the SEC for reasonable crypto rules for Americans. We got legal threats instead." In April, Coinbase did something few regulated companies would dare: it sued the SEC in federal court to force the agency to respond to a rulemaking petition Coinbase had filed the previous year. In June, the SEC sued Coinbase, alleging that it had operated as an unregistered exchange, broker, and clearing agency since 2019 and asking that it be "permanently restrained and enjoined." The stock dropped 14 percent before the market opened.
There had been a preview. In September 2021, the SEC had threatened to sue if Coinbase launched a crypto lending product called Lend, even though, as TechCrunch noted, similar products were already on the market. Coinbase had complained publicly of being singled out—and then canceled the product. Armstrong learned something from that retreat, and he did not retreat again.
His version of the story, told on podcast after podcast, has the clean lines of a Western. Coinbase met with the SEC some 30 times asking for rules and got none. The ambiguity over whether crypto assets were securities or commodities, he told the podcaster David Senra, "was really weaponized by Gary Gensler, the former SEC chair, and Elizabeth Warren, and some people like that who tried to, in my view, unlawfully kill the industry in the United States." Coinbase spent roughly $50 million on lawyers and refused to settle.
He did not only litigate. He organized. Coinbase seeded a grassroots advocacy group, Stand With Crypto, which by his count signed up some 2 million Americans, and poured money into Fairshake, the pro-crypto super PAC—$49 million in 2024 alone. A Public Citizen analysis found that crypto companies, which had given less than $10 million to super PACs across the two previous election cycles combined, had raised more than $200 million in 2024, nearly half of all corporate contributions that cycle. Coinbase accounted for close to a quarter of Fairshake's money. Fairshake spent heavily against Sherrod Brown, the Ohio Democrat who chaired the Senate Banking Committee, and Brown lost. Fortune estimated Armstrong had directed around $100 million, from his own fortune and the company's, into pressuring lawmakers.
The introvert who disliked media appearances began to appear on Capitol Hill. In June 2024 he swapped his black T-shirt for a slim black suit and posted that he had met "with more than a dozen Dem and GOP Senators in DC over the last 48 hrs." The New Republic called him "the most powerful crypto bro in Washington." He told Senra he went about once a quarter. "I don't mind going. In some ways, I like doing it. There's some pretty interesting people there."
Donald Trump won the presidency in November 2024 and embraced crypto. In February 2025, the SEC dropped its lawsuit against Coinbase. In May 2025, Coinbase joined the S&P 500. In October 2025, eight months after the dismissal, Coinbase appeared on the list of donors who funded the demolition of the White House's East Wing and the construction of a new presidential ballroom.
The man who had written that his company would stay out of politics had become one of the most consequential political donors in America. He would say there was no contradiction—that the mission memo had said Coinbase would engage on issues that touched its mission, and nothing touched it more directly than whether the SEC considered it legal. That is defensible. It is also exactly the kind of distinction that only feels clean from the inside.
And he never forgot Argentina. "We're actually losing freedoms," he said on a podcast in August 2024. Later, on stage in Singapore at a Network State conference with Srinivasan, he went further: "I also believe in exit... We need to start developing those backup options."
The billionaire lobbying Congress for clear rules in the morning. The believer in exit dreaming of backup nations at night. He holds both without visible strain.
Wartime All the Time
In May 2025, Coinbase disclosed that criminals had bribed overseas customer-support contractors to steal customer data—names, addresses, account information, though not passwords, private keys, or funds—and then demanded $20 million in ransom to keep quiet. Coinbase refused. Instead it offered a $20 million bounty for information leading to the attackers' arrest and conviction. On May 15 it estimated the incident would cost up to $400 million, much of it to reimburse customers who had been tricked into sending funds to scammers posing as Coinbase. The same month, Coinbase agreed to buy Deribit, the Dubai-based crypto derivatives exchange, for $2.9 billion—$700 million in cash and $2.2 billion in stock. Four days after the hack disclosure, it joined the S&P 500.
That cadence—catastrophe, counterpunch, expansion, all within a fortnight—is how Armstrong now describes his job. "I used to think that I was better as a peacetime CEO," he said in a 2026 interview. "But I think that's not true. It's actually totally flipped. I think it's wartime all the time." And, in a line that reads like an epitaph for his competitors: "The worst companies are basically destroyed in a crisis. Decent companies survive the crisis. The best companies are defined by the crisis."
It was not always a posture he could sustain. In a conversation with the podcaster Kevin Follonier, he described a period when he developed persistent back pain. He tried a new mattress, new ergonomics. Nothing worked. Then he realized that a leader had departed and he had absorbed the vacuum—twelve to fifteen direct reports. The back pain was the org chart. "Some mornings I wake up and I have to get myself into that mindset," he said. "There are some mornings I've woken up and I've been like, everything sucks. It's not working. Why am I even doing this?" On those mornings, by his own admission, his most reliable motivational tool is typing "David Goggins" into YouTube. "Then you'll be working out shortly, otherwise you will feel like a total wimp."
What he claims as his distinguishing trait is not intelligence. "I don't think I'm always like the smartest guy in the room," he told Follonier. "I'm not the best fundraiser ever. But I do have pretty extraordinary levels of determination." He calls himself "stubborn on the goal" and "very flexible on how to get there."
The flexibility has a price, and the price is a willingness to be wrong in public. In February 2023 Coinbase launched Base, its own layer-2 blockchain on Ethereum, built on the open-source OP Stack, led by an engineer named Jesse Pollak, whose hyperactive posting became, for a while, the public face of Coinbase's on-chain ambitions. By 2024 the plan was to turn the Base app into a decentralized social network where every post could become a tradable "content coin." It did not work. The tokens collapsed; critics complained that insiders and Coinbase-affiliated projects got preferential treatment. Pollak later observed, with a candor rare for a big-company executive, that it is hard to "build a decentralized blockchain inside of a big centralized public company."
On July 13, 2026, an X user complained that Base had spent more than a year pushing creator coins with nothing to show for it. Armstrong replied himself: "They didn't work and we pivoted early this year. We messed up, time to turn the page." Pollak moved into a quieter role. The app was refocused on trading, payments, and AI agents, "in that order."
By then, he was rebuilding the company around software that was not human. In May 2026 Coinbase cut 14 percent of its staff in a memo that announced the end of "pure managers" and noted that non-technical teams were now shipping production code. In June, he told the investor Molly O'Shea that 1,200 full-time AI agents worked at Coinbase and that some teams had shrunk to a single person. He talked about 4 billion people worldwide who had no brokerage account. He talked about an "Everything Exchange," where every asset class trades on blockchain rails, around the clock.
The boulder was rolling downhill again. He was chasing it again. He had stopped pretending he would ever stop.
Atoms, Not Bits
After the direct listing, he had money—real money, not paper—and he went looking for something to do with it that was not crypto. "I think we're kind of in this golden age of software where fortunes are being made," he told Ferriss. "But some of that wealth, even in crypto, is now being directed into hard science, hard science problems, atoms, not bits."
He started hosting dinners. Biotech CEOs, investors, scientists. He would go around the table and ask each person what they were most excited about. At one of these dinners, someone brought up epigenetic reprogramming: the line of research descending from Shinya Yamanaka's discovery that four transcription factors could turn an adult cell back into a stem cell, and the more recent finding that a partial dose might make a cell not younger in identity but younger in function. Salamanders regrow limbs. Babies who undergo surgery in the womb are born without scars. Somewhere in the body, he reasoned, was proof that rejuvenation was possible.
He called a friend, Blake Byers, a Ph.D. who had founded a biotech company before and invested in many more as a partner at Google Ventures. Together they funded NewLimit, which by late 2022 had eight or nine people in an office in South San Francisco and a plan to use machine learning to search a combinatorial space of transcription factors he estimated at ten to the fifteenth power. Its intermediate goals were concrete—rejuvenating exhausted T-cells in older people, say, or treating pulmonary fibrosis. Its moonshot was not. "We're trying to help humans live much longer, not just a little bit longer."
He realized, in the course of this, that he did not understand biology. "You get really demoralized when you read a sentence and you're like five of those words, I have no idea what they do." So he did what he had done at twenty-two when he wanted to learn entrepreneurship, except this time he paid someone else to write the book. He hired a graduate student to tutor him for an hour on Sunday afternoons. He wanted, he said, not to "sound like a complete baby" in front of the NewLimit team. "This is kind of how I sound when I try to speak Spanish. I sound like a five year old or whatever."
He also funded ResearchHub, an attempt to make scientific publishing look more like GitHub, with open peer review and a crypto token, ResearchCoin, to reward contributors. His pitch for it carried his core conviction about where change comes from. A lot of young people trying to improve the world, he said, think they should become activists or protestors. "My message to people... is I wanted to show people if you want to try to impact the world, I think one of the best ways to do it is to build a company that's based around science and technology."
There is something poignant in how he talks about his own aging. At thirty-nine, he told Ferriss, he had started noticing that jet lag hit harder, that injuries healed more slowly, that his short-term memory, once excellent, now dropped things two minutes later. "Maybe I only have another ten, 20, 30 years of really productive stuff, and what do I want to do with that?" The longevity company is a hedge against the one exit no network state can provide.
There is a slim self-published biography of him, Brian Armstrong- From Code to Crypto Empire: The Quiet Visionary Who Took Bitcoin Mainstream and Sparked a Financial Revolution, sixty-six pages long, which uses the word quiet in its title as though it were settled. It is half right. He is quiet in the way of a man who has decided where to spend his words. On Sunday afternoons he spends them asking a graduate student what PCR is.
No Bill Is Better Than a Bad Bill
The Clarity Act passed the House of Representatives 294 to 134 on July 17, 2025—a genuinely bipartisan margin, the payoff for the Fairshake money and the Stand With Crypto signatures and the quarterly trips. It was supposed to sail through the Senate. Armstrong visited Washington thirteen times between the 2024 election and the final vote. His policy team held multiple calls a week with industry partners. People involved in the negotiations told the Wall Street Journal that, besides Trump, few people had as much influence over the bill as he did, and that he could effectively veto portions he disliked.
Around the turn of 2026, Senator Angela Alsobrooks, a Maryland Democrat and one of the bill's key negotiators, came back from her holiday break to find a voicemail on her phone. It was from Armstrong, and it was about three minutes long. According to people familiar with it, he was admonishing her over changes that could have curtailed Coinbase's ability to pay stablecoin rewards, and the gist was that if she gave the banks an inch, they would take a mile. Some people working on the bill felt his tone was inappropriate for addressing a U.S. senator. The shy kid from San Jose, the conflict-averse founder, was now leaving senators voicemails that people described in those terms.
On January 14, 2026, hours before a scheduled Senate Banking Committee markup, he posted that Coinbase could not support the draft as written. "We'd rather have no bill than a bad bill." The committee postponed the session almost immediately. His objections, he said, were four: the draft's treatment of decentralized finance, of tokenized equities, of the CFTC's authority, and of stablecoin rewards. It was the week before Davos. It was why Dimon walked over.
Brad Garlinghouse of Ripple put the industry's frustration in a sentence: "We had a lot of momentum in January. Then one group in our industry kind of shot ourselves in the foot."
By Armstrong's account, the delay worked. The Senate Banking Committee advanced a revised bill 15 to 9 on May 14, 2026, with Alsobrooks voting yes; all four of his items, he said, had been fixed. He switched from blocker to backer and lobbied through the summer. And then the window closed. Trump's crypto profits, disclosed in June, raised fresh ethics objections among Democrats. A war in Iran, inflation, and backlash against data centers and rogue AI agents crowded the calendar. On September 15, 2026, a cloture vote on the motion to proceed failed 49 to 50, short of the 60 needed, despite 126 concessions in the final package. Bitcoin exchange-traded funds shed $450 million the same day.
Four days later, on a Saturday, he posted again: "Here we go again! The WSJ is working on a story blaming Coinbase and me personally for the CLARITY Act not passing." He accused the paper of "regurgitating bank lobby talking points." Then he laid out what he called "the boring TLDR"—the four items, the four months, the fixed draft. He was "proud to have done it, and would do it again, because it helped create a better bill."
Did he kill the bill? It is the kind of question that has no clean answer, which is why both sides will keep answering it. What is clear is that the same instinct that made Coinbase—stubborn on the goal, flexible on the path, unwilling to accept a compromise he thought was fundamentally wrong, willing to sit through a hundred bad headlines—was the instinct that cost the industry eight months. He bet that a better bill was worth the delay. The calendar called the bet.
He told Yahoo Finance afterward that the regulators, the SEC and the CFTC, could still deliver much of what Congress had not. Another path. There is always, with him, another path. Jump over the wall, tunnel under it, build an airplane around it.
— Brian Armstrong, Forbes Australia, November 2022You have to be willing to have 99% of people in the world think you're crazy, but not actually be crazy.
He did not leave. That is the thing. Ehrsam left for Paradigm. Carlson-Wee left to run a fund. Srinivasan left to theorize new countries. Reeves left before the first check cleared. Wahi bought a one-way ticket. Five percent of the company took the severance. Bankman-Fried and Zhao were removed. Armstrong, who wrote a book about breaking free and a memo about offering exits and gave a speech in Singapore about believing in exit, is still in the chair he took in June 2012, fourteen years on, nineteen percent of the company, the last big man standing.
Somewhere in an archive there is footage from that October 2022 shoot. An empty stage. A green screen behind him, a makeup sponge just pulled away, a black jacket, a teleprompter scrolling the words to create more economic freedom. He trips on the last phrase. He does not laugh, and he does not leave the stage.
"We'll try that one again."
Part IIThe Playbook
What follows is not a theory of Brian Armstrong but a set of operating principles extracted from his decisions—the ones that worked, and several that did not. Read together, they describe a founder whose edge was never brilliance or charisma but a kind of disciplined stubbornness: unusually clear about what he wanted, unusually willing to look wrong while getting there, and unusually ready to tell people where the door was.
Principles
- 1.Find the missing ingredient, then hide the complexity behind a button.
- 2.Make compliance the moat.
- 3.Be stubborn on the goal and flexible on the path.
- 4.Write to find out what you think.
- 5.Narrow the mission, then fund the exit.
- 6.Treat the first hit piece as a promotion.
- 7.Keep the enemy outside the building.
- 8.When the referee won't define the rules, force the question.
- 9.Turn customers into constituents.
- 10.Admit the miss in one sentence.
- 11.Assume wartime is the default.
- 12.Spend liquidity on problems that require atoms.
Coinbase's product-market fit arrived not through a feature request but through a confession: a user who had signed up for a wallet and left because he had nothing to put in it. The insight was that the product required an input the user could not easily obtain, and that the company—not the customer—should absorb the cost of obtaining it. Armstrong did not ask what features users wanted. He asked why they left.
The second half of the lesson is less discussed. The Buy button was the simplest object on the screen and the hardest thing Coinbase had built: a bank partner, a written anti-money-laundering policy, an ACH integration, a sourcing system for the bitcoin itself. Users saw one click. Behind it were months of the dull, regulated plumbing that most crypto founders were trying to avoid. The difficulty was the defensibility. Anyone could copy the button; few wanted to copy the bank relationship.
The same logic recurs throughout his career: the NFT marketplace that aggregated supply from other marketplaces rather than competing for it, the custody products for people who did not want to manage their own keys, the stablecoin rewards fight. Each time, the question is what step is stopping ordinary people, and whether Coinbase can quietly take that step for them.
Tactic: Call five users who churned, ask what they were missing when they left, and build the hardest part of that answer behind the simplest possible interface.
Principle 2
Make compliance the moat
In an industry whose founding myth was permissionlessness, Armstrong chose permission. Coinbase required taxpayer IDs, reported to the IRS, obtained New York's BitLicense in 2017, declined to list privacy coins, and accepted that it would "move a little bit slower." He told Stratechery in December 2022 that "following the rules is not always fast, but it's been a good validation of our strategy."
The validation came in a single month. When FTX collapsed in November 2022, the qualities that had made Coinbase look plodding—U.S. domicile, segregated custody, licenses—suddenly looked like the only things that mattered. Two years later, BlackRock and most of the U.S. bitcoin ETF issuers chose Coinbase as their custodian. Institutions do not buy the fastest exchange; they buy the one least likely to appear in a federal indictment.
The principle has limits, and the record shows them: a $50 million New York fine over 100,000 unreviewed monitoring alerts, a £3.5 million U.K. fine, an F rating from the Better Business Bureau for five years. Compliance as a strategy only works if the operational follow-through is real. Coinbase's advantage was not perfection but relative seriousness in a field where seriousness was rare.
Tactic: If your industry's incumbents are reckless, make trust your product—and budget the operational headcount to actually deliver it, because a compliance brand without compliance operations is a liability.
Principle 3
Be stubborn on the goal and flexible on the path
Armstrong's self-assessment is unflattering in its specifics—not the smartest, not the best fundraiser—and specific in its one claim: "extraordinary levels of determination." What he means by determination is not rigidity. "I'm stubborn on the goal and I'm very flexible on how to get there." The Android wallet was wrong, so he built a hosted wallet. The hosted wallet stalled, so he added a button. The SEC refused to write rules, so he sued; Congress stalled, so he funded elections; Congress failed, so he turned to the regulators.
The distinction matters because founders often reverse it: they fall in love with a method and drift on the goal. Armstrong's mission line—economic freedom through crypto—has barely changed since 2012. Almost everything else about Coinbase has.
Tactic: Write your company's goal in one sentence you would still defend in ten years, then list every method you are currently using to pursue it and mark which ones you would abandon tomorrow if a better path appeared.
Principle 4
Write to find out what you think
He wrote a book about entrepreneurship before he was an entrepreneur, admitted that he had done it backward, and kept the habit anyway. The mission-first memo went through multiple discarded drafts. His morning routine begins with writing the three things that would move the company forward if nothing else got done. When he wants to learn, he writes; now he blogs.
Writing does two jobs for him. It forces precision—you cannot be vague on a page the way you can be in a meeting—and it creates an artifact that scales his judgment to thousands of people who will never be in the room. The September 2020 memo did more to shape Coinbase's culture than any number of all-hands meetings, precisely because it was a document people could read, argue with, and decide to leave over.
Tactic: Before any decision that will affect the whole organization, write it as a public memo, throw away the first two drafts, and only then decide whether to send the third.
Principle 5
Narrow the mission, then fund the exit
The most famous Armstrong decision is often misread as a culture-war gesture. Operationally, it was a definition of scope plus a severance offer. He concluded that the company was split on whether it should engage broader social issues and that he was "failing as a leader" by not resolving it. He resolved it, and rather than punishing those who disagreed, he paid them to leave. About 5 percent did.
The financial cost of the severance was small relative to the cost of a permanently divided company. The important word in his framing is "aligned": "now we're fully aligned, we're making faster progress, everybody who joins the company knows what they're signing up for." Clarity, in his telling, was a recruiting tool as much as a filter.
The limits deserve equal weight. The memo did not make the underlying grievances disappear; the company later overhauled its compensation process in response to pay-gap reporting. And the "apolitical" frame became harder to sustain once Coinbase became one of the largest political donors in the country. Narrowing a mission is a tool, not an immunity.
Tactic: When internal conflict comes from genuine disagreement about what the company is for, settle the question in writing and offer a generous, no-shame exit to anyone who signed up for a different company.
Principle 6
Treat the first hit piece as a promotion
Armstrong anticipated many hardships in startup life but not this one: "What I didn't anticipate was that people would actually root for you to fail." He describes stress dreams and sleepless nights the first few times he had "a thousand or a hundred thousand people mad at you at the same time."
His coping system has three parts. Protect focus: do the day's one or two most important things before reading any comments. Curate feedback: keep a small circle—family, board, Ehrsam, Choi—licensed to tell him he is being an idiot, and ignore people "playing their own game." Reframe: "If the first time you get a New York Times hit piece or whatever, I mean, congratulations, you earned your wings as an entrepreneur." The reframe is not denial. It is a decision that a CEO who makes choices out of fear of reaction is a weaker CEO.
Tactic: Name three people who will tell you the truth without deference, and make them—not the comments section—your source of correction.
Principle 7
Keep the enemy outside the building
Armstrong borrows Paul Graham's line that "startups die by suicide, not homicide," and he pairs it with an observation about tribalism: if a company lacks an external rival, "tribes will start to form inside the company and people start fighting each other." Competitors, in this view, are useful—they give a naturally tribal species something to aim at besides each other.
He nonetheless tries to stay friendly with rival CEOs and frames the real goal as growing the crypto economy "100x" rather than winning share. The consistent move is to externalize conflict: competitors, regulators, banks, and Gensler's SEC all served, at different moments, as the outside force that kept Coinbase cohesive. The 2020 walkout was dangerous precisely because it turned the conflict inward.
Tactic: Give your team one clearly named external obstacle to beat this year, and treat recurring internal factional fights as a sign that you have failed to name one.
Principle 8
When the referee won't define the rules, force the question
Most regulated companies treat enforcement threats as a reason to settle. Coinbase retreated once—canceling its Lend product in 2021 after an SEC threat—and then never again. When the Wells notice came in March 2023, the company published its objections and sued the SEC to compel rulemaking. When the SEC sued in June, Armstrong, by his account, spent around $50 million on lawyers rather than settle. The case was dismissed in February 2025.
The strategic logic is that ambiguity favors the regulator and the offshore competitor at the same time. A compliant U.S. firm is the easiest enforcement target and the least able to compete with Bahamian exchanges that ignore the rules. Forcing a clear ruling, even an adverse one, converts an unbounded risk into a bounded one. Armstrong's advice to founders at a December 2025 Goldman Sachs event made this general: "you're also probably going to have to bump up against legislation and the government. Don't shy away from those problems."
Tactic: If an existential regulatory question hangs over your business, quantify what a definitive answer is worth to you and spend toward getting one, rather than spending indefinitely on avoiding the question.
⚖
The Long War With Washington
Coinbase's regulatory arc, from licensee to litigant to lobbyist.
2017
Coinbase obtains New York's BitLicense; IRS orders reporting on users with $20,000+ in annual transactions.
2021
SEC threatens suit over the Lend product; Coinbase cancels the launch.
2023
Wells notice in March; Coinbase sues SEC in April; SEC sues Coinbase in June. NYDFS levies a $50M fine.
2024
Coinbase contributes $49M to Fairshake; Armstrong lobbies Senators in person.
2025
SEC dismisses its lawsuit in February; Clarity Act passes the House 294–134 in July.
2026
Armstrong pulls support in January, backs revised bill in May; Senate cloture fails 49–50 on September 15.
Principle 9
Turn customers into constituents
The decisive move in Washington was not the lawsuit but the realization that Coinbase's customer base was also a voting bloc. Stand With Crypto, by Armstrong's count, signed up some 2 million Americans; Fairshake gave politicians a reason to fear being anti-crypto. His message to Congress, as he summarized it, was that there were no votes in opposing crypto and plenty in supporting it.
This is a playbook available to any company whose users are numerous and whose policy environment is hostile: organize the people who already depend on your product into a political force with its own identity, separate from the company's brand. Its risk is the one the Clarity Act exposed. Political capital can be spent badly, and a company with veto power over legislation will be blamed when the legislation dies—fairly or not.
Tactic: If your regulatory fate depends on lawmakers, build a measurable, opt-in constituency of affected users before you need it, and keep it distinct from your marketing list.
Principle 10
Admit the miss in one sentence
Armstrong's apologies are short. On Neutrino in 2019: the company "did not properly evaluate" the deal, and the affected staff "will transition out." On Base's content coins in 2026: "They didn't work and we pivoted early this year. We messed up, time to turn the page." On the June 2020 statement, he reversed course on Twitter within days, then reversed his reversal months later in the mission memo.
The brevity is the point. A one-sentence admission costs little and closes the story; a defensive paragraph keeps it open. The pattern also reflects his "action produces information" ethos: if you ship quickly, you will be wrong often, and the organizational skill that matters is how cheaply you can acknowledge it and move on.
Tactic: When a bet fails, publish a two-sentence acknowledgment that names the mistake and the new direction, and resist the urge to explain why it was reasonable at the time.
B
Conventional Wisdom vs. the Armstrong Approach
| Conventional wisdom | Armstrong's approach | Outcome |
|---|---|---|
| Crypto startups should avoid regulators. | Seek licenses early; report to the IRS. | Scaled |
| CEOs must take public positions on social issues. | Declare mission-first; pay dissenters to leave. | Contested |
| Settle with the SEC. | Sue the SEC and refuse to settle. | Won |
| Take the bill you can get. | "No bill is better than a bad bill." | Bill died |
| Build social features to drive engagement. | Content coins on Base. | Abandoned |
| Pay ransoms quietly. | Refuse $20M ransom; post $20M bounty. | Signal sent |
Principle 11
Assume wartime is the default
Armstrong once believed he was a peacetime CEO. Fourteen years of crypto cycles cured him of it: "It's actually totally flipped. I think it's wartime all the time." His formulation of crisis—"The best companies are defined by the crisis"—is a description of 2025 in miniature. Hit with an extortion demand, Coinbase refused the $20 million ransom, offered a $20 million bounty instead, reimbursed customers at a cost of up to $400 million, and in the same month announced a $2.9 billion acquisition and entered the S&P 500.
Wartime as a default has costs he has also described honestly: burnout, back pain from carrying 12 to 15 direct reports, mornings when "everything sucks." His remedy is equally unglamorous—exercise, nature, friends, sleep, and redistributing the load. A permanent war is survivable only if the general's life is designed for decades rather than sprints.
Tactic: Pre-decide your response to your three most likely crises—ransom, regulatory action, market collapse—so that when one arrives, you are executing rather than deliberating.
Principle 12
Spend liquidity on problems that require atoms
After the 2021 listing, Armstrong did not buy a sports team. He hosted dinners, asked scientists what excited them, co-founded NewLimit with Blake Byers to pursue epigenetic reprogramming, funded ResearchHub, and hired a graduate student to teach him biology on Sundays. His thesis is that the most valuable companies emerge when "true scientific innovation" meets "someone who can go commercialize it"—Genentech, SpaceX, Google, and Coinbase itself, which began with a research paper.
The tactical element is humility in the form of tuition. He did not pretend to understand cell biology; he paid to learn enough not to "sound like a complete baby" in front of the people he funded. The broader message to founders comes from his Goldman Sachs remarks: "people generally don't attack ambitious enough problems, but that's where the value is created and there's fewer people who play in that space."
Tactic: If you have capital and want to move into a hard-science field, hire a tutor before you hire a team, and spend your first months asking experts what they are most excited about rather than telling them what you want to build.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Part IIIQuotes and Maxims
In their words
— Brian Armstrong, on discovering the Bitcoin white paper, Bloomberg Players Technology Summit, August 2018It just grabbed me and I felt like this is the most important thing I have read in like five years.
— Brian Armstrong, The Tim Ferriss Show, October 2022There's a real art to preserving your mental focus. I think that's the most valuable scarce resource you have, and so you have to really preserve that.
— Brian Armstrong, The Tim Ferriss Show, October 2022You don't have to just consume the world. The world is not happening to you. You can go create things in the world.
— Brian Armstrong, Goldman Sachs, December 2025It's probably going to take 10 years to even make a dent in it, but by 20 years you'll really start having an impact.
— Brian Armstrong, The Tim Ferriss Show, October 2022The greatest risk is not taking one.
Maxims
- Action produces information. Ship the wrong thing quickly, because being wrong in public teaches you faster than being right in private.
- Ask why they left. The founding insight is usually hidden in a churned user's explanation of what they were missing.
- Simple outside, hard inside. The button that looks trivial should sit on top of the work your competitors refuse to do.
- Trust is a product. In a reckless industry, the slowest compliant company can become the default choice of every serious institution.
- Stubborn goal, flexible path. Hold the mission fixed for a decade and let every method be disposable.
- Write before you decide. A memo that has survived three drafts will align more people than any meeting.
- Define the mission, and price the door. Clarity creates disagreement; a generous exit turns that disagreement into alignment.
- Celebrate the first hit piece. Once you survive public hostility, you can make decisions without fear of reaction.
- Startups die by suicide. Keep conflict pointed outward, or it will turn inward.
- Admit it in one sentence. "We messed up, time to turn the page" closes a story; a defense keeps it open.
