Anthropic is now arguably the world’s most influential AI company.
Its Claude large language model is deployed at the Pentagon, U.S. intelligence agencies, and national laboratories, supporting military intelligence analysis and target selection for operations against Iran.
In less than three years, Anthropic’s annualized revenue has soared from zero to $14 billion. In February 2026, the company closed a $30 billion Series G round, pushing its post-money valuation beyond $380 billion. Tech giants—Amazon, Google, Nvidia, and Microsoft—are all eager to invest.
In recent weeks, Anthropic has been locked in high-profile negotiations with the Pentagon over the weaponization of AI, a contest watched by the entire world.
One name from Anthropic’s early funding history continues to spark discussion: Sam Bankman-Fried.
In April 2022, before ChatGPT existed and before the AI race became what it is today, SBF used his hedge fund, Alameda Research, to pour $500 million into Anthropic’s Series B round. He took 86% of the round and secured roughly 8% equity. Seven months later, the FTX empire collapsed, and SBF became the central figure in the largest fraud case in crypto history, sentenced to 25 years in prison. That $500 million came from FTX customer deposits.
Had SBF not been arrested, and had those funds been legally sourced, his 8% stake would be worth over $30 billion at today’s $380 billion valuation. That’s a 60x return—one of the highest in venture capital history.
A convicted crypto fraudster serving time in federal prison came within a hair’s breadth of making the wildest bet in AI investment history.
Why did SBF spot Anthropic in 2022? Why did he risk $500 million? Why did Anthropic accept the money?
The answer lies in a community known as Effective Altruism.
A Shared House, a Movement, a Check
In mid-2010s San Francisco, a group of people lived in similar shared houses, attended the same gatherings, read the same academic papers, and embraced a shared philosophy.
That philosophy, Effective Altruism (EA), is simple at its core: charity should be based on calculation, not intuition. Every dollar should flow to where it can mathematically “maximize good.” For a key branch of EA, humanity’s greatest existential risk isn’t nuclear war or pandemics—it’s uncontrolled artificial intelligence.
Dario Amodei was deeply embedded in this circle.
He was the 43rd signatory of the Giving What We Can Pledge, committing to donate at least 10% of his income, and became a fan of GiveWell as early as 2007 or 2008.
He shared a house with Holden Karnofsky—co-founder of GiveWell and Open Philanthropy, and one of EA’s most influential funders—and Paul Christiano, a core researcher in AI alignment. At the time, Dario and Paul both served as technical advisors to Open Philanthropy.
Later, Karnofsky married Dario’s sister, Daniela. After their engagement, the couple lived with Dario for a time. In January 2025, Karnofsky quietly joined Anthropic as “technical staff” responsible for safety policy. When Fortune’s reporter uncovered this, Anthropic hadn’t even announced the appointment.
This is a close-knit social network.
Anthropic’s early employee Amanda Askell is the ex-wife of William MacAskill, one of the EA movement’s founders. She was the 67th signatory of GWWC and wrote her doctoral thesis on a core EA philosophical problem: how ethics should handle infinity.
Anthropic’s main governance body, the Long-Term Benefit Trust, theoretically holds major control over the company. Of its four members, three come directly from the EA system: former GiveWell Managing Director Neil Buddy Shah, Center for Effective Altruism CEO Zach Robinson, and Kanika Bahl, CEO of Evidence Action, a long-term GiveWell grantee.
The three largest donors in EA history were all early investors in Anthropic: Facebook co-founder Dustin Moskovitz, Skype co-founder Jaan Tallinn, and Sam Bankman-Fried.
This is how SBF found Anthropic—not through investment genius or prescient AI insight, but through an internal funding loop: EA money flowing to EA projects to solve EA-defined problems.
SBF followed a more radical EA branch: “earning to give.” He left Jane Street, a Wall Street quantitative firm, to pursue crypto, publicly stating his goal wasn’t personal wealth but “altruism”—to earn as much as possible and direct funds for the greatest positive impact. Anthropic’s mission—“developing powerful AI safely”—is nearly a textbook EA response to AI existential risk.
In May 2021, Jaan Tallinn led Anthropic’s $124 million Series A, with Moskovitz following. In April 2022, SBF led the Series B, writing a $500 million check—86% of the $580 million round. Other participants included Caroline Ellison, Nishad Singh, and Jane Street’s James McClave.
This list is telling: Caroline Ellison was Alameda’s CEO, Nishad Singh was FTX’s Director of Engineering, and Jane Street was SBF’s former employer.
The $580 million Series B round was, in effect, almost entirely funded by SBF and his inner circle.
Red Flags and Compromises
Dario Amodei is no fool.
In a later in-depth interview, he recalled that SBF seemed “bullish on AI and concerned about safety”—well-aligned with Anthropic’s direction. But Dario added a crucial point: he noticed “enough red flags.”
So he made a decision: take the money, but isolate it in the governance structure. SBF received non-voting shares and was excluded from the board. Dario later described SBF’s behavior as “much more extreme and much more egregious than I imagined”—repeating “much more” three times.
This decision proved extremely wise. But it left a sharp question: if there were enough warning signs to require governance isolation, why take the money at all?
In early 2022, the AI funding landscape was not nearly as hot as today. Anthropic needed significant capital to build compute infrastructure, and an investor willing to put up $500 million at once—regardless of “red flags”—was hard to find.
But there’s a subtler reason: in the EA community’s operational logic, the “cleanliness” of funds is never the top priority. What matters is “effectiveness”—whether the money enables greater impact. SBF’s entire wealth narrative rests on this: making money is a means, doing good is the end, so the means can be flexible as long as the ultimate “good” is large enough.
SBF took this logic to a criminal extreme, but at the time of his Anthropic investment, it appeared to be a radical yet legal philosophical choice.
After the Collapse: A Dark Comedy
The rest of the story is well known in crypto circles.
In November 2022, CoinDesk exposed Alameda’s balance sheet. Changpeng Zhao announced the sale of FTT, triggering a run on FTX. Within nine days, the empire collapsed. SBF was arrested, extradited, tried, and sentenced to 25 years in March 2024. Anthropic’s 8% equity, along with all assets, was frozen in bankruptcy proceedings.
One courtroom episode, though excluded from trial, is worth noting.
SBF’s defense tried to present the Anthropic investment as evidence of “foresight”—“See, he wasn’t just squandering funds, he made an investment that multiplied in value.”
Prosecutor Damian Williams was blunt: whether the investment was profitable is irrelevant to the fraud charge. If you steal others’ money to invest, any gains are still stolen. The judge agreed, and Anthropic’s name was excluded from the proceedings.
The prosecution added: wasn’t FTX itself the best counterexample? Valued at $18 billion in 2021, $32 billion in 2022, and now worthless.
Then came the liquidation auction.
In March 2024, the first round valued at $884 million.
The largest buyer, Abu Dhabi’s sovereign fund Mubadala, invested $500 million—the same amount SBF originally put in. The second largest was Jane Street, SBF and Caroline Ellison’s former employer; Jane Street’s Head of Quantitative Research, Craig Falls, personally contributed $20 million. SBF’s first job after graduating from MIT was as a trader at Jane Street, and now his old employer was buying back shares he’d acquired with illicit funds.
Two rounds recovered $1.34 billion. These funds went into FTX’s creditor pool, becoming a key resource for compensating affected users.
What if the liquidation team hadn’t sold?
In February 2026, Anthropic completed a $30 billion Series G, reaching a $380 billion valuation. Without dilution, that 8% could have grown from $1.34 billion to $30 billion. The liquidation team, of course, had to cash out quickly to repay creditors, but the gap—$1.34 billion vs. a potential $30 billion—is central to why this story remains so compelling.
It’s the biggest “what if” in the entire FTX bankruptcy saga.
EA’s Collective Amnesia
Anthropic’s current scale and influence need no elaboration, but an interesting phenomenon has emerged: the company is systematically distancing itself from the EA movement.
All seven co-founders have pledged to donate 80% of their personal wealth. At current valuations, their combined commitments would be worth about $38 billion. Nearly 30 Anthropic employees signed up for EA conferences in San Francisco—more than twice the combined total of OpenAI, Google DeepMind, xAI, and Meta’s superintelligence labs.
But in a Wired interview, Daniela Amodei said: “I’m not an expert on effective altruism. I don’t identify with that term. My impression is it’s a bit of an outdated phrase.” Her husband, one of EA’s most influential funders, had just joined her company.
This “taking EA money, hiring EA people, living in EA shared houses, but not admitting to being EA” posture became understandable after the SBF case. FTX’s collapse tanked EA’s reputation. Anthropic needs to distance itself from the label, just as any savvy company would cut ties with a brand facing negative associations.
But the facts remain: Anthropic’s founding logic comes from EA’s core arguments about AI existential risk; its early funding was almost entirely sourced from the EA network; and its governance is dominated by EA-system personnel.
A Parallel Universe in Prison
Sam Bankman-Fried is now in federal prison, eligible for release as early as 2049—he’ll be 57.
During his incarceration, the AI company he invested in with illicit funds has seen its valuation soar past $380 billion and is locked in a globally watched contest with the Pentagon over AI weaponization. Its founders are regulars at The New York Times and Capitol Hill. Had everything been legal, that $500 million bet would have made SBF one of the most successful venture investors of the era.
SBF’s “earning to give” and Anthropic’s “safe AI development” share the same underlying operating system: for a large enough positive outcome, unconventional means and risks are acceptable.
SBF crossed the line into criminality. Anthropic operates on the safe side of that line, but its core thesis—“we must build the most powerful AI ourselves to ensure AI safety”—is itself a grand, almost self-fulfilling wager.
They grew from the same soil.
In that soil, Dario and SBF once attended the same gatherings, believed in the same philosophy, and lived at different nodes of the same social network. One went on to build a $380 billion AI empire, the other ended up in federal prison.
And the $500 million check connecting them remains the most bizarre chapter in Anthropic’s history.
Disclaimer:
-
This article is republished from [TechFlow], with copyright belonging to the original author [TechFlow]. If you have concerns regarding this republication, please contact the Gate Learn team, who will address the issue promptly in accordance with relevant procedures.
-
Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute investment advice.
-
Other language versions of this article are translated by the Gate Learn team. Without explicit mention of Gate, reproduction, distribution, or plagiarism of the translated article is prohibited.
