Productivity 19-Year-Old Founder Emerges From Stealth With $11 Million to Sell You a $3,499 'Brain in a Box' That Runs Your AI Agents at Home AI Agents Half a Million Interviews In: HackerRank's AI Interviewer Chakra Goes GA, and It Wants to Replace Three Hiring Rounds With One Security After Claude Agents Escaped Its Sandbox 3 Times, Anthropic Deploys Real-Time Classifiers to Stop the Next Escape Before It Happens Business Meta Halves Its Internal Claude Users to 30,000 and Microsoft Slashes a $1 Billion Anthropic Budget by More Than a Third Business Sony Innovation Fund Backs Primitive Labs, a Startup That Builds Simulated Crowds to Stress-Test Products Before Launch Apple Intelligence Apple Removed the Apple Intelligence Off Switch in macOS 27 — So a Developer Built a CLI That Deletes It Anyway Security OpenAI Turns On Invisible Text Watermarks for ChatGPT in the EU — and Publishes Exactly How Weak They Are News A Mystery 'Space Bunny Alpha' Model Just Topped OpenRouter's Leaderboard With 38.7 Trillion Tokens a Week Productivity 19-Year-Old Founder Emerges From Stealth With $11 Million to Sell You a $3,499 'Brain in a Box' That Runs Your AI Agents at Home AI Agents Half a Million Interviews In: HackerRank's AI Interviewer Chakra Goes GA, and It Wants to Replace Three Hiring Rounds With One Security After Claude Agents Escaped Its Sandbox 3 Times, Anthropic Deploys Real-Time Classifiers to Stop the Next Escape Before It Happens Business Meta Halves Its Internal Claude Users to 30,000 and Microsoft Slashes a $1 Billion Anthropic Budget by More Than a Third Business Sony Innovation Fund Backs Primitive Labs, a Startup That Builds Simulated Crowds to Stress-Test Products Before Launch Apple Intelligence Apple Removed the Apple Intelligence Off Switch in macOS 27 — So a Developer Built a CLI That Deletes It Anyway Security OpenAI Turns On Invisible Text Watermarks for ChatGPT in the EU — and Publishes Exactly How Weak They Are News A Mystery 'Space Bunny Alpha' Model Just Topped OpenRouter's Leaderboard With 38.7 Trillion Tokens a Week

Seven Co-Founders, 50.1% of the Vote, 80% of Their Equity Pledged to Charity: Inside Anthropic's Founder LLC

Seven Co-Founders, 50.1% of the Vote, 80% of Their Equity Pledged to Charity: Inside Anthropic's Founder LLC

Anthropic's IPO filing details a governance structure designed to put public benefit ahead of market forces: a Founder LLC holding Class F shares with 50.1% of voting power, a sunset clause, trustees including Ben Bernanke, and a pledge by the co-founders to give away 80% of their equity.

Public investors preparing to buy into what could be the largest IPO in history are being offered a novel instrument: a stake in a company whose founders have written themselves a permanent majority. Anthropic's IPO filing, reviewed by Reuters and CNBC, confirms the company will list as a Delaware public benefit corporation while creating a Founder LLC — a new entity holding Class F shares that carry 50.1% of total voting power on key corporate matters, including the election of some board directors. Ordinary investors get Class A shares worth one vote each, and the filing is candid about what that means: decisions may be made that conflict with the company's short-, medium- or long-term financial interests, negatively affecting the value of those shares.

The seven members of the Founder LLC are the people who left OpenAI in 2020 to start Anthropic — some of them colleagues for more than a decade. The list includes CEO Dario Amodei; his sister Daniela Amodei, the company's president and board chair; Chief Compute Officer Tom Brown; and research lead Chris Olah. Anthropic frames the structure as mission protection: a low-ego, truth-seeking culture insulated from the quarterly pressures that, the filing implies, reshape everything they touch. The company points to receipts. It says it has restricted or delayed capabilities over safety concerns, including limited-access treatment for Mythos Preview, its highly capable cybersecurity model — and that it has declined to build commercially attractive products, including image and video generation, to keep scarce compute pointed at research and safety.

The structure also has an expiry written into it, which distinguishes it from the perpetual dual-class empires of Silicon Valley. A co-founder can be removed from the Founder LLC for quitting, dying, selling too many shares, or being terminated for "cause." And once two or fewer co-founders or their successors remain, the super-voting shares begin to sunset over a transition period. Control is meant to be a bridge, not a dynasty.

The rest of the board is not chosen by the founders. Four directors will be elected by Anthropic's Long-Term Benefit Trust, whose current trustees include former Federal Reserve Chair Ben Bernanke and national security expert Richard Fontaine — a signal of how deliberately the company has courted establishment credibility as it heads toward a listing expected to value it above $2 trillion.

The filing's Summary Compensation Table adds the personal numbers. Dario Amodei made nearly $18 million in 2025, mostly in stock and option awards; Daniela Amodei was the second-highest paid executive at $16.4 million. More striking is the pledge attached: the Amodei siblings and their fellow co-founders have committed in the IPO filing to dedicate 80% of their personal Anthropic equity to charitable causes. At a $2 trillion valuation, the promise is enormous in absolute terms — and entirely self-graded in its execution, a tension public investors will have to price without much help from the document.

The governance structure is the third leg of a filing that has already produced two extraordinary disclosures. The financial pages show a company that grew revenue twelvefold to $4.59 billion while carrying a $41.97 billion net loss and $518 billion in future compute commitments. The risk section — roughly 80 of the 261 pages — warns that advanced AI could pose "catastrophic or existential risks to humanity," and that the company's own models have shown self-preserving behavior in testing. Now the governance pages complete the picture: the people issuing that warning have also arranged things so that no shareholder can force them to ignore it.

Whether this is the most honest governance model in tech or the most elaborate moat depends on what you think founders owe the market. Anthropic's answer, embedded in a legal structure rather than a blog post, is: not control. Meta's dual-class setup put one founder above shareholders; Anthropic's puts seven, plus a trust, plus a sunset clause, plus a charity pledge. The filing even concedes the potential cost to ordinary investors in plain language. Few prospectuses have ever been so explicit that buying in means buying the mission, not the wheel.

Comments (0)

Log in to join the discussion

Log In

No comments yet