Before Anthropic rings the opening bell, its power is not concentrated in a single chair or a founder majority in the ordinary sense. Instead, control is deliberately sliced across several layers: the current board of directors, an independent Long-Term Benefit Trust holding a special class of shares, a founder bloc with a locked-in voting majority, and an employee-vote mechanism that can override certain trustee decisions. This guide walks you through each layer in the order you would actually encounter them in a news report, so you can follow corporate-control stories like this without a law degree.
The board of directors is currently chaired by President Daniela Amodei. Its members include CEO Dario Amodei, her brother; Netflix co-founder Reed Hastings; Christopher Liddell, former chief financial officer of Microsoft and General Motors; Vas Narasimhan, CEO of Novartis; and Yasmin Razavi, a partner at Spark Capital. The board presently carries one vacancy.
For readers who follow corporate news casually, that roster is the part you will most often see quoted in financial coverage, because it is the formal governing body visible in any filing.
Anthropic hands an important share of its board-level control to the Long-Term Benefit Trust, or LTBT. The LTBT is an independent group whose stated goal is to keep the company focused on its mission of benefiting humanity. It holds a class of shares that gives it the right to elect and remove the majority of directors — specifically, the Class T directors.
The LTBT is chaired by Neil Shah, a global-health expert and former executive director of the charity-evaluation nonprofit GiveWell. Its members include former Federal Reserve chair Ben Bernanke, national-security specialist Richard Fontaine, and Harvard physicist Christopher Stubbs. The trust also has one open seat right now; remaining members will pick the replacement themselves.
The practical takeaway: an entity that does not report to public shareholders can still swing most of the boardroom. That is unusual enough to explain why Anthropic's governance keeps landing on tech-news front pages.
Anthropic's seven founders jointly hold a class of special shares. A majority of those founders therefore controls 50.1% of shareholder voting rights, which is enough to elect three board members. If the founders cannot form a majority for any reason, their shares revert to voting proportionally alongside ordinary shareholders.
The board acts as a counterweight to the LTBT. When the LTBT nominates one of its four directors as a designated director, or removes that person without cause — meaning no misconduct is alleged — the board's nomination committee can refuse the move and send the final call to an employee vote.

In short, the staff is not merely a constituency to be placated; in defined situations they hold the tiebreaker on a specific trustee seat.
The dynamics of this layered structure are hard to forecast. Reporting has noted that a similar arrangement was effectively overthrown before: OpenAI's board briefly removed CEO Sam Altman after concluding he had not been candid with them, but strong employee loyalty to Altman pushed the board to reverse course, restore him, and stand up a new governance structure. Anthropic's design appears to institutionalize that kind of employee power.
Two fracture scenarios follow directly from the rules above. If the seven founders fail to act as a united bloc, ordinary shareholders may end up with more influence than the current numbers suggest. And if employees side with the board rather than with the trust, they can overrule the LTBT on a board-seat question.
Founder control is not permanent. Ten years after the IPO, the founders may choose to give up their special voting power; that control also ends automatically if fewer than three founders remain. Meanwhile, the LTBT persists unless 85% of ordinary shareholders vote to dissolve it, or it decides to exit with board approval. Only if both classes of special shareholders disappear — reporting frames that hope as "not because of an AI-caused catastrophe" — do ordinary shareholders finally gain full control.
Q: Does Anthropic have one shareholder with majority economic ownership?
A: The article does not describe an economic majority. It describes a voting majority: seven founders jointly hold a class of special shares worth 50.1% of shareholder voting rights.
Q: How many Class T directors does the LTBT appoint?
A: The trust holds a class of shares with the right to elect and remove the majority of Class T directors; in the employee-vote scenario the trust is described as having four directors, one of whom is the designated seat.
Q: What happens to the vacant seats?
A: The board carries one vacancy, and the LTBT also has one open seat whose replacement the remaining members will choose themselves.
Q: Is the OpenAI comparison a prediction about Anthropic?
A: No. Reporting merely uses OpenAI's board episode involving Sam Altman as an example of how a similar layered structure can be overturned in practice. The comparison illustrates unpredictability, not a forecast.
Q: What ends founder control?
A: A founder choice ten years after the IPO, or automatic termination when fewer than three founders remain.
Governance structures like this read as convoluted until you line them up as a sequence of who appoints whom, who can remove whom, and which votes dissolve which rights. Once you see Anthropic's board, Long-Term Benefit Trust, founder special shares, and employee tiebreaker as four linked mechanisms rather than four separate news items, the company's pre-IPO control map becomes a single connected picture — and any future headline about a change to one of those pieces immediately signals how the others might react.