Anthropic's IPO: Public Shareholders Locked Out
Original: Who’s really in charge at Anthropic after the IPO?
Why This Matters
Sets a precedent for how safety-focused AI labs can structure public listings while retaining mission control.
Anthropic is heading to an IPO with a governance structure that gives common shareholders virtually no control over board composition, routing power instead to a Long-Term Benefit Trust, seven founders, and the board itself.
Anthropic's IPO governance structure is unusually restrictive—even by Silicon Valley founder-control standards. Three interlocking mechanisms keep public shareholders largely sidelined.
First, Anthropic is incorporated as a public benefit company, legally shielding board decisions made in the name of AI safety from shareholder lawsuits. The board is chaired by president Daniela Amodei and includes CEO Dario Amodei, Netflix co-founder Reed Hastings, former Microsoft/GM CFO Christopher Liddell, Novartis CEO Vas Narasimhan, and Spark Capital's Yasmin Razavi.
Second, a Long-Term Benefit Trust (LTBT) holds a share class that lets it elect and remove a majority of board directors. The LTBT is chaired by GiveWell alumnus Neil Shah and includes former Fed chair Ben Bernanke, national security expert Richard Fontaine, and Harvard physicist Christopher Stubbs. Crucially, the LTBT's own members choose who fills any vacancy—no shareholder input required.
Third, Anthropic's seven co-founders collectively hold a share class conferring 50.1% of voting power, which they can use to elect three board seats so long as a majority of founders agree.
One partial check exists: when the LTBT nominates or removes a specific category of director, the board's nominating committee can reject the move and send it to an employee vote instead. Common stockholders don't appear in that chain at all.