Anthropic IPO Governance: Founders Hold 50.1% Voting Power, LTBT Controls Board

Executive Overview

On October 27, 2026, TechCrunch AI published an analysis of Anthropic's complex pre-IPO governance structure, revealing that the company's seven founders hold 50.1% of voting rights and elect three board seats, while the Long-Term Benefit Trust (LTBT) controls the election of the board majority. Ordinary shareholders will not gain effective control until 10 years after the IPO, when fewer than three founders remain, or via an 85% shareholder vote to dissolve the trust. The structure is designed to protect Anthropic's AI safety mission from shareholder pressure, but it raises significant questions about accountability and transparency ahead of the company's public listing.

Deep-Dive Features & Architecture

Anthropic was founded in 2021 as a Public Benefit Company, a legal status that allows the board to balance public mission with profit without facing lawsuits from angry shareholders. The board is currently chaired by company president Daniela Amodei and includes her brother and CEO Dario Amodei, Netflix co-founder Reed Hastings, former Microsoft and General Motors CFO Christopher Liddell, Novartis CEO Vas Narasimhan, and Spark Capital partner Yasmin Razavi, with one vacant seat.

The Long-Term Benefit Trust (LTBT) owns a class of shares that allows it to elect and remove the majority of the board (Class T directors). It is chaired by global health expert and former GiveWell director Neil Shah and includes former Federal Reserve Chairman Ben Bernanke, national security expert Richard Fontaine, and Harvard physicist Christopher Stubbs, with one vacant seat filled by the current members. The seven founders retain a special class of shares granting their majority 50.1% of voting rights and the election of three board members; if no majority exists among them, their shares are voted proportionally with ordinary shareholders.

The board can place a check on the LTBT: when the trust nominates or removes one of its four "designated" directors without cause, the board's nominating committee can reject the decision and refer it to an employee vote. Founder control ends 10 years after the IPO or automatically when fewer than three founders remain, and the LTBT continues until 85% of ordinary shareholders vote to dissolve it or the trust decides to step down with board approval.

Benchmark & Competitive Performance

The report compares this structure to Silicon Valley governance models, where founders like Elon Musk and Mark Zuckerberg have controlled their public companies without owning a majority of shares. However, Anthropic's case is more complex due to its obsession with AI safety, producing a "baroque" governance that limits ordinary shareholder influence. The report invokes the 2023 OpenAI crisis, when the board fired Sam Altman and then reinstated him under employee pressure, noting that Anthropic's structure formalizes employee power by granting them the right to vote on Class T directors if the nominating committee rejects an LTBT decision.

Industry Impact & Enterprise Adoption

For developers and startups in the Arab world, this structure means that transparency about who controls Anthropic—and by extension, the safety policies and deployment priorities of its models—will remain limited for ordinary investors. The founders' 50.1% voting bloc and the LTBT's board majority effectively insulate the company from hostile takeovers and activist shareholder campaigns, but they also concentrate decision-making power in a small group. As Anthropic prepares for its IPO, enterprise customers and partners will need to assess whether this governance model provides the stability and accountability they expect from a public company. The structure may also influence how other AI safety-focused startups design their own governance as they scale.

Conclusion

Anthropic's pre-IPO governance is a bold experiment in balancing mission protection with public accountability. The founders' 50.1% voting rights and the LTBT's board control ensure that the company's AI safety mission cannot be easily overturned by short-term shareholder interests. However, ordinary investors will have limited influence for at least a decade, and the complex checks and balances—including employee votes on Class T directors—may complicate decision-making. As Anthropic enters public markets, the effectiveness of this structure will be tested by market pressures and the evolving regulatory landscape for AI governance.

Media Source: TechCrunch AI | Fact Verification & Analysis: AI Tools Oasis