Anthropic Founders Seek Majority Voting Control Ahead of IPO

Anthropic is asking shareholders to approve a new class of special shares that would give CEO Dario Amodei and his six co-founders a combined 50.1% vote on most corporate matters ahead of an initial public offering. TechCrunch

The voting control would stay in place as long as at least three of the founders retain a minimum stake. It is structured as control without added economics. The proposed shares would carry no extra economic value.

Each of the seven founders owns about 2% of the company, including Amodei. That leaves the group with a minority economic position under a majority voting position. The co-founders have pledged to give away 80% of their wealth.

The proposal also revises board and employee mechanics. Under the plan, Anthropic's Long-Term Benefit Trust would continue to select most of the board. The founders' own board seats would increase from two to three. Employees would receive their own class of stock to break ties on some issues.

The governance request is tied to a public listing process that is already underway. Anthropic filed confidentially for an initial public offering, a step it announced in a company blog post. TechCrunch

Valuation expectations around that listing are large. Anthropic was valued at $965 billion in May 2026. It was more recently valued at $1.5 trillion on the secondary market. Nvidia Corp. is considering investing as much as $10 billion in Anthropic PBC's initial public offering. Bloomberg

Economics and control would diverge. That is the point.

The broader context here is how AI labs are trying to lock governance before public-market pressure arrives. A 50.1% founder bloc on most corporate matters centralizes veto power, even while the Trust retains board selection and employees hold a tie-breaking instrument on defined questions. For engineers and enterprise buyers, the practical question is continuity of research direction and product policy through a listing that will expand the shareholder base and invite quarterly scrutiny.

In my view, the structure is best read as an attempt to separate capital scale from decision scale. Anthropic needs public-market capital at a trillion-dollar order of magnitude, yet wants model development and safety judgments insulated from short-term shareholder shifts. The no-extra-economics provision, the three-founder persistence rule, and the pledge to give away most founder wealth all point in that direction. They do not remove founder power. They frame it as stewardship rather than ownership leverage.

Worth flagging for technologists is what remains unspecified in the verified terms. Minimum stake is not defined in the disclosed facts, nor is the exact scope of most corporate matters or the some issues subject to an employee tie-break. Those definitions will determine whether the Trust, the founders, or staff decide contested cases. Public investors will price that ambiguity.

The optimistic reading, and the one I lean toward, is that durable governance can help long-horizon technical work survive listing. Large-scale training, evaluation, and deployment decisions play out over years, not quarters. If the voting structure holds, Anthropic can raise at IPO scale while keeping its research leadership accountable to a small, named group rather than a shifting majority.