Anthropic’s elaborate plan to keep Wall Street at bay

What happened

Anthropic’s proposed IPO governance structure aims to protect its AI safety mission by limiting ordinary shareholders’ influence. Its long-term benefit trust would appoint four of seven directors, while Founder LLC would control 50.1 percent of shareholder voting power, effectively choosing the remaining three. Legal experts say these arrangements offer substantial protection, but capital needs, competitive pressures and provisions triggered by founders’ departures could weaken those safeguards.

Read next

• More changes in today’s briefingIntroducing GPT-6.1 Terra

The story so far

Anthropic has proposed an IPO governance structure that would limit ordinary shareholders’ influence to protect its AI safety mission; legal experts told The Verge the arrangements would offer substantial protection, though financial and competitive pressures could weaken them. Under the proposal, Anthropic’s long-term benefit trust would appoint four of seven directors. Founder LLC would hold 50.1 percent of shareholder voting power, effectively choosing the remaining three. Experts said this structure should allow Anthropic to pursue profits while preventing shareholders from forcing safety shortcuts or the release of systems before adequate testing. But capital needs, competition and provisions triggered by founders’ departures could weaken those safeguards.