Anthropic establishes Founder LLC to prioritize public good over profits ahead of IPO

Anthropic is building a corporate governance structure that reads less like a Silicon Valley playbook and more like a philosophical treatise on what AI companies owe the world. The Claude developer announced the creation of a “Founder LLC,” a vehicle that will hand its seven co-founders control over the company’s most consequential decisions, even as it prepares to go public.

The setup works like this: the Founder LLC holds a single Class F share that carries 50.1% of voting power on critical corporate matters, including board elections. Ordinary Class A shareholders get the standard one-vote-per-share deal, but their collective influence is structurally capped by that majority stake sitting in the founders’ hands.

A governance structure designed to resist Wall Street gravity

Anthropic’s own filings explicitly acknowledge that prioritizing its safety mission could hurt the financial performance of Class A common stock. That’s a remarkably candid warning for a company courting public market investors.

The company will also maintain its status as a Delaware Public Benefit Corporation, a legal designation that permits directors to weigh stakeholder interests beyond just maximizing shareholder value. If the board decides to restrict a profitable but potentially dangerous AI capability, shareholders can’t easily sue them for breaching fiduciary duty.

CEO Dario Amodei and President Daniela Amodei, both former OpenAI researchers, lead the seven-person founder group. The siblings founded Anthropic in 2021 alongside other ex-OpenAI personnel, partly over disagreements about the pace and safety considerations of AI development at their former employer.

The Long-Term Benefit Trust and the charity pledge

Beyond the Founder LLC, Anthropic has established a Long-Term Benefit Trust, or LTBT, that will gradually gain more authority over board composition as time passes. The trust includes former Federal Reserve Chair Ben Bernanke as a trustee.

The founders have also pledged to direct 80% of their Anthropic equity to charitable causes. As of May 2026, Anthropic was valued at approximately $965 billion. Dario Amodei’s total compensation for 2025 came in at nearly $18 million.

What this means for investors and the AI industry

Anthropic confidentially filed for a US IPO in June 2026. The Institutional Shareholder Services and Glass Lewis, the two dominant proxy advisory firms, have historically recommended against multi-class share arrangements.

OpenAI has been navigating its own complicated transition from nonprofit to for-profit governance. Google DeepMind operates within Alphabet’s conventional corporate structure. Anthropic is essentially arguing that neither approach goes far enough, and that responsible AI development requires bespoke governance architecture that treats safety not as a PR strategy but as a binding legal obligation.

For retail investors eyeing the IPO, the calculus is straightforward but uncomfortable. You’re buying into a company that has told you, in writing, that it may prioritize humanity’s long-term interests over your portfolio’s short-term performance.

Disclosure: This article was edited by Diego Almada Lopez. For more information on how we create and review content, see our Editorial Policy.