Full Breakdown
Anthropic’s Founder LLC Gives Co-Founders Majority Voting Control Ahead of IPO
By Drooid · · How we work
Core Event – Governance Structure Unveiled in IPO Filing
Anthropic’s draft prospectus, filed in late September, creates a Founder LLC composed of the seven co-founders. A majority vote among members directs a single Class F share that carries 50.1 % of total voting power on key corporate matters, including the election of certain board directors. The publicly offered Class A common stock grants one vote per share; strategic-partner shares carry minimal voting weight. The filing warns that decisions made through this structure could conflict with short-, medium- or long-term financial interests of Class A shareholders and potentially depress their value.
Background & Context
Anthropic was launched in 2020 by a team that left OpenAI, positioning itself as a “virtuous” AI company focused on safety. The firm operates as a Public Benefit Corporation (PBC) under Delaware law, allowing leadership to balance investor returns with broader societal interests.
Key Figures & Groups
- Dario Amodei – CEO and co-founder.
- Daniela Amodei – President, board chair, and co-founder.
- Tom Brown – Chief Compute Officer, co-founder.
- Chris Olah – Researcher, co-founder.
- Ben Bernanke – Former Federal Reserve chair, trustee of the Long-Term Benefit Trust that appoints four board directors.
- Richard Fontaine – National-security expert, trustee of the Long-Term Benefit Trust.
Data & Statistics
- Voting control: One Class F share = 50.1 % of voting power; Class A shares = 1 vote each.
- Compensation (2025): Dario Amodei earned nearly $18 million; Daniela Amodei earned $16.4 million (largely stock and option awards).
- Equity pledge: All seven co-founders committed to donate 80 % of their personal Anthropic equity to charitable causes.
- Prospectus length: 261 pages total; ?80 pages devoted to risk factors.
- Model pricing (Sept 22): Opus 5.5 at $4 per M input tokens and $20 per M output tokens; Sonnet 5.5 at $2 per M input tokens and $10 per M output tokens.
Official Statements & Responses
An Anthropic representative declined to comment when approached by Reuters, and the company has refused comment to The Next Web.
Why It Matters – Investor Influence and AI Safety Governance
The Founder LLC arrangement concentrates decisive voting power in a small group of executives, potentially diminishing the voice of ordinary investors who hold Class A shares. Proponents argue that the structure aligns leadership with the public-benefit mission and enables rapid safety-first decisions, such as limiting the release of certain high-capability models. Critics note the filing’s warning that the arrangement could conflict with shareholders’ financial interests, raising questions about the balance between AI safety priorities and fiduciary duties.
Conflicting Reports & Gaps
Both Reuters and The Next Web report the same core governance details, but the prospectus itself has not been made public, limiting independent verification. The filing’s risk section outlines scenarios such as “self-protective behavior” and “ransom-like behavior,” yet no external audits of Anthropic’s safety controls have been disclosed.
What’s Next – Sunset Mechanism and Future Governance
The filing specifies that the special voting share will phase out once two or fewer founders (or their successors) remain, triggering a transition period in which voting power reverts to the broader shareholder base. No specific timeline is provided, but the mechanism suggests a future shift toward more conventional corporate governance as the founder cohort narrows.