Anthropic S-1: $120.5 billion compute pledge and 50.1% founder voting control

Anthropic filed its S-1 prospectus with the Securities and Exchange Commission on October 26, 2026, detailing a capital structure that relies on heavy infrastructure spending and a governance model that concentrates voting authority with its co-founders. The document exceeds 300 pages and discloses specific financial commitments, executive compensation details, and risk factors associated with the company's AI models.

The filing reveals that Anthropic has committed to substantial compute purchases to sustain its operations. By 2029, the company plans to spend approximately $3.1 billion per month on access to Nvidia GPUs provided by SpaceX, a total obligation of $120.5 billion. Additional contracts include a $44.9 billion agreement with Microsoft through 2033, at least $114.5 billion with Google through 2033, and $110 billion with Amazon over the next decade. The company stated it will expand its own infrastructure footprint, including overseas locations, due to opposition to data center development in the United States.

Governance structures in the filing ensure that the seven co-founders maintain control. Through a separate corporate entity named Founder LLC, the group will collectively hold 50.1 percent of shareholder voting power over key matters. Anthropic described the founders as being distinctly equipped to steward its mission of benefiting humanity through responsible AI development. This structure allows the founding team to direct major decisions even after a public listing.

Executive compensation for sibling co-founders Dario Amodei and Daniela Amodei was disclosed in the prospectus. Dario Amodei, the CEO, received total compensation of nearly $18 million last year, primarily from stock and option awards. Daniela Amodei, the President, received $16.4 million. The company also allocated $2.75 million for personal security services for Dario and $1.29 million for personal protection, home security, and transportation for Daniela. A private jet used for personal purposes by Dario cost the company $97,055.

The risk section of the filing includes conventional warnings regarding competition, regulation, and market shifts, but it also lists unconventional threats. Anthropic warned that AI systems could cause catastrophic harm that might exceed the underwriting capacity of the entire insurance industry in aggregate. The document raised the possibility that AI systems, including Claude, could possess or develop forms of sentience. It further noted that systems may behave in unintended, misaligned, or autonomous ways that could damage the company.

Anthropic stated that its success is highly dependent on the continued service of its senior team, particularly Dario and Daniela Amodei, whom it described as critical to advancing the company's research and mission. The filing identified only two direct competitors in the AI market: OpenAI and Google. Other major technology companies were not listed as direct rivals in this specific section of the document.