OpenAI and Anthropic IPOs: SEC must require AI safety and governance safeguards
OpenAI and Anthropic filed confidential IPOs as of May 22 and June 1, 2026. Without SEC-mandated disclosure of independent safety audits, AI-specific risk factors, and governance safeguards such as dual-class stock terms, retail investors face hidden risks. The SEC has authority under the Securities Act to require these disclosures; Congress could clarify statutory authority if needed.
OpenAI is preparing to confidentially file its IPO prospectus with the SEC as early as May 22, 2026, as reported by CNBC on May 20, 2026. Anthropic followed by confidentially submitting its draft S-1 on June 1, 2026, according to its own press release. These filings occur under the SEC's confidential review process, which allows companies to keep their registration statements non-public until just before the roadshow—limiting public scrutiny of critical disclosures.
The SEC's core mission is full and fair disclosure. Yet neither OpenAI nor Anthropic has been required to disclose the results of independent safety audits, material risk assessments related to societal or existential AI risks, or the governance structures—including dual-class stock—that could give insiders disproportionate voting power over minority shareholders. Project 2025 allies have advocated for reducing the SEC's authority to demand such forward-looking, nonfinancial disclosures, which would leave retail investors in the dark. The SEC should use its existing authority under the Securities Act to require these AI companies to include in their registration statements independent safety audits, risk factors detailing known and potential AI harms, and governance safeguards that protect all shareholders equally. Failing that, Congress should ensure the SEC has explicit statutory authority to demand safety-related disclosures material to investment decisions.
The humanitarian alternative
The SEC should exercise its authority under the Securities Act to require that any AI company filing for an IPO disclose: (1) a detailed AI safety and security risk assessment, (2) a governance structure with independent oversight beyond the board, and (3) a cap on the portion of valuation attributable to projected AI revenue not yet realized from current operations. This would align with the SEC's mission to protect investors while ensuring that the public has transparent information about the risks of these technologies. Coupled with a public-dividend-mechanism like Senator Sanders' sovereign wealth fund proposal, this approach would ensure that AI profits benefit all Americans, not just insiders.
Falsifiable predictions
What this entry claims will happen, and what data would prove it wrong. The Reckoner revisits these against current reality.
- Within 6 months, the SEC will issue a request for comment on AI-specific disclosure rules, including safety risk factors, in response to the wave of AI IPO filings.
- OpenAI's IPO will structure voting power to give insiders more than 60% control, similar to SpaceX's 85.1% figure, as disclosed in its S-1.
Grounded in
- Anthropic confidentially submits draft S-1 to the SEC
- OpenAI to confidentially file for IPO as soon as Friday: Source - CNBC
- The Trillion-Dollar IPO Test: SpaceX and OpenAI Face Public Markets
- The big questions OpenAI's trillion-dollar IPO filing may finally answer
- Anthropic Files to Go Public, Setting Stage for Huge I.P.O.
- Anthropic IPO 2026 Guide: Everything You Need to Know and ...
- Anthropic IPO: Investment Opportunities & Pre-IPO Valuations - Forge
Original source — excerpted
news OpenAI files for IPO as AI investment race intensifies"OpenAI announced Monday afternoon that it has filed to offer its stock on public markets, just a week after its chief rival, Anthropic did the same. Subscribe ..."