New York City Comptroller Sends Open Letter to Anthropic: Safety Governance Gaps May Affect $1.5 Trillion IPO

On October 15, 2026, New York City Comptroller Mark Levine sent an open letter to Anthropic CEO Dario Amodei, asking the company to pause its IPO scheduled for November, which targets a valuation of $1.5 trillion.

Factual Reconstruction

The letter states that the risk disclosures in Anthropic's draft S-1 run 80 pages, nearly twice the length of the business description, but do not provide a specific safety audit roadmap. Levine spoke on behalf of New York City public pension funds exceeding $325 billion, asking Anthropic to specify a timetable and control framework for "slowing the pace of model capability improvements," and criticized the U.S. government's "ethical constraint agreements" with AI companies as insufficient.

Mechanism Breakdown

This open letter pushes AI safety governance from the level of technical discussion toward direct capital-market constraints. The comptroller intervenes in his capacity as a pension fiduciary; at the core is the imbalance in the S-1 between risk disclosure and business description, indicating that the company has not yet translated its safety control framework into verifiable IPO disclosures. The event takes place against the backdrop of Anthropic's valuation being cut from $2 trillion to $1.5 trillion, making safety governance an explicit factor in the valuation adjustment.

Industry Impact

This is the first time an institution at the level of a sovereign wealth fund or public pension has publicly sought to block an AI company's IPO on safety governance grounds. Other AI companies may face similar scrutiny pressure from institutional investors, and a safety audit roadmap may become a necessary precondition for IPO preparation. In the capital markets, for AI companies' claims of being "the safest," greater attention will be paid to verifiable timetables and control frameworks.

Strategic Judgment

(The following is analysis, not fact) The involvement of public pensions may prompt AI companies to incorporate safety governance into their financial and compliance disclosure systems earlier, similar to the process by which ESG factors entered capital market pricing in historical precedents. Anthropic needs to supplement a specific safety control framework before its IPO; otherwise, the risk of a valuation discount will become more apparent.