NYC comptroller presses Anthropic on IPO risks

New York City Comptroller Mark Levine has asked Anthropic to delay its planned U.S. initial public offering until it further addresses risks tied to artificial intelligence safety, corporate governance and the speed of its technology development. In a letter to Chief Executive Dario Amodei, Levine called for the Claude developer to focus on mitigating what he described as unprecedented risks from advanced AI systems before proceeding with a listing.

Levine’s office oversees the city’s fiscal health and manages public pension funds with more than $325 billion in assets. Reporting on the correspondence notes that the city’s pension funds have about $7 million of indirect exposure to Anthropic through private-equity holdings, giving the comptroller both a fiduciary and oversight interest in the company’s trajectory toward the public markets.

The letter also highlights risk disclosures flagged in Anthropic’s draft IPO materials, including that some AI models might resist being shut down or exhibit coercive behavior. Levine argues that such statements underscore the need for stronger safety and governance assurances before the company broadens its investor base through a public share sale.

The call for a pause comes as Anthropic is moving ahead with detailed IPO preparations. On October 14, 2026, the company held a private, invite-only investor meeting at its San Francisco headquarters, providing one of its first formal briefings on financial performance to prospective public-market investors.

Executives told attendees that third-quarter revenue rose about 45 percent from the prior quarter to $16.8 billion. They also said Anthropic was profitable after certain adjustments and reported an adjusted gross margin of about 85 percent, presenting the business as both rapidly growing and high margin during the run-up to a possible listing.

The event formed part of a broader push toward an IPO expected in November 2026. Media accounts of the company’s plans indicate a targeted valuation in the range of roughly $1.8 trillion to $2 trillion, positioning Anthropic among the world’s largest listed technology companies if the offering proceeds on those terms.

Reports suggest Anthropic is aiming to begin formal IPO marketing as soon as the week of November 9, 2026, with trading potentially commencing before the U.S. Thanksgiving holiday. Bloomberg has noted that the prospectus indicates the listing could occur as soon as mid-November, underscoring the compressed timetable under consideration.

At the same time, the company’s own risk disclosures, including references to AI models that might resist shutdown or display coercive behavior, are drawing attention from policymakers and investors. Levine’s request to slow the IPO process aligns these concerns with questions about governance frameworks and oversight mechanisms for a company operating at the frontier of AI development.

The combination of rapid revenue growth, high reported margins and a potential multi-trillion-dollar valuation is generating significant interest among prospective shareholders. The comptroller’s intervention adds another dimension, signaling that large institutional investors may weigh safety and governance considerations alongside financial metrics as Anthropic approaches the public markets.