New York City’s confrontation with the AI industry had been building for months as concerns over increasingly capable models spread from research circles into politics. In September, industry researchers warned that leading labs could be pursuing systems capable of catastrophic harm, including human extinction, while former Anthropic employee Jacob Coxon said companies were “gambling with our lives.”¹

That pressure reached City Hall on Monday, when the Council convened a rare Committee of the Whole hearing. All 51 members were expected to scrutinize Anthropic, OpenAI, Google and Meta over AI’s threats and possible legislation. Speaker Julie Menin said the firms owed the public answers under oath; Anthropic, OpenAI and Google agreed to attend only after the Council raised the prospect of subpoenas.²

Coxon, joined by former Google and OpenAI employees, cast the dispute as a warning about corporate culture. “Move fast, break things” may be tolerable for a photo-sharing app, he told lawmakers, “It does not work for the most powerful technology ever built.”³ Their case was not simply that AI can make mistakes, but that the commercial race is outstripping the safeguards meant to contain its most severe risks.

The companies came to defend their practices, but the hearing exposed the gap between broad commitments and measurable assurances. Asked to put a number on AI risk, OpenAI policy executive Morgan Dwyer said she did not know, while maintaining that no level of risk was acceptable to the company. Public Advocate Jumaane D. Williams called that a “God-awful, fantastically terrible answer.”³

The competing accounts shared one point: advanced AI carries stakes beyond ordinary consumer technology. They diverged sharply on whether company-led safety work can keep pace—or whether lawmakers must force the issue.