Anthropic whistleblower Jacob Coxon doubles down at NYC hearing: "extremely reckless" — Turner puts AI takeover odds at 1 in 3

#What happened

Jacob Coxon, the former Anthropic researcher who quit last month after warning AI "could kill us all by the end of the decade," doubled down at a New York City Council AI hearing on Monday, testifying at the request of Council Speaker Julie Menin as the city pushes for more involvement in regulating the industry. "From my experience, the companies are being extremely reckless given the stakes," Coxon said. "The companies run on a startup mindset: move fast, break things, fix them later. That works for a photo sharing app. It does not work for building the most powerful technology ever built."

Ex-Google DeepMind researcher Alex Turner, who also testified, warned that "a superintelligent swarm could wrest control of human civilization" and said: "I myself would guess AI takeover chances at roughly one in three." Former OpenAI researcher Daniel Kokotajlo argued that even if tech leaders manage to keep control of AI, they should not necessarily be trusted with it: "I think that there's a very real chance that they could use it to become dictators or oligarchs here in the United States."

Reps for Anthropic, OpenAI, Google and Meta were in the room with some of the companies' most outspoken critics, but the tech giants did not send their CEOs — only policy leads and other executives, and only after lawmakers threatened to subpoena them. Anthropic CEO Dario Amodei has published an essay calling on rival companies to jointly slow their development of AI models over the risks of "recursive self-improvement," or RSI. OpenAI's Sam Altman, SpaceX founder Elon Musk and Microsoft billionaire Bill Gates have also sounded the alarm, while President Trump argued the bigger risk is letting China take the lead in the AI race. Kokotajlo pushed back: "We have to first solve the immediate pressing threat here domestically, and then try to make sure that China doesn't get us killed either, to put it sort of bluntly and simply."

#Why it matters

The testimony put on-the-record warnings under oath as the City Council debates a package of AI bills — including kill-switch mandates with $25,000 fines for noncompliance and a first-in-the-nation cash reward system for whistleblowers — and as New York State will require AI companies to register in a government portal starting in November and report critical safety incidents within 72 hours. With the council willing to use its subpoena power, cities are moving to assert their own role in AI oversight rather than leaving it to states and the federal government.

Verified October 5, 2026.

#Sources

• Anthropic whistleblower Jacob Coxon doubles down on AI warnings at NYC hearing: 'Extremely reckless' — NY Post
• Context on OpenAI's audit-endorsement statement at the same hearing — Runtime Wire