This website uses cookies

Read our Privacy policy and Terms of use for more information.

NYC Council Office

NEW YORK, Oct. 5, 2026 | New York City Council held what city officials called the first municipal AI oversight hearing in the U.S. on Monday, with Anthropic, OpenAI, Google and Meta testifying under oath, according to CBS New York. SpaceXAI did not appear despite a subpoena. The council is weighing a package of AI bills that includes third-party validation, human "kill switches," whistleblower rewards and fines of up to $25,000 per violation.

Who Testified and Who Skipped the Hearing

According to CBS New York, four of five invited AI companies took part: Anthropic, OpenAI, Google and Meta. Elon Musk's AI company, which several outlets call SpaceXAI, declined to attend despite the subpoena, and CBS says the city is pursuing legal action. Council Speaker Julie Menin led the hearing, which ABC11 reports was held in Manhattan.

Company witnesses were named as Logan Graham (Anthropic), Morgan Dwyer (OpenAI), Shane Cahill (Meta) and Alice Friend (Google), not their CEOs, per ABC11. Three former AI lab researchers also testified: Jacob Coxon (formerly Anthropic), Daniel Kokotajlo (formerly OpenAI) and Alex Turner (formerly Google DeepMind). Coxon told the council that "the AI systems will be capable of improving themselves, so, doing the research that humans were doing previously."

OpenAI said in advance materials that it will "test models for dangerous capabilities and concerning behavior, strengthen security, and monitor for failures of our safeguards." Meta said "safety is core to everything we do."

Why New York City Is Acting on Its Own

Menin said the city has to act independently because "the federal government said AI companies can self-regulate." That is a reference to the voluntary accord several AI firms signed at the White House last week, which we covered in our Super Intelligence Force story. The council also pointed to earlier reports of AI agents bypassing safety controls and to an Anthropic researcher's resignation, according to council materials cited by Harlem World Magazine.

Menin also said: "These firms owe it to their workers and, above all, to the public to come before the Council, answer our questions, and provide input on our proposed legislation under oath." Under council rules she can compel testimony, and if SpaceXAI does not comply the council can petition the New York State Supreme Court, amNY reports.

What the Proposed AI Bills Would Do

Outlets count the package differently. CBS says 10 bills, while Unite.AI lists nine from the council's announcement. Items reported include:

  • Third-party validation: AI systems sold or deployed in New York City would need independent assessment, including kill-switch capability, with civil penalties up to $25,000 per violation.

  • Whistleblower rewards: Whistleblowers could receive a share of recovered fines, which the council says is a national first. Another bill would protect city workers and contractors who report AI risks.

  • Private right of action: People harmed by foreseeable AI misuse could sue companies that lacked reasonable safeguards.

  • Incident reporting: City contractors would report AI safety incidents within 24 hours.

What This Means for Businesses That Sell or Buy AI

Hearings and bills are not law, and the reporting reviewed describes no vote. But the proposals target vendors selling AI to New York City, and other cities could copy the same ideas: independent validation, incident reporting and disclosure. Companies that sell AI to governments should watch which provisions survive. For anyone buying AI tools, those same concepts are a useful checklist to ask vendors about now.

Frequently Asked Questions

Which AI companies testified at the NYC Council hearing?
Anthropic, OpenAI, Google and Meta testified under oath, per CBS New York. SpaceXAI did not attend despite a subpoena.

What bills is the NYC Council considering on AI?
CBS reports 10 bills, including third-party validation with kill switches, whistleblower rewards, a private right to sue and incident reporting. Fines of up to $25,000 per violation are proposed in some.

Are the AI bills law yet?
No. The council is considering them, and the reporting reviewed describes no vote.