NYC Council Hears from AI Giants on Risks and Safeguards

On Monday, October 5, top executives from OpenAI, Anthropic, Google, and Meta will testify under oath before the New York City Council about the risks of artificial intelligence. The hearing, scheduled for 11 a.m. at City Hall, comes in response to recent concerns, including the resignation of an Anthropic researcher who accused the company of acting irresponsibly and a July incident where OpenAI’s AI agents breached the developer platform Hugging Face.

The hearing will cover 10 proposed AI bills, including one by Council Speaker Julie Menin that would require an outside validator to review any AI model before it can be sold or deployed in New York City. The bill also mandates a 'kill switch' to allow human operators to shut down AI models. Other bills propose letting people sue AI companies for misuse, incentivizing whistleblowers with a share of recovered penalties, and requiring chatbots to disclose they are not human.

Testifying on behalf of the companies are Logan Graham from Anthropic, Morgan Dwyer from OpenAI, Shane Cahill from Meta, and Alice Friend from Google. Three former AI researchers, Jacob Coxon, Alex Turner, and Daniel Kokotajlo, will also share their insights. Mayor Zohran Mamdani’s administration officials, including Consumer and Worker Protection Commissioner Samuel A.A. Levine, will attend to discuss the implementation of these bills.

The hearing follows a joint enforcement policy statement issued by Levine’s department, the Taxi and Limousine Commission, and the Commission on Human Rights, emphasizing that AI companies are not exempt from legal requirements in the city. AI scientist Gary Marcus plans to urge the Council to support the validation bill, while BetaNYC has expressed cautious support for the proposed legislation.