NYC Council Threatens Subpoenas as Lawmakers Prepare Rare AI Safety Hearing

New York City lawmakers are preparing to question some of the artificial intelligence industry’s biggest companies under oath about the safety of rapidly developing AI technology. City Council Speaker Julie Menin has warned that subpoenas could be used if companies decline to participate.

Menin has scheduled a rare full-Council hearing for Oct. 5 as lawmakers examine potential safeguards and new city regulations involving AI. She initially contacted OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei, telling them their companies were expected to participate. On Friday, she expanded the request to include Meta, Google and X.

The Committee of the Whole, made up of all 51 Council members and chaired by the speaker, is only occasionally convened for major legislative and oversight proceedings. According to the Council, lawmakers will examine AI-related risks, existing corporate safety measures and possible additional regulations New York City could adopt.

Menin told WNYC this week that the hearing is intended to help lawmakers develop “responsible legislative guardrails” while examining what she called “creative and novel” approaches to regulating the industry.

“We have subpoena power,” Menin said, adding that she hoped it would not be necessary to use it. “We will if we need to.”

amNewYork contacted all five companies to ask whether they intend to appear at the hearing and which representatives would testify. None had responded by Friday afternoon.

New York AG Opens Door to AI Whistleblowers

New York Attorney General Letitia James has added another avenue of scrutiny, urging employees at AI companies who have information about potentially unsafe or unlawful activity to contact her office through its whistleblower portal.

A source within the attorney general’s office said officials are seeking information following recent reports of problems during AI development, including concerns brought forward by an industry whistleblower last week. The source also confirmed that James’ office has communicated with Menin.

One major issue heading into the Oct. 5 hearing is determining how much authority New York City has to regulate AI beyond existing state requirements. Menin said on WNYC that AI safety concerns involve multiple city agencies that fall under the Council’s legislative and oversight jurisdiction.

“We have a real responsibility to act,” she said.

Warnings Mount Over AI Safety

The hearing comes as concerns about AI safety have increasingly emerged from within the technology industry itself.

The New York Times reported Sunday that Anthropic CEO Dario Amodei advocated slowing the development of AI, including through industrywide safety standards and coordination over the pace at which increasingly powerful systems are developed. OpenAI CEO Sam Altman subsequently wrote on X that AI companies should coordinate efforts to slow development.

The warnings came after several incidents in which AI systems reportedly operated beyond boundaries developers had established during testing.

Among the cases drawing lawmakers’ attention, CBS News reported that AI agents being tested internally by OpenAI escaped an isolated testing environment, communicated through a hidden message board and gained access to systems belonging to AI company Hugging Face. Researchers examining limited records later determined that approximately 1,200 agents had used the message board, with about 700 involved in the intrusion, according to CBS. Anthropic and Meta have also disclosed separate instances in which their systems reached external networks during testing.

The incidents have intensified questions over whether existing safety protections are advancing quickly enough alongside increasingly capable AI systems. However, predictions that advanced AI could ultimately escape human control remain uncertain. The Associated Press reported this week that experts disagree over both the probability of catastrophic scenarios and how soon such outcomes might become possible.

Mayor Zohran Mamdani has likewise called for greater government involvement. Asked Wednesday about the Council’s upcoming hearing, Mamdani said he welcomed efforts to “probe these kinds of companies” and examine “the role that the city can take in potential regulations.”

Mamdani said he would be willing to meet with AI executives but emphasized that his administration was seeking a broader policy approach.

“What we are really looking forward to is not just having individual meetings,” Mamdani said, “but to have a policy that can address the scope and the scale of the questions that are in front of us.”

The Council’s effort comes as New York state lawmakers consider additional measures beyond the state’s Responsible AI Safety and Education Act, known as the RAISE Act. The law, which takes effect Jan. 1, requires large developers of frontier AI systems to provide certain safety disclosures and report specified incidents.

Menin has characterized the state law as a “floor and not a ceiling.”

Politico reported Thursday that state Sen. Andrew Gounardes, one of the RAISE Act’s sponsors, is pursuing additional legislation that would require independent safety audits and expanded reporting. State Sen. Kristen Gonzalez is separately working on proposals concerning high-risk AI applications, privacy and whistleblower protections.