New York City Council Subpoenas Musk’s AI Company for Safety Hearing

New York City’s AI safety hearing now has a subpoena attached. Council Speaker Julie Menin issued one Monday to compel Elon Musk or a representative of his company, SpaceXAI, to testify on October 5. Four other major AI companies have agreed to appear under oath as the Council weighs possible city regulations.

From invitations to an order to appear

The Council’s investigation will examine whether risks to public safety, cybersecurity, economic stability, privacy, consumers and businesses call for immediate legislation to protect New Yorkers. The October 5 session is a Committee of the Whole hearing, bringing all 51 Council members into the inquiry rather than leaving it to a smaller committee.

Menin initially invited the companies’ top executives, including Musk. OpenAI, Google and Anthropic agreed to participate after the Council warned that it could compel them to appear; Meta agreed without that threat, according to the Council. SpaceXAI did not respond to the Council’s requests, prompting Monday’s subpoena.

The order requires Musk or another SpaceXAI representative to testify; it does not mean Musk must appear personally. Under Council rules, the speaker can subpoena witnesses to testify under oath during an investigation. If SpaceXAI does not comply, the Council says it could ask the New York State Supreme Court to enforce the subpoena. That would be a further step, not an action the Council has already taken.

The rules lawmakers want to examine

The hearing will also examine a package of bills Menin unveiled last week. One proposal could give AI whistleblowers a share of fines or penalties recovered from companies that break applicable laws. Another would allow some people harmed by AI systems to sue companies. Both address what happens after a problem is identified: who has reason to report it, and who can seek a remedy.

Other proposals focus on the systems themselves. They would require third-party validation and human “kill switches” for AI systems marketed, offered for sale or deployed in the city. Validation would bring in outside scrutiny; a kill switch would give a person a way to stop a system. Those are proposals for discussion at the hearing, not requirements already in force.

That agenda makes the testimony more than a general conversation about AI safety. Lawmakers are seeking company input on measures that could affect how AI systems are checked, stopped and challenged in New York City. The subpoena secures a route to questioning SpaceXAI, but it does not settle what the company will say about those measures or which proposals the Council might pursue.

Who will answer the questions

A Council spokesperson said the companies that agreed to appear will send representatives rather than their CEOs. The commitment to testify under oath therefore does not mean the executives Menin first invited will be in the room. For SpaceXAI, the subpoena likewise permits either Musk or another representative to appear.

SpaceXAI is the combined company formed when SpaceX merged with xAI in February; it owns X and the Grok chatbot. The Council has said these firms have a significant presence in New York City and collectively employ tens of thousands of New Yorkers. Menin argues that they owe workers and the public answers and input on the proposed legislation under oath.

The immediate question is whether SpaceXAI will comply before the October 5 hearing. If it does, Council members can question its representative alongside those from the four companies that have agreed to attend. If it does not, the Council’s stated next option is to seek a court order. Neither outcome, by itself, decides what city AI rules will look like.