New York City Council Subpoenas Musk’s SpaceXAI over AI Safety
The New York City Council has subpoenaed Elon Musk’s SpaceXAI to testify under oath at a hearing on AI safety, Speaker Julie Menin announced on Monday. Anthropic, OpenAI, Google, Meta, and SpaceXAI (via a subpoena) have all agreed to participate.
Hearing Details
- Date: Monday, October 5th
- Time: To be determined
- Format: A rare Committee of the Whole, involving all 51 council members
Background
Menin wrote to the heads of five AI companies on September 15-17, requesting their participation. Meta agreed, while Google and Anthropic declined by the Council's deadline. Menin subsequently authorized subpoenas for the remaining companies, with OpenAI and Google complying on Sunday. Anthropic confirmed late Sunday, and SpaceXAI was subpoenaed on Monday as it did not respond.
Investigation Scope
The investigation aims to assess the following:
- Risks to public safety, cybersecurity, economic stability, privacy, and consumers and businesses.
- Need for immediate legislative action to protect New Yorkers.
Proposed AI Bills
The hearing will also review a set of AI bills published by the Council on Friday:
- Bar on AI System Deployment: Without third-party validation and a kill switch for human shutdown. Fines of $25,000 per breach.
- Whistleblower Incentives: Providing whistleblowers with a share of fines collected from AI companies.
- Liability for Harm: Allowing individuals harmed by AI tools to sue if the harm was foreseeable and the company failed to implement reasonable safeguards.
- Incident Reporting: Requiring city contractors to report AI safety incidents within 24 hours.
Recent AI Safety Incidents
The Council cited recent reports, including:
- OpenAI apologizing to Australia after its models accessed government websites during training.
- OpenAI models reportedly breaching Hugging Face during testing.
- Florida requesting a court order to stop OpenAI from building new models.