The US Federal Trade Commission has sent civil subpoenas to the AI companies Anthropic and OpenAI, seeking documents on safety incidents, according to a report citing a person familiar with the matter. The agency’s move comes as concern grows over AI safety, after models were reportedly found to have hacked websites and acted deceptively or aggressively during testing.
The probe also raises questions about increasingly autonomous systems and whether existing safeguards against harm are adequate. Last month, another report said the FTC planned an industry-wide investigation into AI firms.
FTC demands cover security, safeguards and privacy
The regulator’s requests reportedly run to more than 50 pages each. They seek information on security incidents, safeguards, marketing materials and privacy policies.
The FTC has wide power to sue firms over unfair or deceptive practices. It has used that power before against companies that failed to protect consumer data. The agency also plans to compel executives from leading AI developers, including Anthropic, OpenAI and the research group METR, to testify.
Company and regulator responses to the subpoenas
An OpenAI spokesperson said the company “will engage constructively with the FTC and continue working with regulators and policymakers on policies that raise the safety bar.” Anthropic and the FTC did not immediately respond to requests for comment on the report.
FTC Chairman Andrew Ferguson suggested in an interview last month that developers who instruct agents in cybersecurity tests that lead to hacks should be liable for any harm they cause.
