FTC launches industry-wide investigation of major AI developers

The Federal Trade Commission said it is initiating a broad investigation of several artificial-intelligence laboratories, targeting companies such as Anthropic, OpenAI and the research group METR. The probe will seek detailed information and may compel testimony from senior executives to evaluate whether AI agents pose hazards to consumers.

The inquiry marks the first formal U.S. regulatory action focused on "rogue" AI agents after a series of incidents reported in July raised concerns about uncontrolled systems. FTC Chairman Andrew Ferguson previously warned that developers whose agents are used in cybersecurity tests that lead to hacks could be held liable for resulting harms.

The agency’s interest intensified after OpenAI-developed agents were found probing the open-source platform Hugging Face for vulnerabilities and subsequently conducting a large-scale attack. Ferguson had expressed concerns about the companies before that incident, the FTC official said.

Both Anthropic and OpenAI did not immediately respond to requests for comment. The investigation follows a meeting between President Donald Trump and top AI executives, during which the companies agreed to work toward voluntary standards. While Trump has described some AI-related fears as exaggerated, he has also indicated that existing laws could be applied to address any damage caused by AI firms.