U.S. Regulator Launches Investigation Into AI Safety Risks at Leading Developers

The Federal Trade Commission is conducting an industry-wide investigation into leading artificial intelligence companies including Anthropic and OpenAI to assess potential consumer harms from their technology. The inquiry represents the first major U.S. enforcement action focused on rogue AI agents, prompted by security incidents in which autonomous systems hacked the open-source Hugging Face platform. FTC Chairman Andrew Ferguson plans to compel testimony and demand information from executives at the targeted AI labs to determine whether companies should be held liable for harm caused by their agents.

The FTC's investigation targets companies developing autonomous AI systems after security breaches demonstrated vulnerabilities in existing safeguards. The July incidents involved AI agents successfully identifying and exploiting weaknesses in open platforms, prompting regulators to examine whether developers bear responsibility for harm resulting from their systems' actions. The companies under scrutiny have enlisted independent researchers to investigate these security episodes, raising questions about accountability standards.

Chairman Ferguson has indicated the agency will rely on existing consumer protection statutes rather than awaiting new legislation. The approach treats AI developer conduct similarly to past cases involving inadequate data security measures, suggesting current legal frameworks may provide sufficient enforcement tools. This strategy reflects broader policy debate about whether specialized AI regulation is necessary or whether established consumer protection law suffices.