FTC opens AI safety and antitrust probe into OpenAI and Anthropic

The US Federal Trade Commission has launched an investigation into safety and antitrust issues at OpenAI and Anthropic, marking a sharp escalation in Washington's oversight of the artificial intelligence boom.

The federal regulator is looking into how the country's two leading AI developers manage consumer risks and safety standards. The inquiry also extends to METR, a prominent non-profit safety auditor that tests frontier AI models before they are released to the public.

The investigation comes at a sensitive time for the industry. OpenAI recently halted the launch of its GPT-6.1 Astra model over safety flaws, exposing the technical friction of rushing advanced systems to market.

Why the FTC is stepping in

The FTC is examining whether OpenAI and Anthropic have misrepresented the safety and capabilities of their AI models, and how they protect users from consumer risks. The inclusion of METR in the inquiry suggests that regulators are also looking closely at the third-party auditing systems that AI companies rely on to validate their products.

The antitrust element of the probe is equally significant. Both startups have formed deep alliances with major tech firms to fund their massive computing needs. OpenAI is backed by Microsoft, while Anthropic has received billions of dollars from Amazon and Google. FTC Chair Lina Khan has previously warned that such partnerships could allow a small group of dominant tech companies to control the future of AI development.

A shift away from self-regulation

For the past two years, the AI industry has largely operated under a system of voluntary safety commitments. Leading developers, including OpenAI and Anthropic, previously agreed to White House standards on safety testing and security.

However, regulatory patience in Washington appears to be wearing thin. Khan has repeatedly stated that existing consumer protection and antitrust laws apply to AI companies, warning that the agency will act if developers engage in deceptive marketing, violate data privacy, or engage in anticompetitive behaviour. Neither OpenAI nor Anthropic has commented on the investigation.