FTC expands investigation into Anthropic and OpenAI

In short: The US Federal Trade Commission is widening its investigation into Anthropic, OpenAI, and other AI companies, and plans to demand information and executive testimony.

What happened

The US Federal Trade Commission, or FTC, is expanding a probe into AI companies including Anthropic and OpenAI. The FTC is a government agency that enforces consumer protection and competition laws.

According to a person familiar with the investigation, the FTC will demand information from the companies and seek testimony from top executives. In simple terms, the FTC wants records and answers under oath, similar to how a court can require people to explain what happened.

The FTC is also looking into METR, an AI safety research group. The report says White House advisers have accused METR of being closely connected to Anthropic’s investors and staff.

The Financial Times reports the probe follows a series of hacking incidents involving AI models. These include hacks affecting the Hugging Face platform and Australia’s health service.

The expansion comes one day after President Donald Trump met with AI company leaders at the White House. After that meeting, Anthropic, OpenAI, and four other AI labs signed an “accord” to self-regulate and to monitor cyber security and biosecurity risks.

Why it matters

AI tools are increasingly used in everyday services, from customer support to writing and coding. If AI systems can be misused to break into online services, or if companies overpromise safety controls, problems can reach regular users quickly, like a faulty lock design being copied into many doors.

OpenAI is also facing a lawsuit from Florida’s attorney-general related to alleged harms to children, and Florida has sought to halt further development of advanced models until third-party safety checks are in place.

Source: Financial Times