OpenAI and Anthropic face investigation as rouge AI agents raise security concerns, check details

OpenAI and Anthropic are facing an investigation by the US Federal Trade Commission (FTC) over potential AI risks. A senior FTC official told Reuters that the agency is carrying out an industry-wide probe into AI companies to uncover the possible dangers their technology could pose to consumers. The move comes after a series of incidents involving agentic AI systems raised concerns about their ability to carry out unauthorised actions.

The FTC plans to send formal demands for information to leading AI developers, according to the report. It may also compel executives from these companies to provide testimony. The companies named in the investigation include Anthropic and OpenAI. The agency also plans to seek information from the research group METR.

Also read: Google introduces Gemini 4 Argon AI model, claims it outperforms GPT 6 Astra on several benchmarks 

METR has worked with both Anthropic and OpenAI on independent investigations into security incidents involving their AI agents. 

FTC Chairman Andrew Ferguson had already raised concerns about AI companies before OpenAI’s AI agents hacked Hugging Face. The incident increased concerns around the risks of AI agents and added urgency to the FTC’s probe. Ferguson had also said last week in an interview with Reuters that AI developers could be held responsible when agents used for cybersecurity testing end up causing hacks. 

Also read: Vivo X Fold 6 vs Samsung Galaxy Z Fold 8 Ultra: Price in India, camera, processor and other expected differences  

Ferguson had also argued that the US should first use existing laws to address problems linked to AI instead of creating new regulations. The FTC already has powers to take action against companies over unfair or deceptive practices. It has previously used these powers against companies that failed to take reasonable steps to protect consumer data.

Earlier this week, US President Donald Trump met with leading AI executives, with companies agreeing to establish voluntary standards. 

Anthropic also warned about the legal risks linked to agentic AI in documents prepared for its stock market debut, as per the report. The company described these risks as significant and unpredictable.

Also read: US President Donald Trump meets AI giants: From self regulation pact to new name for AI, here are 3 key takeaways