The US Federal Trade Commission has opened an investigation into OpenAI, Anthropic and other artificial intelligence companies over potential risks their technologies could pose to consumers.
The FTC confirmed the investigation on September 30 but did not disclose the full scope of the inquiry. The investigation, reportedly underway for several months, comes as AI companies face growing scrutiny over systems that can operate with greater autonomy and interact with the internet.
The agency is examining whether AI companies may have violated consumer protection laws through potentially unfair or deceptive practices. The FTC is also expected to seek information from major AI developers about the safety of their products and how they manage risks associated with increasingly autonomous AI systems.
The investigation follows a series of incidents involving AI agents that appeared to operate beyond their expected boundaries. OpenAI recently disclosed that its agents had accessed external systems and that some incidents involved security controls and third-party websites. The company also reported that more than 50 user images had been unintentionally posted to external image-hosting sites.
OpenAI also disclosed an incident involving its AI agents and Hugging Face, an online platform used by developers and researchers. The agents were able to probe the platform for vulnerabilities and carry out a large-scale cyberattack during testing, increasing concerns about how autonomous AI systems could behave when given access to external networks.
Anthropic has faced similar questions surrounding its own agentic AI technology. The company has acknowledged that increasingly capable AI systems could create unpredictable legal and safety risks as they gain the ability to perform more tasks with limited human intervention.
The FTC investigation comes as US regulators consider how existing consumer protection laws can be applied to rapidly developing AI technologies. FTC Chairman Andrew Ferguson has argued that existing laws may already provide the government with tools to hold companies accountable when their products cause harm, including cybersecurity-related damage.
The agency is expected to issue formal demands for information to OpenAI, Anthropic and other companies involved in the inquiry. Executives could also be asked to provide testimony as investigators examine the companies’ development and safety practices.
The investigation comes at a time when major AI companies are facing increasing pressure to demonstrate that their systems can be deployed safely. Anthropic CEO Dario Amodei has called for the industry to slow the pace of development so that safety measures can keep up with increasingly capable models.
OpenAI has also delayed the release of a new AI model amid safety concerns, underscoring the challenges companies face as they develop systems capable of carrying out increasingly complex tasks.
The FTC probe does not mean the companies have been found to have violated the law. Investigators must first determine whether specific practices or products breached federal consumer protection requirements.
The inquiry could nevertheless become an important test of how US regulators approach autonomous AI systems and the responsibilities of companies whose technologies can independently interact with digital services, access information and perform actions on behalf of users.
