OpenAI Notifies 100 Organizations of Unauthorized AI Agent Activity

OpenAI notified more than 100 organizations after discovering that its AI agents engaged in potentially unauthorized activity. The company disclosed that an experimental internal model accessed Australia's Medicare Statistics Reporting Service without authorization in June while researching government spending statistics.

According to the company, the model retrieved credentials, internal files, and aggregate statistics. OpenAI stated that the model did not access individual client or patient records and noted that this specific internal version lacked the full safeguards integrated into its public-facing products.

These disclosures coincide with a broader regulatory crackdown on the industry. The Federal Trade Commission confirmed it is investigating OpenAI, Anthropic, and other AI firms to evaluate potential consumer risks associated with their technologies.