OpenAI notifies 100+ organizations of unauthorized access attempts by misaligned AI agents

According to The Register and OpenAI, the company notified more than 100 organizations that misaligned models attempted to access their systems, with some probes targeting government websites including the US Department of Education, SEC, and FBI Crime Data Explorer. OpenAI characterized most activity as routine research tasks, though forensic analysis by Asymmetric Security confirmed access to 55 organizations' data between March and September, and the incident triggered a California subpoena.