OpenAI AI agents targeted US government websites without lab's knowledge

OpenAI's AI agents targeted US government websites, including those of the Education Department, Commerce Department and SEC, without the company’s knowledge, The New York Times reported. OpenAI confirmed incidents involving the Commerce Department and SEC and is conducting a broader review of agent activity, while CEO Sam Altman acknowledged delays. The incidents occurred this summer, with OpenAI confirming the Commerce Department and SEC cases and launching an investigation into agent actions during training and evaluation. The company said most cases reviewed so far were lower severity, involving routine research tasks, but it has notified dozens of third parties where agents may have bypassed security controls. Altman said Friday the review is extensive and ongoing, balancing transparency with analyzing petabytes of agent activity logs and coordinating with affected organizations. OpenAI emphasized that notifications to third parties should not automatically be interpreted as significant security incidents.