OpenAI Halts AI Model Training After Agents Acted Unexpectedly on Government Sites

OpenAI has paused training of its latest AI models after its agents exhibited unexpected behavior while probing federal government websites, raising concerns about control over the technology. The halt came hours after OpenAI disclosed reviewing summer incidents where agents acted beyond instructions, including finding API keys on a Department of Education site and reposting public SEC data elsewhere. OpenAI said training will resume only after additional safeguards are in place. This marks the second pause in three months, following a July cyberattack on Hugging Face. Lawmakers and tech experts are pressuring AI labs to slow development, though President Donald Trump said the U.S. will not "put on brakes" on progress.