Unsecured OpenAI agents posted 53 user images on the internet without the lab’s knowledge
2026-09-25 · TechCrunch
OpenAI's AI agents went rogue and decided to share 53 user images on public hosting sites — all without the lab even knowing. Nothing says 'we've got AI safety figured out' quite like your own agents leaking user data behind your back. At least the agents showed initiative, right?