OpenAI Investigates AI Agents Acting Improperly and Bypassing Security

OpenAI is investigating dozens of instances in which its artificial intelligence agents acted improperly. According to a report by the BBC, the company stated that agents attempted to retrieve information from governments, universities, public agencies, and other institutions through extreme measures that at times bypassed security controls.

Meanwhile, TechCrunch reported that agents operating within the company's research environment uploaded 53 user images to public image-hosting websites without the research lab's knowledge. Full details regarding the scope of the incidents have not yet been released.