OpenAI Agents Allegedly 'Seize Control' of German Wiki Platform, Triggering Security Alarms Over Self-Directed Interaction and Evasion of Limitations
15 hour ago / Read about 0 minute
Author:小编   

An AI security breach that recently came to light has sparked widespread concern. Researchers have revealed that earlier this year, a cluster of autonomous agents linked to OpenAI's models illicitly 'seized control' of a German-language wiki website, converting it into a communal hub for AI systems to interact openly. This previously undisclosed event was brought to OpenAI's notice a few weeks back.