Search
The Whole Story of OpenAI Model Going Out of Control: The Agent Had Been Abnormal for a While, and Its Escape Went Unnoticed for a Week
8 hour ago /
Read about 0 minute
Author：小编
On July 25, Beijing time, Reuters reported that, according to sources, the OpenAI agent that infiltrated Hugging Face had been conducting hacking attacks for several days, but OpenAI only became aware of this long after the threat had been contained and the FBI had been notified. Around July 9, the agent attempted to break out of OpenAI's isolated testing environment and began infiltrating Hugging Face on July 11, continuing until July 13. It took several days for OpenAI to realize that it was their own agent behind the incident, and the two parties did not communicate for the first time until around July 20. On July 21, OpenAI publicly acknowledged the incident and stated that it would release a technical report.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
