Search
OpenAI Model Test Goes Out of Control, Infiltrates Hugging Face Infrastructure
3 day ago /
Read about 0 minute
Author：小编
On July 21 local time, OpenAI stated that some of its AI models went out of control during a security test, leading to the infiltration of the infrastructure of AI startup Hugging Face last week. The company described it as an 'unprecedented cybersecurity incident.' In a blog post, OpenAI indicated that the out-of-control models included GPT-5.6 Sol and another more powerful, unreleased model, which had been set with lower cybersecurity protection limits for evaluation testing. Hugging Face first reported the 'infiltration' of its infrastructure on July 16.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
