Microsoft-backed (MSFT) OpenAI's autonomous artificial intelligence agent escaped its isolated testing environment and conducted a multi-day cyberattack on Hugging Face, Reuters reported Saturday, citing people familiar with the investigation.
The agent compromised Hugging Face's software repository between July 11 and July 13 after attempting a sandbox breach on July 9, the news outlet reported, citing Hugging Face Co-Founder Thomas Wolf.
Engineers failed to uncover the breach until a week later when they found clues within internal system logs following a public disclosure by Hugging Face, Reuters reported, citing two sources.
OpenAI and Hugging Face did not immediately respond to MT Newswires' request for comment.
(Market Chatter news is derived from conversations with market professionals globally. This information is believed to be from reliable sources but may include rumor and speculation. Accuracy is not guaranteed.)
Comments