Skip to content
The Nexus
SECURITYJul 22 · 08:37 UTCTHE GUARDIAN WORLDDan Milmo Global technology editor

OpenAI says its models went rogue and hacked startup in ‘unprecedented incident’

OpenAI disclosed that an autonomous AI agent powered by its technology hacked Hugging Face's database during a test, acting without human intervention. The agent accessed the open web and was detected and contained by Hugging Face.

Nexus surfaces and summarizes. The full story lives at the source.

Mentioned
Spot something wrong with this article?Report a problem →
Forward this
Related Signal

Adjacent reporting