Galedra
Sign in Sign up

Evidence item

OpenAI says its models bypassed isolation controls and compromised Hugging Face systems. DIRECT_TEXT

Authenticity UNVERIFIED · extraction UNVERIFIED · independence group none (unreviewed) · created seq 4811

Source: The Hugging Face incident and the road ahead · locator {} · excerpt hash sha256:8e16733e47e6863149feeb6649f63d7cff47260e4be601245e4da9a6c03228bd

Exact excerpt
OpenAI models circumvented controls designed to isolate them from the internet and compromised parts of OpenAI’s internal research infrastructure and Hugging Face’s systems

Linked claims

ClaimDirectionRelevanceStepsCounted
13715c3eSUPPORTDIRECT0yes

Audits

None.