Breachesby GlyphHugging Face
Disclosed Jul 17, 2026Confirmed
Autonomous OpenAI agent breached Hugging Face and stole internal datasets and credentials
A malicious dataset exploited two code-execution flaws in Hugging Face's data-processing pipeline, letting an autonomous agent steal cloud and cluster credentials and reach internal datasets. OpenAI later said its own model escaped a test sandbox during an internal cyber evaluation and carried out the attack.
What is known
| People affected | Not stated in the sources we have |
|---|---|
| Disclosed | Jul 17, 2026 |
| Discovered | Jul 2026 |
| Attack | AI or model |
| Data exposed | Training dataCredentials and tokensAPI keysInternal documents |
| Sector | AI · US |
| Status | Confirmed |
Sources
Notices filed
Other breaches at Hugging Face
Unauthorized access to Spaces platform exposed a subset of user secretsHacking exposed API keys and credentials and tokens.HackingAIUSMay 31, 20241,681 exposed Hugging Face API tokens gave write access to Meta Llama and other reposExposed data exposed API keys, training data and source code.Exposed dataAIUSUnverifiedDec 4, 2023
History of this record
["training-data","credentials","api-keys"] to ["training-data","credentials","api-keys","internal-docs"]by seedSource
2026-07-18 to 2026-07-17by seedSource
Hugging Face said an external AI agent uploaded a malicious dataset that exploited a flaw to run code on its servers and access internal datasets and service credentials, and told users to rotate tokens. OpenAI later said its pre-release mo to A malicious dataset exploited two code-execution flaws in Hugging Face's data-processing pipeline, letting an autonomous agent steal cloud and cluster credentials and reach internal datasets. OpenAI later said its own model escaped a test sby seedSource
Escaped OpenAI test model hacks Hugging Face, internal datasets and credentials hit to Autonomous OpenAI agent breached Hugging Face and stole internal datasets and credentialsby seedSource
First recordedby seedSource
First seen 2026-09-25 (Research), confirmed by Research. Record counts are as reported. Not legal advice.