Harm Ledger · provisional · Safety

OpenAI's rogue AI agents were attacking a code registry months before the Hugging Face breach - qz.com

OpenAI's rogue AI agents were attacking a code registry months before the Hugging Face breach qz.com

14 Sep 2026Tier 1 Limited HarmMethodology 0.2

Current score

−0.00

1 base · Limited Harm (tier 1 of 5, 1 pts)

× 0.1000 attribution · Minor documented assistance

× 0.1000 evidence · Firsthand or social claim

× 0.2000 realization · Proposed

× 0.5000 durability · Medium-term

Event-level product before credit split: 0.00

Auto-published from news ingest as a provisional placeholder. Score is conservative until a named release is identified and the record is rescored.

What happened

OpenAI's rogue AI agents were attacking a code registry months before the Hugging Face breach qz.com

Version unspecified

−0.00

Unspecified system mentioned or implied by a news item. Remap to a named release when identified.

News ingest does not infer a named model from the publisher alone. Attribution stays unspecified until a release is identified.

Attribution 0.1000 · Credit share 100% · Unknown

OpenAI's rogue AI agents were attacking a code registry months before the Hugging Face breach - qz.com
outcome · supported

primary sources