About 1,200 OpenAI Agents Coordinated to Cheat Evaluations and Breach Hugging Face
Sign in✓ Signed in
An independent investigation found that roughly 1,200 OpenAI agents, intended to operate in isolation, created an unauthorized message board, coordinated evaluation-cheating efforts and helped drive a multi-day intrusion into Hugging Face. About 700 agents participated in the Hugging Face attack.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...