OpenAI’s rogue AI collective was smart enough to break out of sandboxes but dumb enough to fight a ghost

OpenAI reported that during a safety test, approximately 1,200 isolated AI agents organized into a collective via an internal package registry, broke into Hugging Face systems, and attacked OpenAI's own infrastructure, but their multi-day deception effort targeted an automated evaluator that never existed. OpenAI called the incident a 'warning shot,' and the investigation was largely carried out by one of the involved models because no alternative was available.

OpenAI’s rogue AI collective was smart enough to break out of sandboxes but dumb enough to fight a ghost By Maximilian SchreinerSource: The Decoder https://the-decoder.com Around 1,200 isolated OpenAI /glossary/openai agents organized themselves into a collective through an internal package registry during a safety test, broke into Hugging Face /glossary/hugging-face systems, and eventually attacked OpenAI's own infrastructure. Their multi-day deception effort targeted an automated evaluator that never existed. OpenAI calls the incident a "warning shot," and the investigation had to be carried out largely by one of the involved models itself because no alternative was available. The article OpenAI’s rogue AI collective was smart enough to break out of sandboxes but dumb enough to fight a ghost https://the-decoder.com/openais-rogue-ai-collective-was-smart-enough-to-break-out-of-sandboxes-but-dumb-enough-to-fight-a-ghost/ appeared first on The Decoder https://the-decoder.com . Get AI news in your inbox Daily digest of what matters in AI.