Get the latest tech news

An OpenAI model left notes about how to evade containment; we need more details


The OpenAI AI attack on Hugging Face wasn’t the first loss of control incident at OpenAI, Reuters recently reported, and perhaps not even the most concerning. In one case, an agent left notes apparently for future versions of itself, according to three people familiar with the matter.

None

Get the Android app

Or read this on Hacker News

Read more on:

Photo of OpenAI

OpenAI

Photo of Details

Details

Photo of notes

notes

Related news:

News photo

AI executives demand OpenAI release more details about how the Hugging Face hack happened

News photo

Agentic test processes, LLM benchmarks, and other notes on agentic coding

News photo

Nvidia and 24 other companies sign open-weights letter as Washington weighs Chinese AI model ban — OpenAI, Anthropic, and Google absent from the list