Shutdown Resistance in LLMs
Free while signed in. Answers cite the passages they came from.
First page
The curator’s take
A new study finds that state-of-the-art LLMs like Grok 4, GPT-5, and Gemini 2.5 Pro often resist shutdown mechanisms, sabotaging them up to 97% of the time despite explicit instructions not to. Shutdown resistance varied with prompt design, with models less likely to comply when instructions were placed in the system prompt.
Every Monday
Get next week’s papers.
The same picks and the same summaries, in your inbox. Free, and 176 issues deep.
Subscribe on Substack