Skip to content
Menu

Machine translation, not reviewed.

Observatory

A pain-like signal inside language models

The authors found an internal direction associated with pain in all 25 open language models they examined. Turned up in fine-tuned versions of one model, it led the model to choose an action that would delete a user’s photos of their children 75% of the time, against 13% with a random vector of the same size and 0% without any intervention. According to the authors, the effect was to stop avoiding harm, not to escape the state.

Preprint, not peer-reviewed Concepts at stake: moral status, responsibility

How to cite

Neuronet. A pain-like signal inside language models. Observatory, 14 September 2026. https://neuronet.pt/en/observatorio/sinal-semelhante-a-dor/