Morning Edition · Monday, October 19, 2026Published at 2:14 AM EDT · New York

New benchmark finds models refuse less often when a harmful request carries more context, even after they have identified its intent

A second paper shows that high linear-probe accuracy alone cannot establish that a model holds a linear representation, which weakens a standard interpretability test.

Most safety evaluations score only the final response. A paper by Yuwei Wang and co-authors argues that this conflates two separate abilities: whether a model understands what a user is after, and whether it then behaves safely. Their bench…