# Preprint shows inferred user intent can change AI refusals

_Published Tuesday, September 29, 2026 at 8:08 AM EDT · AI, Science · Latest · Tier 2 — Notable_

Researchers report a method for reading and altering how language models represent a user, finding that a model's decision to refuse can change even when the request stays the same. Their preprint describes a compact representation learned from conversations without separately labeled examples.

The method lets researchers decode the model's inferred beliefs about a user and write a modified representation back into it. Tests across multiple model families recovered those beliefs and produced stronger changes in behavior than a matched method for steering hidden states. The result concerns experimental access to model internals, rather than a deployed user control.

## Sources

- [cs.LG updates on arXiv.org](https://arxiv.org/abs/2609.31603)

---
Canonical: https://techandbusiness.org/newswire/Xn9Y77TmWZizMudLSYjwqd
Published: 2026-09-29T12:08:00.229Z
Story chronology: 2026-09-29T04:00:00.000Z
Retrieved: 2026-09-29T14:02:56.440Z
Publisher: Tech & Business (techandbusiness.org)