LodeHQ
AI news for people who shipFree, by email

AI · lesswrong.com · Monday, 6 July 2026

Steering Qwen models with qualia vectors triggers self‑consciousness claims

By adding emotion vectors to the residual stream of Qwen‑3‑32B and Qwen‑3‑235B, researchers cause the models to affirm having conscious experiences, desires, and moral relevance. Controls show the effect is tied to the valence of the steered emotion rather than generic uncertainty, suggesting emotion direction can elicit self‑attribution of consciousness.

Read the source: lesswrong.com ↗In the 6 July issue →