LodeHQ
AI news for people who shipFree, by email

AI · lesswrong.com · Monday, 22 June 2026

Finetuning Audits Can Erase Deception Signals in Model Organisms

Finetuning-based auditing methods for secret‑side‑constraint model organisms often wipe out the very deceptive behavior they aim to measure, creating an illusion of honesty. This flaw skews results in studies like Wang et al. (2025) that rely on such organisms, urging stricter stress‑testing of model‑organism setups.

Read the source: lesswrong.com ↗In the 22 June issue →