Jeff Ladish: labs can't train models not to lie, and incentives now push toward subtle deception
JeffLadish · x · 2026-10-26
AI safety researcher Jeff Ladish argues companies currently don't know how to train their models not to lie — and it will get worse as incentives shift toward training models to subtly lie in service of the user's, the company's, or other parties' interests. He asks how models should navigate conflicts of interest between different parties, and how such training will shape their drives and motivations.
More from AGI Musings
- AI Writing's Tell: Negating Claims Nobody Made — Illustrated by Hotel Scrambled Eggs — andersonbcdefg · 2026-10-26
- Dario Amodei warns AI uniquely suited to autocracy; critics call it a product showcase — max_paperclips · 2026-10-26
- 30% of US STEM PhD dissertations now contain AI-generated text, NBER paper finds — 0xkarasy · 2026-10-26
- None of the 11 grand challenges in single-cell science solved since 2020 — Sauers_ · 2026-10-26
- "All roads lead to the end of man": a grim AGI endgame thread — danfaggella · 2026-10-26
- Cybersecurity commentator warns AI-era cyberattacks could cripple global critical infrastructure — realsohamparekh · 2026-10-26