The evidence behind the guides

Sources

Find a study, check what it supports, and follow it back to a concern.

Sources for Corrigibility failure

1 source

Guide citations are still under editorial review.

Hadfield-Menell, Dragan, Abbeel and Russell · 2017 Formal model

The Off-Switch Game (original source, new tab)

Analyzes shutdown incentives under specific mathematical assumptions; does not prove a safeguard works for deployed language models.

Sources retain their original publishers’ rights. Inclusion does not imply endorsement. Saved sources stay in this browser on this device. Dates and source descriptions come from the existing editorial records.