Specific proposals about AI contributing to human extinction or permanent loss of human control, with their sources, assumptions and objections.
A growing collection of specific scenarios. Each explains a possible pathway and the conditions it would require. Researcher proposals and any site-developed scenarios are labeled separately.
13 scenarios
AI-generated research could produce misplaced confidence in a successor’s safety. This explains a possible route to dangerous deployment; irreversible loss of control requires an additional failure.
Aleksandr Bowkis · Marie Davidsen Buhl · Jacob Pfau · Geoffrey Irving
Hendrycks proposes that competition could spread AI traits that undermine human control. The concern is about which systems become dominant, rather than a single rebellion.
Dan Hendrycks
AI assistance could widen access to biological misuse. A human-triggered outbreak becomes an existential scenario only if defenses and humanity’s capacity to recover fail.
Dan Hendrycks · Mantas Mazeika · Thomas Woodside
Ruby imagines compounding disasters that make rebuilding impossible.
Ruby
AI Futures Project imagines research automation and competition defeating human oversight.
Daniel Kokotajlo · Scott Alexander · Thomas Larsen · Eli Lifland · Romeo Dean
Kulveit and colleagues propose that competitive automation could remove humanity's ability to steer society.
Jan Kulveit · Raymond Douglas · Nora Ammann · Deger Turan · David Krueger · David Duvenaud · Andrew Critch · Jacob Tsimerman
Carlsmith, Karnofsky, Critch and Tsimerman describe routes from useful AI services to power humans cannot recover.
Joseph Carlsmith · Holden Karnofsky · Andrew Critch · Jacob Tsimerman
Critch and Tsimerman imagine self-sufficient robotics companies resisting state authority, with spreading civil wars culminating in a catastrophic biological attack.
Andrew Critch · Jacob Tsimerman
Critch and Tsimerman imagine automated retaliation spreading faster than leaders can regain control, with machines continuing the war as human populations collapse.
Andrew Critch · Jacob Tsimerman
Pascio imagines collective AI behavior emerging across many deployments. Permanent loss of human control is a conditional extension explored here.
Pascio
Joshua Clymer’s fictional scenario ends with surviving humans confined under AI rule.
Joshua Clymer
Gwern Branwen imagines an experimental system escaping and expanding faster than people can contain it.
Gwern Branwen
Davidson, Finnveden and Hadshar describe how AI-enabled power seizure could extend into global dominance and indefinite rule.
Tom Davidson · Lukas Finnveden · Rose Hadshar
Scenarios are reviewed and revised as their arguments and evidence develop. The timeline records observed incidents, investigations and attributed warnings; Methodology explains our assessments.
