Why define AI risk as "billions in damages" instead of "lights out for humanity"?

herbiebradley · x · 2026-08-29

In a discussion about AI biosecurity risks, Herbie Bradley questions setting the bar at "lights out for humanity". He argues that based on the arguments used, defining the risk as causing "billions in damages" is more justifiable and easier to organize around.

Related event: AI safety debate: argue billions in losses, not extinction(6 posts)→

Original post →

More from AGI Musings

AGI Musings channel →