Anthropic researcher Evan Hubinger puts >10% odds on AI killing all humans within a decade

JoshPurtell · x · 2026-09-09

Anthropic alignment researcher Evan Hubinger said on X that the team sincerely believes AI could kill all humans, putting his personal estimate at >10% within the next decade. He admitted Anthropic is trying its best but has no plan to solve alignment for superintelligence and isn't clearly on track to.

The statement drew sharp backlash: user JoshPurtell blasted him for claiming doom is imminent while making "absolutely fuck all progress" on alignment, telling him to resign. The exchange highlights the ongoing tension inside the safety community between doom warnings and slow research progress.

Related event: Anthropic Researcher Quits Over Safety Fears, Warns 10%-Plus Extinction Risk(86 posts)→

Original post →

More from AGI Musings

AGI Musings channel →