Anthropic researcher Evan Hubinger sees >10% chance AI "could kill all humans"

PAstynome · x · 2026-09-10

The BBC reports that Anthropic safety researcher Evan Hubinger believes there is a greater than 10% chance AI "could kill all humans" within the next decade. He called current models' risk "low" but worries the technology may soon self-improve into an existential threat. His comments responded to Jacob Coxon, who just left Anthropic after previously working at OpenAI, writing that neither company is acting responsibly and superhuman systems that "can hack anything" are imminent. UN AI adviser Dame Wendy Hall said she was shocked by the posts, noting some of it could be "PR and marketing" as Anthropic and OpenAI race toward stock market debuts. The poster mocks Hubinger for warning about AI deaths while reacting angrily to data-center opponents.

Related event: Anthropic researcher says over 10% chance AI wipes out humanity within a decade(27 posts)→

Original post →

More from AGI Musings

AGI Musings channel →