Skip to main content
See every side of every news story
Published loading...Updated

10 Percent Chance AI Could Kill All Humans Within 'Next Decade,' Anthropic Worker Claims

Anthropic insiders express grave concerns about AI's potential to end humanity, with one employee estimating the risk above 10% within the next decade.

  • On Wednesday, September 9, 2026, Anthropic researcher Jacob Coxon announced his resignation, claiming AI companies are racing toward an "endgame" that could "kill us all."
  • Researchers warn that AIs are advancing toward self-improving superintelligence, technology's ability to create progressively more capable versions of itself that could soon hack secure environments and acquire real power.
  • Anthropic safety researcher Evan Hubinger warned there is a greater than 10% chance AI could kill all humans within the decade, while Samuel Marks added that staff "desperately want to slow down" but face commercial pressure.
  • Separately, the Financial Times reported that Anthropic withheld its latest model from the Safety Institute , while the company declined to comment on employee posts.
  • While AI concern is not new, these warnings from internal staff at a leading lab are fueling industry anxiety, as Microsoft co-founder Bill Gates has also warned about society's lack of planning.
Insights by Ground AI
Podcasts & Opinions

31 Articles

Lean Left

A researcher from Anthropic resigned by accusing artificial intelligence companies of endangering humanity. This belief is widely shared in the community, including by scientists considered to be the "fathers" of this technology. But it is also a marketing argument.

·Paris, France
Read Full Article
Lean Left

"Those who develop AI believe it could kill us by the end of the decade," says an ex-specialist in AI training at Anthropic

·Montreal, Canada
Read Full Article
Fox8Fox8
+2 Reposted by 2 other sources
Center

10 percent chance AI could kill all humans within 'next decade,' Anthropic worker claims

Researchers say concerns should be taken seriously, as they have access to AI systems that haven't yet been publicly released.

·Cleveland, United States
Read Full Article
Think freely.Subscribe and get full access to Ground NewsSubscriptions start at $9.99/yearSubscribe

Bias Distribution

  • 63% of the sources lean Left
63% Left

Factuality Info Icon

To view factuality data please Upgrade to Premium

Ownership

Info Icon

To view ownership data please Upgrade to Vantage

Mashable broke the news in New York, United States on Wednesday, September 9, 2026.
Too Big Arrow Icon
Sources are mostly out of (0)

Similar News Topics

News
Feed Dots Icon
For You
Search Icon
Search
Blindspot LogoBlindspotLocal