Ex-Anthropic researcher Jacob Coxon says AI could kill all humans by 2030; X post goes viral
Coveragetap to expand ▾Spectrum: Mixed🌍US: 3 · Other: 2
- Ex-Anthropic researcher Jacob Coxon resigns, says AI development is 'gambling with our lives' - tippahnews.com
- Jacob Coxon is a former researcher at OpenAI and Anthropic who said he spent three years doing pre-training research at both companies.
- Coxon posted on X on Sept. 8 and the post has attracted over 150 million views as of this writing.
- Coxon wrote: “The people building AI earnestly believe that it could kill us all by the end of the decade.”
- Anthropic’s current head of alignment reposted Coxon and wrote that he is “correct” that “we really do earnestly believe AI could kill all humans!”
- An individual in the conversation estimated a greater than 10% chance that AI could kill all humans within the next decade.
- Coxon immediately went on a media tour the same day he posted on X and his message captured headlines at major news organizations.
- Country singer Sheryl Crow urged her Instagram followers to take Coxon seriously.
Jacob Coxon, a researcher who worked on pretraining at both OpenAI and Anthropic, resigned from Anthropic and used a post on X to warn the public that he believes the industry is "racing straight to self-improving superintelligence" and that "the people building AI earnestly believe that it could kill us all by the end of the decade" (per Washington Examiner, Time Magazine).
His departure and public statement amplified internal alarm about the pace and governance of frontier AI development, a theme emphasized by left-leaning outlets as existential risk and by center and right outlets as credibility coming from an industry insider (per news.google.com, Time Magazine, Washington Examiner).
Coxon argued in his post that attempting to "speedrun alignment" from private companies is a "hubristic gamble" and said Anthropic understands the stakes but is "locked in a race to get there first," while he characterized OpenAI staff as more varied in how deeply they have internalized the civilizational risks (per Washington Examiner).
Time Magazine and Common Dreams foregrounded his role helping build powerful systems as context for why his warning matters; the Washington Examiner provided more detail about Coxon’s three years doing pretraining research and quoted his direct language about companies "gambling with our lives" (per Time Magazine, news.google.com, Washington Examiner).
The immediate consequence is heightened public scrutiny of Anthropic and OpenAI’s development pace and internal safety cultures, as Coxon’s post has circulated widely and been reported across political outlets (per news.google.com, Time Magazine, Washington Examiner).
Confirmed facts include Coxon’s resignation and his quoted claims on X; claims about internal company deliberations and the likelihood of civilization-scale outcomes remain assertions from Coxon rather than independently verified findings from the companies (per Washington Examiner, Time Magazine).
Expect debate now to focus on whether independent oversight, slower release practices, or coordinated standards can be implemented to address the specific concerns Coxon raised about private-sector "race" dynamics and the feasibility of aligning self-improving systems (per the framing in all sources).
Left- and right-leaning outlets are covering this story differently — in which facts to emphasize, which context to include, and how to frame causes and consequences.