NEWS | Jacob Coxon, a 27 year old AI researcher who spent the past three years working on pretraining at both OpenAI and Anthropic, resigned from Anthropic this week and published a warning on social media accusing both companies of racing toward self improving superintelligence without acting responsibly. His post has since gone viral, gathering hundreds of thousands of views.
Coxon said future AI systems could become powerful enough to hack almost anything, transform entire industries overnight, and gain real world resources and influence. He claimed people building this technology privately believe it could kill everyone by the end of the decade, even when they publicly downplay that risk to sound more measured. Anthropic's own Alignment Science Lead, Evan Hubinger, publicly backed him up, saying he personally believes there is more than a 10 percent chance of this kind of catastrophe within the next ten years, while adding the company does not yet have a solution for keeping a superintelligent system aligned with human interests.
BMN sees this as a rare moment where an actual insider, not an outside critic, is raising the alarm publicly, and getting confirmed by someone still working at the company he just left. That combination is different from the usual AI doom predictions coming from people outside the labs themselves.
Coxon is calling for AI labs to agree on a temporary slowdown in how fast these systems get scaled up, arguing competition between companies is the real force pushing everyone toward danger faster than anyone actually wants to go.
If the people actually building this technology are privately admitting they are scared of what they are creating, why does the race to build it keep moving faster instead of slower?
Your picture beside a comment comes from Gravatar, matched to the email you signed in with. Set one there and it appears here; without one you get your initials. How we handle this