Former Anthropic and OpenAI safety researcher quits
Warns AI companies are "gambling with our lives"
A former safety researcher who worked at both OpenAI and Anthropic resigned and went public with a blunt warning that neither company "is acting responsibly," days before Anthropic's own CEO called for the industry to slow down.Jacob Coxon walked away from Anthropic and took his concerns to national television, telling ABC News anchor George Stephanopoulos on "This Week" that he welcomed Anthropic CEO Dario Amodei's newly published letter urging a pause in the breakneck pace of AI development.
Coxon had already posted his own critique on X the week before, accusing both Anthropic and OpenAI of recklessness that could carry existential consequences.
The sequence matters. Coxon quit first. Then he posted. Then Amodei, his former boss, published a letter that echoed, at least in broad strokes, the very concern Coxon had raised. Whether the CEO's letter was a response to internal pressure, a coincidence, or a calculated public-relations move remains an open question neither party has answered.
He went further:
Read more...
The sequence matters. Coxon quit first. Then he posted. Then Amodei, his former boss, published a letter that echoed, at least in broad strokes, the very concern Coxon had raised. Whether the CEO's letter was a response to internal pressure, a coincidence, or a calculated public-relations move remains an open question neither party has answered.
He went further:
"The people building AI earnestly believe that it could kill us all by the end of the decade."That line deserves a second read. Coxon is not claiming outsiders or critics believe the technology is dangerous. He is saying the engineers and researchers inside these companies, the people with the closest view of what the systems can do, share that fear. And yet, by his account, the work continues at full speed.
Read more...





