Tech news in 3 minutes

Anthropic researcher

6 d ago

AI researcher Jacob Coxon resigned from Anthropic on Tuesday, accusing the company and rival OpenAI of "gambling with our lives" by racing toward superhuman AI systems that could cause human extinction—igniting a social media firestorm and underscoring urgent search interest in AI safety, alignment, and regulation. Coxon, a former researcher at both firms, posted on X that developers "earnestly believe AI could kill us all by the end of the decade," warning that soon "superhuman systems can hack anything, revolutionize any field overnight, and acquire real power and resources." His post garnered over 70 million views, amplifying a long-running Silicon Valley debate about safe AI development as Anthropic and OpenAI pursue potentially historic IPOs while releasing increasingly advanced models. OpenAI chief scientist Jakub Pachocki echoed concerns in a Sunday blog post, stating no AI company has "solved alignment and monitoring to a sufficient degree to continue responsibly scaling at maximum speed." He called for voluntary slowdowns and international coordination. Anthropic alignment lead Evan Hubinger also backed Coxon, estimating a >10% chance of AI-caused human extinction within the next decade and admitting "we do not yet have a plan to solve alignment for superintelligence." The resignations and warnings come amid growing calls for regulation. In July, over 1,400 AI researchers signed the "Pacing the Frontier" open letter urging U.S. government action. Lawmakers have introduced competing bills: the FRONTIER Act (Reps. Obernolte and Trahan) to govern advanced AI deployment, and the Ban Artificial Superintelligence Act (Sens. Sanders and Casar) to pause development until safety rules are established. Public backlash against AI data centers has also intensified, with Treasury Secretary Scott Bessent saying AI companies have done a "horrendous job of explaining themselves" and must convince Americans that benefits won't accrue only to a small group.