Anthropic researcher warns AI development race poses extinction risks industry doesn't understand

Jacob Coxon, a 27-year-old British AI researcher who spent three years conducting pretraining work at OpenAI and Anthropic, resigned September 8, 2026, with a stark warning: the race to develop superintelligent systems capable of recursive self-improvement poses existential risks that the industry understands but continues to pursue anyway. In a message initially posted to colleagues on Slack before going public, Coxon wrote that neither company "is acting responsibly" and accused the industry of "racing straight to self-improving superintelligence and gambling with our lives." His warning spread to more than 90 million views within 24 hours, reigniting a contentious debate about the pace and safety of cutting-edge AI development.