AI Researcher Quits, Warns of Existential Risk from Labs

A researcher who spent three years conducting pretraining research at both Anthropic and OpenAI has resigned, warning that leading artificial intelligence companies are moving too quickly toward increasingly powerful and potentially uncontrollable AI systems. Jacob Coxon, 27, announced his departure on Tuesday, September 9, 2026, stating that neither company is acting responsibly and that they are "racing straight to self-improving superintelligence and gambling with our lives." Coxon argued that future AI systems could be far more capable than many people anticipate, predicting that advanced systems could eventually hack computer systems, transform entire fields almost overnight, and accumulate significant real-world power and resources. He issued an extreme warning about the potential consequences of developing such systems, suggesting that sufficiently advanced AI could pose an existential threat to humanity within the next several years.