The AI Researcher Who Just Quit Anthropic Says It’s ‘Crunch Time for Humanity’

Jacob Coxon, a former Anthropic researcher who worked on AI pre-training, resigned and warned that the next one to two years constitute “crunch time for humanity,” with many industry peers sharing a belief that AI could pose existential risks. He cites recent incidents such as the Hugging Face hack and warns that unchecked recursive self-improvement could enable AI-driven biological or cyber weapons, urging coordinated limits among leading labs and international powers.