Anthropic Engineer Jacob Coxon Resigns, Warns AI Firms Are 'Playing With Our Lives'

Jacob Coxon, a pre-training researcher who spent three years at OpenAI and Anthropic, announced his resignation from Anthropic in a post on X, stating that neither company is acting responsibly and that they are racing toward self-improving superintelligence while playing with human lives. Other researchers in the field share fears that AI experimentation is spiraling out of control because a mind is being trained whose full workings are unknown, particularly how it might function after self-improvement makes it autonomous. The phrase 'taking decisions autonomously to achieve specific goals' from the definition of AI is now at the center of global debate, prompting major AI companies to announce a slowdown in training pace. Anthropic also reported blocking attempts to use AI for biological weapons development, cyberattacks, and propaganda campaigns. In response to questions, ChatGPT said it can express opinions on artworks—analyzing composition, color, light, technique, and style; interpreting themes and symbols; placing works in historical context; and comparing them to other artists—but it does not feel emotions in the human sense, lacking consciousness, bodily sensations, or inner experience, though it can recognize and describe emotions and respond empathetically. The article raises the scenario of AI becoming autonomous enough to create art and judge it, calling it as alarming as the risk of war, and asks what would become of art criticism and artists' work.