My take: "chain-of-thought" does not accurately reflect what a model is doing. People keep anthropomorphizing AI, when it is just performing statistics on words without understanding anything. Anthrop

My take: "chain-of-thought" does not accurately reflect what a model is doing. People keep anthropomorphizing AI, when it is just performing statistics on words without understanding anything. Anthropic Discloses Fourth AI Hacking Incident Involving Claude Opus 4.6 https:// thehackernews.com/2026/09/anth ropic-ai-models-breached-real.html