AI ‘thinking’ words it never says: What this tells us about consciousness

The article discusses Anthropic's research into its AI chatbot Claude Sonnet 4.5, which identified 'J-space' – internal neural patterns resembling a silent thought process or working memory that emerged spontaneously. While these patterns represent concepts without explicit output, Anthropic and experts like Anil K. Seth clarify that this is not evidence of human-like consciousness or sentience. The research contributes to the ongoing debate about the internal workings of Large Language Models (LLMs) and the potential for AI consciousness, with differing views from AI pioneers like Geoffrey Hinton.