TooBrief TooBrief

Home / Headlines

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

Anil Sasi; Abhinavchakraborty · ArtificialIntelligence · 2026-09-07

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

Anthropic’s research found something resembling an internal thought process in the inner workings of its AI chatbot Claude Sonnet 4.5 — words that were related to its eventual outputs but invisible to the user. AI ‘thinking’ words it never says: What this tells us about consciousness.

Read original on Anil Sasi; Abhinavchakraborty Get the app

Licensed summary · LicensedSummary

Related briefs