Read original ↗
newsMIT Technology Review AITrust 88 · LabPublished 5d agoLive · 3d ago

Anthropic found a hidden space where Claude puzzles over concepts

The AI firm Anthropic has developed a technique that has given it the clearest glimpse yet at what’s really going on inside large language models as they answer questions or carry out tasks. What they found ranges from the mundane to the unnerving. Researchers at the company built a tool called the Jacobian lens (or…

ResearchMIT Technology Review AIverified

Covers

paperHow Anthropomorphic Language Impacts Public Perceptions of AIpaperSituation Perception: A Necessary Primitive to Artificial Superintelligencerepothevseprod/humanizer-rurepoetkecc/baibotrepoahumblenerd/tour-of-agents

Covers (incoming)

repoThibautMelen/agentic-ai-systemsrepotryAGI/Anthropicreposmixs/humanizer-rurepoExtraltodeus/J-Wash

Related across the graph

repoThibautMelen/agentic-ai-systemsrepoetkecc/baibotreposmixs/humanizer-rupaperHow Anthropomorphic Language Impacts Public Perceptions of AIpaperSituation Perception: A Necessary Primitive to Artificial SuperintelligencerepoExtraltodeus/J-Washrepothevseprod/humanizer-rurepotryAGI/Anthropicrepoahumblenerd/tour-of-agents