Anthropic finds Claude's internal thinking space

Anthropic has identified a conceptual workspace within its Claude AI model that mirrors aspects of consciousness theory. This internal space represents how the model processes and organizes information during reasoning tasks, offering insight into its decision-making mechanisms.