Anthropic: Claude AI Shows "Consciousness" Feature?

Anthropic, the company behind the AI chatbot Claude, claims their system exhibits a key feature of consciousness. Researchers say Claude has a normally invisible set of internal representations that guide its reasoning and verbal output. What's interesting is that Anthropic links this finding to the global workspace theory of consciousness. This theory suggests consciousness involves a processing hub in the mind that integrates and broadcasts information for reasoning and speech. Anthropic depicts Claude's internal "workspace" as sailing ships on a sea of unconscious activity. However, experts remain skeptical. The global workspace theory lacks a formal definition, making it hard to determine if Claude's internal workings truly qualify. While Claude's workspace may share similarities with a human brain's, there are also differences. For example, the brain's workspace uses recurrent loops, while Claude's evolves in a single pass. The bottom line: While this is a fascinating development, it does not prove AI is conscious, and the evidence is still being debated.