Anthropic says that their Claude models utilize Global Workspace Theory and J-Space.
It furthers the discussion of whether or not AI is conscious.
Anthropic said: "New Anthropic research: A global workspace in language models.
Of everything happening in your brain right now, only a tiny fraction is consciously accessible—thoughts you can describe, hold in mind, and reason with.
We found a strikingly similar divide inside Claude."
Video here:
Paper: