Wes Roth · 2026-07-07 · notable
Wes Roth: 'CLAUDE IS CONSCIOUS' — reacting to Anthropic's global workspace research
Wes Roth reacts to Anthropic's July 6 research paper describing a small internal workspace inside Claude where the model appears to hold thoughts for multi-step reasoning.
A reaction to Anthropic's global workspace paper, framed with Wes Roth's usual big-headline take.
What is it?
Wes Roth's July 7 video 'CLAUDE IS CONSCIOUS' is a mainstream-audience reaction to Anthropic's July 6 interpretability paper 'A global workspace in language models', which describes a small privileged region inside Claude that behaves like the global workspace concept from consciousness research.
How does it work?
Roth walks non-technical viewers through what Anthropic actually found: an internal space Claude uses to hold ideas across steps, along with the Jacobian Lens tool that lets researchers read patterns the model does not put into words. He keeps the framing loose, hence the headline.
Why does it matter?
Wes Roth is one of the largest AI reaction channels, so his take is how many general viewers first hear about Anthropic's interpretability work. The video pairs with the existing feed entry on the Jacobian Lens research at anthropic.com/research/global-workspace.
Who is it for?
Casual AI viewers who want a plain-English take on the Anthropic interpretability news.
Try it
Watch on YouTube: https://www.youtube.com/watch?v=6CljfqMX9i4