Poster and lightning talk

The Case for Brain-Anchored Interpretability

Noa Weiss · Eleos Conference on AI Consciousness and Welfare (ConCon) · Berkeley, September 2026

There is still no test for consciousness, but there is one system we are sure has it (the brain), and we can look inside our models for the machinery it uses.

Open the model, as mechanistic interpretability does, but ask a different question. Not what did training build, but did training build what the brain uses?

Poster

Poster: The Case for Brain-Anchored Interpretability. Sections: every way we study AI consciousness has a hole in it; brain-anchored interpretability and how it works; a caveat; why it travels; the ask.
The poster image opens the full-resolution PDF.

Lightning talk slides

Slide 1: The Case for Brain-Anchored Interpretability. Noa Weiss, AI Researcher. ConCon, September 2026.
Slide 2: AI consciousness research: four ways, four holes. Ask it (self-reports), watch it (behavior), open it (interpretability), audit it (architecture vs. the theories), each with its hole.
Slide 3: The fifth way. Brain-anchored interpretability: mech interp, looking for mechanisms we already know from our own mammalian brains. Rely on known theories from neuroscience and look for those in AIs.
Slide 4: One axis, or two? One example, forthcoming work. RL math treats reward and punishment as one axis; in the dominant model of the brain they split into two pathways, D1 go and D2 no-go. Train agents among goals and hazards, then look inside.
Slide 5: Back to the holes. How brain-anchored interpretability addresses mimicry, the choice of theory, and the gap between represented and felt.
Slide 6: How do you know that I'm conscious?
Slide 7: Why it travels. The inference everyone already runs: built like me, feels like I feel.
Slide 8: The evidence runs one way. Nothing brain-like turning up is weak evidence; the brain's machinery turning up is strong evidence.
Slide 9: Noa Weiss. Let's connect: x.com/NoaWeissAI, linkedin.com/in/noa-weiss, weissnoa.com, hi@weissnoa.com. QR codes for my survey of the field and for X. Agree? Disagree? Let's talk.
Noa Weiss

Noa Weiss is an AI/ML researcher working on AI consciousness. She also wrote The State of AI Consciousness Research, a survey of the field. More of her work is at weissnoa.com.