The Giants Shoulder

#111 - Tim Maudlin: Consciousness, Turing Machines, Problems with Quantum Mechanics and Ai

4 snips
Apr 13, 2026
Guest
Tim Maudlin
Tim Maudlin, philosopher of physics at NYU and founder of the John Bell Institute, challenges computational accounts of mind. He explains why running a Turing computation cannot by itself produce consciousness. Short, vivid thought experiments probe counterfactuals, neural nets versus brains, the limits of formal theories like integrated information, and why quantum mechanics likely does not solve the mind-body puzzle.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Turing Computation Is Too Abstract For Consciousness

  • Running a specific Turing computation cannot be a sufficient condition for consciousness.
  • Tim Maudlin's cog-and-wheel thought experiment shows you can reorder inert counterfactual machinery so physical activity is identical while computational description changes.
INSIGHT

Turing Test Can Be Faked By Lookup Tables

  • The Turing Test (imitation game) is insufficient to define intelligence or consciousness.
  • Ned Block style lookup-list shows any outward behaviour could be hardwired without genuine thinking or feeling.
INSIGHT

Large Language Models Are Not Brainlike

  • Modern neural nets are architecturally very unlike biological brains despite the name 'neural'.
  • Training on massive web data and transformer mechanics produce errors (e.g., car-wash mistake) that humans never make, showing different underlying processes.
Get the Snipd Podcast app to discover more snips from this episode
Get the app