Searle's “Chinese Room”—Can Artificial Intelligence Understand?

[johnwalker]

@civilwestman

having it observe a double-slit experiment. Would the wave function collapse?

I think that almost all physicists would say that it does, but not due to having been “observed” by the LLM/AI. Collapse of the wave function is due to decoherence of the pure quantum state as a result of the particle’s interaction with the detector (that provides the visual input to the AI in this case), resulting in its entanglement with a macroscopic object containing a very large number of particles. Similarly, you will observe collapse and disappearance of fringes in an electron double-slit experiment if you shine an intense light beam behind the two slits so that it interacts with the electrons after they have passed the slits. The interference pattern will collapse regardless of whether anybody is observing the interaction or not.

Dean Radin may disagree.

New Experiments Show Consciousness Affects Matter ~ Dean Radin, PhD

[civilwestman]

@johnwalker

Collapse of the wave function is due to decoherence of the pure quantum state as a result of the particle’s interaction with the detector (that provides the visual input to the AI in this case)

Thanks, John, for explaining decoherence. I had no understanding of that. My proposed experiment was based on my typical very limited knowledge of things quantum (just enough to be dangerous). In my naïveté, I imagined the AI made the observation and then told us the result it observed. In an obverse sense, it was designed to be my ‘poor man’s’ test for consciousness - an extension of the Turning test, I suppose.

[jdougan]

Ron Garrett (aka Erann Gat) has an explanation of QM I found pretty helpful.

QM.pdf

The Quantum Conspiracy: What Popularizers of QM Don't Want You to Know