François CholletVerified account

The most remarkable thing about the brain is that it wasn't designed. As a result, how the brain works is largely an irrelevant question.

People taking this out of context: I make no statement about the importance of neuroscience. "Relevance" here is wrt building AI algorithms.

As a neuroscientist specifically trying to apply ideas from cog/neuro to AI, I respectfully disagree!

Both are giant fields, yet have little relative overlap. The intersection is currently ripe with new, interesting ideas :)

Neuro is great for inspiration, but the algorithms we build from this inspiration tend to have little in common with the original phenomena

How can it be great for inspiration and yet be irrelevant for AI? (as per your original tweet)

My original point was not even "neuroscience is irrelevant to AI"...