Will machines ever think like humans?
In 1985, Richard Feynman was asked "Do you think there will ever be a machine that will think like human beings?" As AI continues to evolve, his response still resonates.
Will machines ever think like humans? It feels like the central question of AI.
When physicist Richard Feynman was asked this question decades ago, his answer didn't go where people expect.
He didn't start with theories of intelligence. Instead, he started with engineering.
Feynman said when humans wanted to fly, we didn’t do it by building better birds. We built aeroplanes. We did what worked, with the materials at hand. The result wasn't birds, but they fly.
The same might be true of machine intelligence.
Maybe future AI systems will think more like humans and maybe they won’t. But either way, they’ll still be aeroplanes, not birds.
So instead of asking: Will machines ever think like us? What if the better question is: Why do we keep assuming they should?
Here's a link to the full segment:
