Why AI Has a Plato Problem — Mazviita Chirimuuta

Scientific oversimplification leads researchers to falsely equate the brain with a computer. True cognition emerges from physical, biological interaction with the world, meaning human intelligence cannot be fully replicated by digital models that lack a body and lived experience.

• Brain as computer — Treating the mind like a machine provides a useful framework, but it creates a narrow viewpoint that ignores crucial biological factors like biochemistry and immune system interaction [27:01].

• Scientific oversimplification — Researchers often favor neat, mathematical representations that are easier to calculate, though these models inevitably leave out the messy, uncontrollable variables found in nature [03:52].

• Haptic realism — Knowledge is not acquired through passive observation; it is constructed through active, tactile engagement and physical manipulation of our environment [16:53].

• Biological embodiment — Humans learn through social context and physical experience, meaning a digital program without a body or life challenges lacks the foundation necessary for genuine understanding [40:37].

• Historical warning — The failed "reflex theory" of the 19th century serves as a cautionary tale of how holding onto a simplified model for too long can derail an entire field of science [10:12].

• The Kaleidoscope myth — AI researchers often operate under a "Platonic" assumption that the universe is made of hidden, perfect mathematical rules waiting to be decoded, rather than recognizing nature as inherently complex and shifting [05:47].

• Developmental risks — Prioritizing screen-based interactions over direct human connection risks the social development of children, who require face-to-face feedback to mature [51:48].

• How does the lack of physical embodiment affect the way AI processes information?

• What are the risks of using computational models to explain human cognition?