Thoughts as Thinkers: the biology & computer science of inspiration
- Date published: 2026-10-16
- Speakers: Michael Levin
- Format: Talk
- Watch: https://www.youtube.com/watch?v=P803O6opd3o
- Topics: Platonic space and ingressing minds, AI and biology, Xenobots and anthrobots
In one paragraph
Levin describes a research program to study inspiration and intuition in the lab. His two points: there is a non-physical component that standard physical and computational models don't capture, and inspirations and the agents who get inspired are not different kinds of thing but points on one continuum.
Main ideas (plain words)
- He frames inspiration as a "free lunch": the gap between the competence a system shows and the effort put in through design, evolution or training.
- Examples of that gap from biology: people with very reduced brain volume and normal IQ, tadpoles that see through an eye grafted on the tail with no brain connection, a turtle on a skateboard that suddenly plays like a cat, molecular networks that learn, and xenobots.
- Examples from computer science: even bubble sort shows behaviors that are "nowhere in the algorithm", such as delayed gratification, which he calls "spontaneous side quests".
- He proposes a structured space of patterns (he calls it Platonic space so mathematicians recognize it) that feeds into physical bodies, which act as "front end" interfaces. He has five or six people in his lab running experiments on it.
- He suggests the line between thought and thinker, pattern and agent, may depend on the observer, and that machines may receive inspirations too.
Key quotes (word for word from YouTube's auto-generated captions)
"what I'm going to talk about today is a really strange thing I'm trying to do, which is to develop laboratory model systems for studying inspiration and intuition."
https://www.youtube.com/watch?v=P803O6opd3o&t=0s"inspirations, so the specific content of inspirations and those who are inspired, so agents of all kinds that the humans and and otherwise that become inspired are not actually different things."
https://www.youtube.com/watch?v=P803O6opd3o&t=30s"It has spontaneous side quests that it does that nobody had noticed until until you actually looked to see what else it might be doing that isn't anywhere in the algorithm, okay?"
https://www.youtube.com/watch?v=P803O6opd3o&t=1380s"I propose the following experimental agenda, which I have now five or six people in my lab working working on this of doing experiments."
https://www.youtube.com/watch?v=P803O6opd3o&t=1650s"I think bodies are just thin clients or front end interfaces that host massive ecosystems of these patterns."
https://www.youtube.com/watch?v=P803O6opd3o&t=1980s"so don't feel don't feel too smug because I think machines get them, too."
https://www.youtube.com/watch?v=P803O6opd3o&t=2010sSource: YouTube auto-generated captions, speakers not labeled; wording is provisional. Timestamps point to the start of the 30-second stretch that holds the quote.