Oct 2, 2026 · 1h 17m
Krakauer separates intelligence from capability in the AI age
What Is Intelligence? With David Krakauer
The episode asks whether systems that process information at enormous scale are intelligent—or merely powerful, energy-intensive pattern machines.
- 1Intelligence emerges as adaptive problem-solving, from biological sensing and collective behavior to human language and culture.
- 2Language models can produce capable answers without the dreams, embodiment, curriculum, and contextual understanding that shape biological intelligence.
- 3Outsourcing cognition may expand human reach while eroding competence, making imagination and better questions increasingly important.
Don't miss
Krakauer’s contrast between complementary tools and replacement tools turns GPS, calculators, and other conveniences into a warning about outsourced competence.
The brief
David Krakauer, Neil deGrasse Tyson, and Chuck Nice begin with intelligence as adaptive problem-solving, tracing it from environmental sensing and evolution to brains, bacteria, swarms, and societies.
The discussion separates intelligence from consciousness and language from signaling: animals communicate, but human symbols can generate infinitely many thoughts from finite rules.
Krakauer contrasts living intelligence with language-first AI, arguing that capability is not enough when systems rely on statistical pattern matching, vast energy use, and limited context.
Tools can complement cognition, like maps and abacuses, or replace it, like GPS; the convenience of outsourcing skills may leave people less competent and less intuitive.
The episode closes at the Santa Fe Institute with a challenge: as machines handle more technical work, humans must protect imagination and the ability to ask profound questions.
Featuring
Chuck Nice
Santa Fe Institute
The Invention of Nature
Blade RunnerBooks & mentions
Listen to the full episode and explore every guest, topic, and moment on PodLume.