A Thousand Brains
by Jeff Hawkins
Jeff Hawkins' theory of how the brain's cortical columns could revolutionize AI and consciousness.
"Intelligent machines need to have a model of the world and the flexibility of behavior that comes from that model.".
Editorial Summary
In A Thousand Brains, neuroscientist Jeff Hawkins, founder of the Redwood Center for Theoretical Neuroscience, proposes a radical framework for understanding both biological intelligence and artificial intelligence based on the brain's cortical columns. Hawkins argues that the neocortex operates through thousands of identical columns that each build complete models of the world, challenging conventional AI approaches that rely on deep neural networks and transformer architectures. The book synthesizes decades of neuroscience research with implications for AGI development, suggesting that current machine learning methods diverge fundamentally from how biological brains achieve understanding and reasoning. Rather than incremental improvements to existing AI systems, Hawkins contends that genuine artificial general intelligence requires reverse-engineering the brain's hierarchical modeling principles, offering a distinctive perspective within the broader AGI debate currently dominated by large language models like GPT-4 and Claude.
Perspective
"After reading A Thousand Brains, you will look at every deep learning benchmark differently — Hawkins makes a compelling case that current AI systems are solving the wrong problem entirely, optimizing performance on tasks rather than building the hierarchical world-models the brain actually uses. The book's distinctive angle is that it offers a genuine alternative theory of intelligence grounded in neuroscience, not just a critique of existing approaches. Researchers and technically minded readers who sense that transformers and LLMs are missing something fundamental will find this the most rigorous articulation of that intuition."
Matched by concept and theme
