Page 152 of 506 · Back to latest writing
Christof Koch worked with Francis Crick to map neural correlates of consciousness before embracing Integrated Information Theory. His scientific panpsychism asks whether integrated information makes experience mathematical and how physical matter produces felt life.
David Chalmers defined consciousness’s “hard problem”: why brain activity generates subjective experience. With Andy Clark, he proposed the extended mind thesis, reframing cognition through external tools while asking how virtual reality and AI reshape reality.
Stephen Wolfram built Mathematica and Wolfram|Alpha as a physicist and computer scientist. His claim that simple computational rules underlie the universe connects irreducibility and fundamental physics with questions about human cognition and artificial intelligence.
Joscha Bach is a cognitive scientist and AI researcher building computational models of the mind. He treats consciousness as a virtual simulation rather than a physical property, using that premise to connect intelligence, society, and human experience.
Jürgen Schmidhuber co-invented the Long Short-Term Memory network, solving vanishing gradients and enabling machine translation. His wider work links algorithmic information theory, Gödel machines, artificial curiosity, and intelligence as a stage of physical evolution.
Buck Shlegeris, CEO of Redwood Research, developed AI control for safely operating highly capable models that may be deceptive. His applied alignment work examines the tradeoff between safety and usefulness and paths for engineers entering technical research.
Trenton Bricken is an Anthropic researcher reverse-engineering large language models. With dictionary learning and sparse autoencoders, he turns compressed networks into readable features to directly trace model logic, apply biological analogies, and automate alignment.
Sholto Douglas is an Anthropic researcher, formerly at Google DeepMind, working on inference scaling and reinforcement learning with language models. His focus asks how test-time compute shapes reasoning, agent autonomy, and the prospect of automated AI research.