fMRI research revealed that averaging multiple brain scans creates a composite image that represents no single individual's brain activity. This fallacy of averages extends across society, from education to medicine, proving that systems designed for the 'average' fail to serve the individual.
Related Insights
Society prioritizes the left brain's focus on the individual "me," logic, and social norms. This creates an imbalance, neglecting the right brain's capacity for connection and presence. This neurological imbalance contributes to widespread issues like individualism and unhappiness.
The assumption that superintelligence will inevitably rule is flawed. In human society, raw IQ is not the primary determinant of power, as evidenced by PhDs often working for MBAs. This suggests an AGI wouldn't automatically dominate humanity simply by being smarter.
We live in "communities of knowledge" where expertise is distributed. Simply being part of a group where others understand a topic (e.g., politics, technology) creates an inflated sense that we personally understand it, contributing to the illusion of individual knowledge.
Work by Kahneman and Tversky shows how human psychology deviates from rational choice theory. However, the deeper issue isn't our failure to adhere to the model, but that the model itself is a terrible guide for making meaningful decisions. The goal should not be to become a better calculator.
Todd Rose ate grapefruit daily based on its average health benefits, only to discover through personalized testing that it was the single worst food for his blood sugar. This demonstrates that relying on population-level averages for personal decisions can be dangerously counterproductive.
Economics-based rational choice theory frames decisions as a calculation of "expected utility," multiplying value by probability. This analogizes complex life choices—from careers to partners—to casino bets, oversimplifying non-quantifiable factors and reducing judgment to mere calculation.
The common belief that AI can't truly understand human wants is debunked by existing technology. Adam D'Angelo points out that recommender systems on platforms like Instagram and Quora are already far better than any individual human at predicting what a user will find engaging.
Milton Friedman's 'as if' defense of rational models—that people act 'as if' they are experts—is flawed. Predicting the behavior of an average golfer by modeling Tiger Woods is bound to fail. Models must account for the behavior of regular people, not just theoretical, hyper-rational experts.
The central challenge for current AI is not merely sample efficiency but a more profound failure to generalize. Models generalize 'dramatically worse than people,' which is the root cause of their brittleness, inability to learn from nuanced instruction, and unreliability compared to human intelligence. Solving this is the key to the next paradigm.
The popular assumption that the brain is optimized solely for survival and reproduction is an overly simplistic narrative. In the modern world, the brain's functions are far more complex, and clinging to this outdated model can limit our understanding of its capabilities and our own behavior.
