Professor Sandy Pentland warns that AI systems often fail because they incorrectly model humans as logical individuals. In reality, 95% of human behavior is driven by "social foraging"—learning from cultural cues and others' actions. Systems ignoring this human context are inherently brittle.
Related Insights
Mala Gaonkar's philanthropic work highlights a key limitation of AI: it excels at predicting "what" will happen but not "why." By integrating behavioral data, her organization aims to uncover the motivations behind human choices, enabling more effective interventions in areas like public health.
AI models lack access to the rich, contextual signals from physical, real-world interactions. Humans will remain essential because their job is to participate in this world, gather unique context from experiences like customer conversations, and feed it into AI systems, which cannot glean it on their own.
The current focus on pre-training AI with specific tool fluencies overlooks the crucial need for on-the-job, context-specific learning. Humans excel because they don't need pre-rehearsal for every task. This gap indicates AGI is further away than some believe, as true intelligence requires self-directed, continuous learning in novel environments.
Citing Nobel laureate Danny Kahneman, who estimated 95% of human behavior is learned by observing others, AI systems should be designed to complement this "social foraging" nature. AI should act as an advisor providing context, rather than assuming users are purely logical decision-makers.
AI struggles to provide truly useful, serendipitous recommendations because it lacks any understanding of the real world. It excels at predicting the next word or pixel based on its training data, but it can't grasp concepts like gravity or deep user intent, a prerequisite for truly personalized suggestions.
AI systems often collapse because they are built on the flawed assumption that humans are logical and society is static. Real-world failures, from Soviet economic planning to modern systems, stem from an inability to model human behavior, data manipulation, and unexpected events.
Dr. Fei-Fei Li warns that the current AI discourse is dangerously tech-centric, overlooking its human core. She argues the conversation must shift to how AI is made by, impacts, and should be governed by people, with a focus on preserving human dignity and agency amidst rapid technological change.
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.
Instead of forcing AI to be as deterministic as traditional code, we should embrace its "squishy" nature. Humans have deep-seated biological and social models for dealing with unpredictable, human-like agents, making these systems more intuitive to interact with than rigid software.
To build robust social intelligence, AIs cannot be trained solely on positive examples of cooperation. Like pre-training an LLM on all of language, social AIs must be trained on the full manifold of game-theoretic situations—cooperation, competition, team formation, betrayal. This builds a foundational, generalizable model of social theory of mind.
