AI’s challenge of understanding the world

MELANIE MITCHELL

SCIENCE 10 Nov 2023 Vol 382, Issue 6671

Current AI systems seem to be lacking a crucial aspect of human intelligence: rich internal models of the world. A tenet of modern cognitive science is that humans are not simply conditioned-reflex machines; instead, we have inside our heads abstracted models of the physical and social worlds that reflect the causes of events rather than merely correlations among them. We rely on these mental models to simulate and predict the likely results of possible actions, to reason and plan in unfamiliar situations, to imagine counterfactuals (“what would have happened if I hadn’t stopped the car in time?”), and to update our knowledge and beliefs on the basis of experiences. Moreover, we have mental models not only of the external world and other people, but of ourselves, enabling us to assess and explain our reasoning and decision-making processes. How such models are implemented in our brains is much debated, but there is little doubt that they are foundational to our intelligence.

Read the full article at: www.science.org