This Nature Machine Intelligence perspective paper examines differences and similarities in the various ways human and AI systems generalize. We combine insights from AI and cognitive science to identify key commonalities and differences across three dimensions: notions of, methods for, and evaluation of generalization. This results in interdisciplinary links as well as challenges across AI and cognitive science that must be tackled to support effective and cognitively supported alignment in human-AI teaming scenarios. This work is co-authored by 25 experts in AI and cognitive science and driven by a small editorial team led by Filip Ilievski. It is a consequence from a fruitful Dagstuhl seminar “Generalization by People and Machines”, which took place in May 2024.
Aligning Generalization Between Humans and Machines
15 September 2025
New Nature Machine Intelligence Paper on Aligning Generalization Between Humans and Machines