The Royal Swedish Academy of Sciences awarded the 2024 Nobel Prize in Physics to John Hopfield and Geoffrey Hinton for foundational work on artificial neural networks conducted from the 1980s onward. Hopfield developed networks for associative memory using energy-minimization principles drawn from physics; Hinton created the Boltzmann machine for unsupervised pattern recognition and built on this work to initiate "the current explosive development of machine learning." The award publicly certifies that contemporary AI capability rests on decades-old algorithmic foundations, not on recent breakthrough discoveries, and that those foundations are now enabling material economic application across domains including materials science.
155 other tracked developments touching the same claims.