A Perfect Chess World Model Still Won't Beat Magnus Carlsen, Researcher Argues

Machine Learning Street Talk · youtube · 2026-09-22

In a short clip from a Machine Learning Street Talk interview, Alexander Mattick (Fraunhofer IIS, UTN) argues that even a perfect world model of chess still won't beat Magnus Carlsen, because predicting the future and choosing the right action are fundamentally different problems. He draws a parallel to humanoid robotics: robots can do backflips on camera yet still can't fetch a carton of milk — world modeling alone doesn't translate into decision-making and execution.

Original post →

More from Embodied

Embodied channel →