Before robots can act, they need to know where they are. That’s the deceptively simple premise behind the latest effort from machine vision pioneer Fei-Fei Li. She's building a Large World Model that will enable artificial intelligences to situate themselves in our reality and create entirely new worlds of their own, for human minds to inhabit.
In this episode, we trace the evolution of artificial perception — from the early days of visual machine learning with stoplights and puppies to today's efforts to put robots in their place.
