World Labs’ Atlas signals a possible shift in AI from large language models to “world models” that understand 3D environments. Built by Fei-Fei Li’s team, Atlas can reconstruct scenes from images. Unlike LLMs trained mainly on internet data, world models aim to understand space, movement and causality, enabling robots, autonomous vehicles, digital twins and simulations to interact more intelligently with the physical world.
aiweekly.co.in
“Where Machines Make Headlines.”