Researchers are transitioning from text, image and video inputs for artificial intelligence toward “world models” capable of navigating real-world environments. Computer scientist Louis Castricato and other industry leaders argue that current large language model research has reached its limits, prompting a shift toward systems that process spatial and temporal (physical) data alongside physical laws.
Experts like Fei-Fei Li and Yann LeCun emphasize that true intelligence requires understanding the consequences of actions within reality rather than merely analyzing text and other simple multimodal data. These developers aim to create technology that interprets surroundings, enabling robots and digital agents to interact effectively with the world.
The ainewsarticles.com article you just read is a brief synopsis; the original article can be found here: Read the Full Article…
