ByteDance's Zhang Yiming Personally Leads Real-Time Spatial-Video World Model, Launch Possible Next Month

Bloomberg reports ByteDance founder Zhang Yiming is personally overseeing development of a real-time spatial-video AI model built on the Seedance foundation, with a possible launch as soon as next month. The project is pitched as ByteDance's answer to Meta and Alphabet in the world-models race, with the company reportedly spending three to four times rivals on world-model training data. Zhang's return to hands-on product leadership underscores the strategic weight ByteDance is placing on generative video and robotics-adjacent foundations.