Chinese tech heavyweight ByteDance is developing a real-time spatial video AI model as it enters the growing race to build systems capable of simulating interactive three-dimensional worlds, Bloomberg reported Monday.
Zhang Yiming, ByteDance’s founder, is overseeing the project and coordinating resources across the company. The model could debut as soon as next month, although the launch schedule remains subject to change.
The system would build on ByteDance’s Seedance video model and could generate virtual environments that respond to users’ voices and movements. Potential applications include livestreaming, games and short form entertainment, as well as robotics and autonomous systems.
ByteDance’s model is part of an industry-wide push toward so-called world models, with Google also developing Genie to generate interactive environments in real time. The company hopes to use its cloud infrastructure to handle the heavy computing demands of spatial content generation.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.
