ByteDance founder Zhang Yiming is personally overseeing development of a real-time spatial video generation AI model expected to launch as early as next month. Built on ByteDance’s existing Seedance video generation model, the system would let users create interactive virtual worlds for live streaming, short dramas, and games. The model places users in a three-dimensional space with latency of about 0.05 seconds and 20 frames per second. ByteDance aims to reduce VR hardware costs by shifting compute-intensive content generation to the cloud, while moving a large portion of its B2B business toward video production and generation after previously securing a $30 billion loan for AI capability expansion.