China-developed text-to-video large AI model unveiled in Beijing

Vidu, a text-to-video large AI model capable of creating a 16-second, high-definition video in 1080p resolution with a single click, was unveiled on Saturday at the 2024 Zhongguancun Forum in Beijing.

Developed by Tsinghua University and Chinese AI firm ShengShu Technology, Vidu is China's first text-to-video large AI model with "extended duration, exceptional consistency and dynamic capabilities."

As a large AI model developed in China, Vidu can understand and generate Chinese content, such as the panda and the Chinese dragon, according to Zhu Jun, deputy director of the Tsinghua Institute for Artificial Intelligence.

AI school established

Tsinghua University established the School of Artificial Intelligence on the same day, with a focus on core fundamental theories and architectures of AI and "AI + X" compound talent cultivation.

It aims to build a high-level base for China's top AI talent and original innovation through new mechanisms, offering strong support for achieving high-level technology self-reliance.

Turing Award winner and member of the Chinese Academy of Sciences Andrew Chi-Chih Yao will serve as the first dean of the school.

The newly established college will develop top talent by innovating talent recruitment mechanisms and fostering a training system based on AI theory while accommodating interdisciplinary "AI + X" talents.

It is expected to achieve breakthroughs in basic research and key core technologies, solidify China's new-generation AI development foundation and promote the application of AI in various industries. The school also plans to uphold high-level international exchanges and collaborations in the field, leveraging the university's comprehensive disciplinary advantages and Beijing's industrial development advantages to establish a top-notch hub for AI talent and innovation worldwide.

(With input from Xinhua)