AI models to propel investments in computing infrastructure in China
Chinese tech heavyweight Baidu Inc is doubling down on artificial intelligence-powered video generation models, as fast-growing AI-generated content is reshaping the landscape of short video, film, animation and advertising industries.
Experts said image-to-video generators will improve the quality and efficiency of video generation and significantly reduce the production costs of video, while creating new business opportunities for online searches, cloud computing and the content ecosystem.
Baidu has launched its updated AI-driven video generation model MuseSteamer 2.0, which is designed to empower businesses with the ability to create high-quality and synchronized audiovisual content from simple image inputs and prompts.
Featuring computer vision and natural language processing capacities, the model is able to generate multiple voices and videos simultaneously and overlay realistic ambient sounds with dialogue from various characters, lowering the barrier to premium video production for businesses and content creators.
The company said robust application demands from various commercial scenarios are driving the speedy development of video generation models, and the emergence of AIGC has greatly bolstered multimodal content creation with broad application prospects in fields such as marketing, as well as film and television creation.
The move comes as Baidu aims to carve out a niche in China’s growing enterprise-level AI services market, where demand for multimedia content creation tools is rapidly rising.
“AIGC-related technologies will improve the productivity of content production, but the process of developing image-to-video generation models necessitates higher requirements for computing capacity, algorithms and high-quality data,” said Pan Helin, a member of the Expert Committee for Information and Communication Economy, which is under the Ministry of Industry and Information Technology.
Chinese tech companies should beef up self-developed and proprietary abilities in underlying computing power chips and programming software, as well as step up investments in basic scientific research, to catch up with foreign counterparts in the AI chatbot race, he said.
Noting that talent, data and computing power are key to image-to-video generation models, Pan added that more efforts are needed to bolster the efficient circulation of data elements, try to achieve breakthroughs in key technologies and expand application scenarios of video generation models in a wider range of segments.
Chen Duan, director of the Digital Economy Integration Innovation Development Center at the Central University of Finance and Economics, said AIGC will unleash people’s demand for tools that can help them express their creativity and drive the evolution of commercial models in content production.
Chen said the multimodal large language model, which possesses the ability to generate high-resolution video clips based on given prompts, is an undeniable future development direction for AI technology.
AIGC will lead to a new revolution in the field of digital content production and boost innovation in the digital culture industry, Chen said, adding that although AIGC is still nascent, she is bullish on its prospects.
Other Chinese tech companies are accelerating steps to launch video generators. Video-sharing platform Kuaishou Technology has updated its video generation model Kling that can mimic the physical world and create imaginative scenes from text instructions, while ByteDance has unveiled its AI model for text-to-video generation.
Experts said the text-to-video or image-to-video AI models will further propel China’s investment in computing infrastructure, such as data centers and cloud computing platforms, and boost the integration and upgrade in China’s AI industrial chain.
Global market research firm International Data Corp said the scale of China’s AI market is expected to reach $26.44 billion in 2026, with a compound annual growth rate of over 20 percent between 2021 and 2026.
While there are many positives, the use of such models raises concerns about ethics, copyright protection, personal privacy leakage and data security.
How to ensure the authenticity and transparency of the content has become an important issue, and more efforts are needed to formulate rules and regulations to ensure the healthy development of such technology, said Liu Xingliang, director of the Beijing-based Data Center of China Internet.
