Alibaba bets $290 million on ShengShu's world model AI to move beyond text-based chatbots

Alibaba Investment Fuels ShengShu's Ambitious World Model Vision

Alibaba Cloud has led a 2 billion yuan ($290 million) Series B funding round for ShengShu Technology, marking one of the largest recent investments in China's competitive AI landscape 1

1

2

2

. The Alibaba Cloud funding round also drew participation from TAL Education Group, Baidu Ventures, and Luminous Ventures, while existing investors including LINK-X CAPITAL and Delta Capital increased their stakes 2

2

. This capital injection comes just two months after the Chinese AI startup secured 600 million yuan from Qiming Venture Partners and other backers, demonstrating rapid investor confidence in ShengShu's technology trajectory 1

1

3

3

.

From Language Models to Physical Intelligence

The ShengShu Technology funding represents a strategic pivot in AI development, shifting focus from large language models trained primarily on text to world model systems built on multimodal data including vision, audio, and touch 3

3

5

5

. Zhu Jun, the Tsinghua University professor who founded ShengShu in March 2023 and serves as chief scientist, explained that "a general world model, built on multimodal data such as vision, audio, and touch, more naturally captures how the physical world works than large language models" 5

5

. The general world model processes sensory information to simulate human perception and interaction, which ShengShu describes as a step toward artificial general intelligence in physical environments 2

2

4

4

. The company aims to bridge two currently separate domains: the digital world of games and AI video generation, and the physical world of autonomous driving and robots 3

3

.

Vidu's Position in the AI Video Generation Race

ShengShu became the first Chinese company to release a video generation model when it launched Vidu in April 2024, positioning itself as a competitor to OpenAI's Sora, which the U.S. company later discontinued 2

2

4

4

. The startup has since released several updated versions, including the Vidu Q3 model announced earlier this year 4

4

. Vidu currently ranks ninth on the Artificial Analysis chart for text-to-video services, trailing ByteDance's Seedance 2 and the recently released Happy Horse generator 1

1

. The platform has reached users in over 200 countries and regions worldwide, spanning industries such as animation, advertising, and film 1

1

. In 2025, ShengShu reported more than tenfold growth in both users and revenue, though specific figures were not disclosed 1

1

.

Expanding Into Robotics Applications Through Motus

ShengShu's ambitions extend beyond AI video generation into robotics applications. In December 2025, the company open-sourced Motus, a new AI model line designed to control robots by processing multimodal data including video and audio 2

2

4

4

. The Motus system enables robots and other machine-intelligence systems to better perceive and understand real-world environments 1

1

. This move positions ShengShu alongside Chinese companies ranging from industry giants like ByteDance to startups such as humanoid robot specialist Unitree, which have begun exploring similar world model technologies 2

2

.

Strategic Implications for Alibaba's Cloud Business

For Alibaba, the investment in ShengShu aligns with its broader strategy as one of the most active backers of Chinese artificial intelligence startups. The company hopes such financing will drive usage of its cloud computing platform, which has overtaken e-commerce as its fastest-growing revenue source 1

1

. Alibaba is also backing other AI video generation competitors, including PixVerse, highlighting the capital-intensive race that has pulled in tech heavyweights like ByteDance and Kuaishou Technology 1

1

. Internationally, companies such as Google and startups including Runway are developing similar technologies, intensifying global competition 4

4

. ShengShu did not provide a timeline for when its general world model system would be commercially available, leaving observers to watch how quickly the company can translate this substantial funding into market-ready products 2

2

.