China’s orbital AI data center push, led by startup Orbital Chenguang (Beijing Orbital Twilight Technology Co., Ltd.) with massive state-aligned backing, represents an ambitious, hybrid public-private effort to sidestep terrestrial constraints on AI scaling. The goal is a gigawatt-scale (1 GW+) space-based computing infrastructure in low Earth orbit by 2035, leveraging near-constant solar power and vacuum cooling. This aligns directly with national priorities in the 15th Five-Year Plan (2026–2030) and broader AI/space ambitions.
Orbital Chenguang was incubated by the Beijing Astro-future Institute of Space Technology, which is backed by Beijing’s municipal science and technology commission and Zhongguancun Science Park. It leads a consortium of 24 organizations across the supply chain (including potential ties to entities like Lenovo and Harbin Institute of Technology for energy systems).
On April 20, 2026, it completed a Pre-A1 equity round with investors including Haisong Capital, CITIC Construction Investment, Cathay Capital, and others (amount undisclosed).
It secured $8.4 billion (57.7 billion yuan) in strategic credit lines from 12 major banks, including Bank of China, Agricultural Bank of China, CITIC Bank, and others. These are framework agreements for potential future financing, not immediate cash disbursements, but they signal extraordinary institutional confidence and de-risking for a high-capital project.
Technical Plans and Timeline
The vision centers on a constellation in dawn-dusk Sun-synchronous orbit (SSO) at 700–800 km altitude. This orbit provides near-continuous sunlight (minimal eclipses) and enables passive radiative cooling into space’s thermal environment. The target over GW power capacity for AI training/inference workloads that would be impractical on the ground due to land, electricity, and cooling limits.
Phased rollout (per Beijing Astro-future Institute plans)
2025–2027: Core technology development and initial constellation launches. Demo satellite Chenguang-1 (roughly equivalent to one ground server in compute) was targeted for late 2025/early 2026. Some early test sats were reportedly lost on Ceres-2/Tianlong-3 flights in 2026.
2028–2030: Hybrid Earth-space integration (cloud-edge-terminal architecture).
By 2035: Full large-scale operational gigawatt-class space data center with modular in-orbit docking, laser inter-satellite links, and ground relays.
Chief scientist Zhang Shancong has highlighted ground-based bottlenecks are heavy land use, soaring energy consumption and limits on atmospheric cooling.
The Astro-Future Institute has received backing from electronics giant Lenovo and the municipal government of Beijing (北京), with at least 140 million Yuan (20.5 million United States Dollars) in funding. More recently, Orbital Chenguang signed a cooperation agreement with the Harbin Institute of Technology, Shenzhen (哈尔滨工业大学, 深圳) to jointly research needed energy management systems for larger space-based data centers.
Elsewhere during this week, newer space enterprises have made first steps to prove their early plans for space-based computing. Xingyong Space Chengdu Technology Co Ltd (星用空间成都科技有限公司) increased its overall available capital to 218.6 million Yuan (32 million United States Dollars), while Comospace (中科天算), Hangsheng Satellite (航升卫星), and Yanhe Technology (炎和科技) have teamed up to perform a two-satellite high-powered in-space compute demonstration by March 2027, leveraging their expertise in small spacecraft, advanced solar cells, and powerful space-rated processors.
ADA Space (国星宇航) and Zhejiang Laboratory (之江实验室) have placed twelve satellites into orbit for the Three-Body Computing Constellation, which has run several different AI models and agents to date. 2026 should see the constellation triple in satellite count.
China Ground Based Power
China currently has 339 gigawatts of wind and solar capacity under construction — roughly two-thirds of the world’s total.
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
You ever get the impression that China’s industrial policy is just, “Do whatever Musk is doing.”?
True, Brett. And that is what makes it so dangerous. Just one small example. Tesla introduced gigacasting for cars, and European and (other) American car makers just stood by, but the chinese car makers just copied it as fast as possible. And of course, that is one small reason why their cars are so much cheaper than Europen/American cars.
Same thing with megapacks, FSD SW, only camera suite (no LIDAR), Optimus… you name anything Elon Musk does, they copy it.
I hate so say it, but as long as you are within the law (and many times the Chinese companies are not, of course), the intelligent strategy is to swallow your pride and copy best in class as soon as possible. European and American companies practice the “not invented here” and suffers as a consequence.
Love to see a comparison of this with Elon’s space datacenter plans.