Forget about the land. The most competitive battlefield for AI in the future lies in the "space data center". Elon Musk will build an ecological closed-loop.
Senior technology investor Gavin Baker believes that space data centers will become the most important technological breakthrough in the next three to four years, and it may take five to six years for them to become the mainstream for deploying computing power. From a first - principles perspective, the advantages of space data centers stem from three core elements: energy, cooling, and chips. The in - depth integration of SpaceX, Tesla, and xAI under Elon Musk's leadership will provide unique advantages for the development of space data centers.
Gavin Baker believes that space data centers will become the most important technological breakthrough in the next three to four years.
On December 9th, senior technology investor Gavin Baker stated in a recent podcast interview that from a first - principles perspective, space data centers are superior to those on Earth in every aspect.
According to Gavin Baker's judgment, it may take five to six years for space data centers to become the mainstream for deployed computing power. Before that, ground - based infrastructure will still be the main support for AI development.
The advantages of space data centers stem from three core elements: energy, cooling, and chips. In the space environment, the cost structure and efficiency of these three elements will undergo fundamental changes. Satellites can stay in sunlight 24 hours a day, with the solar radiation intensity 30% higher than that on Earth, and the total irradiance reaching six times that of Earth, without the need for battery energy storage.
Baker pointed out that this development trend is closely related to the technological progress of SpaceX under Elon Musk's leadership. When SpaceX, Tesla, and xAI are deeply integrated, they will provide unique advantages for the development of space data centers.
01 Technological Advantages of Space Data Centers
Based on first - principles analysis by Gavin Baker, the core inputs for operating a data center are electricity, cooling, and chips. In the space environment, the cost advantages of the first two are extremely significant.
In terms of energy, satellites can maintain 24 - hour sunlight through orbital design, and the sunlight intensity is 30% higher than that on the Earth's surface, making the total irradiance in space six times that of Earth. More importantly, due to continuous power supply, the system does not need to be equipped with batteries, which account for a huge cost in ground - based data centers. Therefore, the lowest - cost energy in the solar system is actually space solar energy.
Cooling systems account for most of the weight and volume of racks in ground - based data centers, involving complex HVAC systems, cooling distribution units, and liquid - cooling technologies. However, in space, cooling is almost free - simply install a radiator on the back - lit side of the satellite and use the space environment close to absolute zero to achieve efficient heat dissipation. This eliminates one of the most complex and costly components in ground - based data centers.
In terms of network connection, space data centers also have advantages. The racks in ground - based data centers are connected by optical fibers, which is essentially the transmission of lasers in optical cables. In space, lasers can be used to directly connect satellites in absolute vacuum, which is faster than fiber - optic transmission and can build a faster and more coherent network than Earth - based data centers.
From the perspective of user experience, space data centers can also shorten the data transmission path. Current AI queries need to pass through mobile phones, base stations, fiber - optic networks, and metropolitan aggregation facilities before reaching the data center. After the calculation is completed, the data returns along the same path.
If satellites can directly connect to mobile phones - Starlink has already demonstrated the ability to directly connect to cellular networks - data transmission will be greatly simplified, providing users with a lower - latency and lower - cost experience.
02 Implementation Paths and Technological Challenges
The main obstacle for space data centers is launch capacity. A large number of Starships are needed to transport data center equipment into orbit, which is the key bottleneck for realizing this vision.
SpaceX leads in this field, and the Starship is the only launch vehicle that can economically support the construction of space data centers.
Notably, Elon Musk recently indicated that Tesla, SpaceX, and xAI are moving towards integration. Baker believes that this provides unique advantages for the development of space data centers.
In this integrated system, xAI will become the intelligent module of Tesla's Optimus robots, Tesla's vision system will provide perception capabilities, and SpaceX will build data centers in space to provide computing power support for xAI, Tesla, and other customers. This vertical integration creates significant competitive advantages, with each company creating value for the others.
For training models, due to their large scale, the application of space data centers may take a long time.
For inference tasks, the application prospects of this technology are clearer. Each satellite can be regarded as a rack, or larger satellites can accommodate equipment equivalent to three racks. Baker said that as the technology matures, even training tasks will eventually be migrated to space.
This article is from the WeChat official account "Hard AI", author: Focusing on technology R & D. Republished by 36Kr with authorization.