Deploying high-energy-consumption AI computing power into orbit is transitioning from theoretical discussion to the stage of economic feasibility verification. According to a report from Deutsche Bank on January 14th, although current space deployment costs are high, this gap is narrowing at a remarkable pace. Analysts predict that with the plummeting cost of launches and the optimization of satellite design, the cost of building a space data center will converge with ground-based construction costs within the next decade. This suggests that the energy and heat dissipation bottlenecks plaguing terrestrial data centers might find their ultimate solution in space.

The rapid closing of the cost gap. Deutsche Bank's model indicates that deploying computing power in space currently remains an expensive bet, but a tipping point is already in sight. According to their calculations, the current cost of building a 1-gigawatt (GW) capacity space data center is at least 7 times that of a ground-based equivalent. However, this multiple is expected to shrink rapidly to 4 times in the latter half of this decade and ultimately reach cost parity by the 2030s. The report points out:

This cost decline is primarily driven by reductions in launch costs and further optimization in satellite design and energy efficiency, which will lead to a significant decrease in the mass required for orbital deployment.

Data shows that under a 2026 estimated scenario, space deployment costs are as high as $114 billion, compared to just $16 billion for ground-based, a difference multiple of 7.2 times. But under an "optimized 2032 scenario," space deployment costs are projected to plummet to $18 billion, only 1.2 times the ground-based cost of $16 billion, nearly reaching parity.

Plummeting launch costs are the key. The core variable enabling this economic reversal lies in the cliff-like drop in launch costs. The model assumes that the cost per kilogram to launch will plunge from $1,600 in 2026 to a mere $67 by 2032. Deutsche Bank's report emphasizes the importance of full rocket reusability and operational scaling:

We assume the current market price for launch to Low Earth Orbit (LEO) is approximately $70 million, or $4,000 per kilogram. Over time, with the achievement of full rocket reusability and operational scaling, this cost could decrease significantly to $10 million or even below $70 per kilogram.

Radical optimization of hardware specifications. Beyond the launch side, hardware evolution on the orbital end is equally aggressive. Deutsche Bank anticipates that by the 2030s, the cost of a single satellite will drop below $2 million (or just $10,000/kW). This optimized version would be equipped with a 150kW power system (including solar arrays and thermal management) and carry 150 custom chips specifically designed for space-based AI infrastructure, interconnected via optical laser terminals. It is worth noting that this model is not without its blind spots. Deutsche Bank acknowledges that its comparison baseline is built on the assumption that "ground-based capacity costs remain unchanged." Simultaneously, the model primarily compares the deployment costs of infrastructure like power, cooling, and mass, and does not include the expensive procurement costs of GPU/TPU chips. The report cautions:

Should rapid and inexpensive methods of power generation (such as nuclear energy) emerge on the ground, this assumption may no longer hold true.

This提示投资者 that the rationale for space data centers depends not only on advancements in space technology but equally on the stagnation—or lack thereof—of the ground-based energy revolution.