Elon Musk’s SpaceX could merge with xAI in a move that may accelerate his ambitions to develop space-based AI data centres, according to a Reuters report. The proposed partnership aligns with Musk’s broader goal of creating satellite clusters capable of handling intensive artificial intelligence workloads, positioning him against competitors like Alphabet’s Google, Meta, OpenAI, and other technology leaders investing in AI infrastructure.
Space-based data centres remain largely conceptual, but the model would rely on fleets of solar-powered satellites in orbit, offering continuous power and reducing the cooling challenges that drive up costs for Earth-based facilities. Advocates argue that orbital AI computing could make processing more efficient, while critics highlight technical and economic hurdles, including space debris management, cosmic radiation protection, limited maintenance options, and the high cost of satellite launches. Deutsche Bank predicts that small-scale orbital data centre tests may begin in 2027–28, with larger constellations only emerging in the 2030s if early deployments succeed.
Musk has long emphasized the strategic advantage SpaceX holds due to its established rocket and satellite capabilities. With thousands of Starlink satellites already in orbit, SpaceX could create AI-ready clusters or facilitate deployment for other operators. Speaking at the World Economic Forum in Davos earlier this month, Musk described space as the “lowest-cost place to put AI” and predicted operational viability within two to three years. Sources also report that SpaceX is considering an initial public offering in 2026 with a potential valuation exceeding $1 trillion, with a portion of the proceeds earmarked for funding AI satellite development.
The interest in space-based AI computing extends beyond SpaceX. Blue Origin has explored orbital data centres, following Jeff Bezos’ vision of solar-powered facilities that bypass Earth-bound energy constraints. Nvidia-backed Starcloud recently launched Starcloud-1 aboard a Falcon 9, demonstrating AI model training in orbit using an Nvidia H100 chip. Starcloud envisions a modular “hypercluster” of satellites capable of delivering computing power equivalent to multiple hyperscale data centres. Google is pursuing a similar approach through Project Suncatcher, planning to deploy satellites equipped with Tensor Processing Units to create an orbital AI cloud by 2027. China has also announced plans for “Space Cloud,” aiming to establish gigawatt-scale space-based AI infrastructure within five years.
As global tech leaders invest in orbital computing, Musk’s SpaceX-xAI collaboration could serve as a significant step in shaping the next frontier of artificial intelligence. By leveraging SpaceX’s launch capabilities and xAI’s AI systems, the project represents a convergence of space exploration and AI innovation, potentially redefining how large-scale computing is delivered and maintained. While commercial deployment remains several years away, early demonstrations may provide valuable insights into the efficiency, reliability, and feasibility of orbital AI operations.
Follow the SPIN IDG WhatsApp Channel for updates across the Smart Pakistan Insights Network covering all of Pakistan’s technology ecosystem.
