Blue Origin Seeks FCC Approval for Project Sunrise, a 51,600-Satellite Orbital Data Center Constellation

Blue Origin Files to Launch Massive Space-Based Data Center Network March 20, 2026 The race to establish computing infrastructure in orbit has intensified as Blue Origin, the aerospace company founded by Jeff Bezos, has formally applied to U.S. regulators for a constellation designed to host data centers in space. The move signals a pivotal shift in how major tech and aerospace players envision the future of scalable, low-latency computing, seeking to bypass terrestrial limitations of land, power grids, and environmental regulations. According to an application filed with the Federal Communications Commission (FCC), the initiative, dubbed Project Sunrise, involves deploying a network of up to 51,600 satellites into sun-synchronous orbit. These satellites are designed to function as interconnected data centers, utilizing optical links to communicate within the constellation and connect to Blue Origin’s planned TeraWave satellite communications network. In its filing, the company argued that “space-based data centers will be a complement to terrestrial infrastructure by introducing a new compute tier that operates independently of Earth-based constraints,” highlighting the advantages of perpetual solar power and the absence of land costs or grid dependencies. Blue Origin’s business model and technical approach bear a striking resemblance to that of its primary competitor, SpaceX. Earlier this year, SpaceX submitted its own proposal to the FCC for a staggering one million “orbital data center” satellites, which would integrate with its existing Starlink internet constellation via laser links. Elon Musk has previously claimed Starlink’s laser transfer capabilities would surpass TeraWave’s projected data rate of up to 6 terabits per second. Both companies contend that orbital data centers could offer cheaper, more efficient compute capacity while navigating a simpler regulatory landscape compared to ground-based builds. The broader industry is mobilizing to support this emerging frontier. This week, Nvidia announced it had developed a space-optimized module of its Vera Rubin GPU-CPU platform. They are joined by a cohort of smaller firms, including Axiom Space, NTT, Ramon.Space, Aetherflux, and Sophia Space, all exploring orbital data center concepts. A preliminary step was taken last November when Starcloud (formerly Lumen Orbit) launched a single Nvidia H100 GPU into space as a technology demonstrator. Despite the growing momentum, the concept faces significant skepticism from industry analysts and leaders. Critics, including OpenAI CEO Sam Altman, analytics firm Gartner, and short-seller Jim Chanos, have respectively dismissed the plans as “ridiculous,” “AI Snake Oil,” and “peak insanity,” citing immense technical hurdles and prohibitive costs associated with operating and maintaining sensitive server hardware in the harsh environment of space. The debate underscores the high-stakes gamble as the industry weighs the potential for a fundamental architectural shift against formidable practical and economic challenges. Source: datacenterdynamics