Amazon took a major step forward in its satellite connectivity strategy on Friday, July 24, 2026, filing an application with the Federal Communications Commission (FCC) to deploy up to 5,105 next-generation satellites designed to provide direct cellular connectivity to ordinary mobile devices — no special antenna or hardware required. The filing, revealed by multiple sources on Monday, represents the company's most ambitious attempt yet to compete directly with SpaceX's Starlink in the emerging direct-to-cell satellite market.
The new constellation, operated by Amazon Leo (formerly Project Kuiper), will use the radio frequency spectrum acquired through Amazon's $11.6 billion purchase of Globalstar in April 2026. The acquisition, approved by Globalstar shareholders in June, gave Amazon access to L-band and S-band spectrum — globally licensed frequencies ideal for direct communications with mobile devices. Unlike the broadband satellites of the original Kuiper constellation (which require dedicated terminals), these new satellites will be optimized to communicate directly with standard 4G and 5G phones.
The timing of the filing is no coincidence. SpaceX already offers its Starlink "Direct to Cell" service in partnership with carriers like T-Mobile in the US, and AST SpaceMobile has its own test satellites in orbit. Amazon enters this market with one significant advantage: Globalstar's spectrum is already used by emergency and tracking devices worldwide, which could accelerate regulatory approval. Additionally, the company already has 3,236 Kuiper satellites authorized by the FCC for broadband, hundreds of which are already in orbit — providing an infrastructure base that competitors lack.
Amazon's proposal is ambitious not only in the number of satellites but in the breadth of planned services. According to FCC documents obtained by the press, the D2D constellation will initially offer text messaging and emergency calls, later evolving to voice, data, and industrial IoT. The company states the system will allow mobile operators to extend coverage to remote areas currently without signal — a market estimated at over 400 million people globally.
Deployment is expected to begin in 2028, but the timeline depends on regulatory approval and hardware qualification tests. The FCC will need to evaluate complex issues including spectrum coordination with existing satellites, potential interference with terrestrial services, and space debris mitigation. Amazon argues its satellites will be designed to deorbit within five years of end of life, following industry best practices.
The move also carries geopolitical implications. The ability to connect any ordinary mobile phone directly via satellite has obvious applications in defense, disaster response, and national security. With SpaceXAI also active in the sector and AST SpaceMobile holding contracts with the US Department of Defense, Amazon needs to demonstrate it can deliver comparable capability. The Globalstar acquisition — whose spectrum is used by emergency devices like SPOT beacons — aligns perfectly with this narrative.
The open question is whether Amazon can execute on its promised timeline. Its Project Kuiper track record is mixed: early prototypes faced delays and higher-than-expected costs, and the company has yet to offer commercial broadband satellite service. If Amazon can launch the D2D constellation by 2028, it will have done in two years what took SpaceX four to build. If it cannot, the Globalstar spectrum and the authorized 5,105 satellites will remain valuable — but underutilized — assets.
Sources: SatNews, CNBC, ISPreview UK, About Amazon
