Starlink crosses 10k sats

SpaceX’s Starlink network has passed a threshold that would have sounded absurd a few years ago. The company now has more than 10,000 active satellites in orbit, a figure it crossed on March 17 after launching two Falcon 9 missions in a single day. Tracker Jonathan McDowell counted 10,049 Starlinks in orbit after those launches, with only 10 not in working order. More than 1,500 earlier Starlinks have already reentered and burned up since the network began in 2019. (space.com) That number matters because it turns Starlink from a big satellite system into a new kind of infrastructure. The old mental model for space operations was a handful of expensive spacecraft, each treated like a national asset. Starlink works the other way. It is a mass-produced swarm in low Earth orbit, refreshed constantly, expanded constantly, and managed as a living network rather than a fixed fleet. The FCC’s January 9, 2026 authorization for additional Gen2 deployments shows how far that model has spread. SpaceX is now authorized for a 15,000-satellite second-generation constellation, with shells as low as 340 kilometers and as high as 525 kilometers. (docs.fcc.gov) Once a constellation gets that large, collision avoidance stops being a side function and becomes the job. Starlink has spent years building for that reality. SpaceX describes its system as automatic and multi-tiered, with trajectory planning designed to keep satellites away from crewed stations before launch and onboard collision-avoidance logic ready to react later if predictions change. NASA’s conjunction process still depends on regular screenings and warning messages from the U.S. Space Force. Starlink adds another layer on top of that because a network this dense cannot wait for humans to hand-fly every close approach. (starlink.com) The maneuver counts show what “dense” now means. SpaceX reported 50,666 collision-avoidance maneuvers in the six months ending November 30, 2024. Then it reported 144,404 maneuvers in the six months ending May 30, 2025. New Scientist reported that the next filing, submitted on December 31, 2025, covered about 149,000 maneuvers from June through November 2025. Those figures are not just signs of danger. They are signs of a system that has accepted constant motion as normal operating practice. (kratosspace.com) That is why the card’s detail about chip strategy matters. In low Earth orbit, radiation is real but the economics are different from deep-space missions that must survive for many years without replacement. The broader industry has moved toward mixes of radiation-hardened parts, redundancy techniques, and carefully screened commercial components to cut cost and speed production. Triple-modular redundancy is a standard way to keep logic working through faults. COTS-based radiation-tolerant designs are now explicitly marketed for “New Space” systems that need volume and fast iteration more than exquisite longevity. (militaryaerospace.com) That trade is one reason low Earth orbit now dominates so much operational thinking. LEO wins where latency matters, where capacity must scale fast, and where replacing hardware is easier than preserving it for 15 years. GEO still has advantages, and deep-space networks still do jobs that Starlink cannot. NASA’s Artemis II communications plan, for example, still hands primary support to the Deep Space Network after Orion leaves near-Earth space. But for broadband, direct-to-device service, and proliferated sensing or relay architectures, the center of gravity has shifted downward. Even SpaceX now frames Starship’s next role in those terms, saying its larger V3 Starlink satellites will soon be launched in batches with more than 20 times the capacity of current Falcon-carried V2 satellites. (nasa.gov)