American military space closed around one company in seven days

By Nick David, Editorial Lead, SatNews

Four events landed in defense space between May 26 and May 29. The U.S. Space Force awarded SpaceX a $2.29 billion Other Transaction Authority contract on May 26 to build the Space Data Network Backbone on its Starshield platform. The same day, the House Armed Services Committee released the chairman’s mark of the fiscal 2027 National Defense Authorization Act, dissolving the Space Development Agency and the Space Rapid Capabilities Office as standalone organizations. On the evening of May 28, Blue Origin’s NG-4 — the fourth New Glenn flight article — exploded during a static fire test at Space Launch Complex 36, severely damaging the only commercial heavy-lift pad outside SpaceX’s certified inventory available to absorb the NSSL Phase 3 manifest. On May 29, the Space Force awarded SpaceX a second contract worth $4.16 billion for the Air Moving Target Indicator (AMTI) constellation, also on Starshield. The trade press has covered each on its own terms.

Read together, these four events are the seven days in which American military space lost its remaining alternatives, and in which the Pentagon’s twenty-year attempt to keep defense space acquisition disaggregated and multi-vendor came to an end.

The Space Development Agency was established in 2019 by then Under Secretary of Defense for Research and Engineering Mike Griffin with one mandate: prevent any single contractor from owning the LEO defense layer. Its design principle was simple. Buy hundreds of small satellites from competing vendors on two-year spiral cycles, refuse to lock in any single prime, and force traditional defense contractors to compete with new entrants on cost and cadence. Through Tranches 0, 1, and 2 the model worked on its own terms. York Space Systems, Lockheed Martin, Northrop Grumman, Rocket Lab, L3Harris, and Sierra Space all won material work. Bus suppliers like Terran Orbital and Blue Canyon gained scale they would not otherwise have reached, scale that has since been absorbed into Lockheed Martin and RTX respectively.

That model is being unwound on four fronts at once.

First, the Tranche 3 Transport Layer is gone. When the Space Force unveiled its fiscal 2027 budget in April, Tranche 3 Transport was unfunded; its requirements were rolled into the new Space Data Network program. The Tranche 2 incumbents face a procurement that has been functionally replaced, not re-baselined.

Second, the $2.29 billion SDN Backbone award places SpaceX’s Starshield variant at the center of that architecture. The contract is a firm-fixed-price OTA delivery order with an operational prototype due by the end of 2027. The Space Force itself describes the SDA Transport Layer as an enclave within the SDN. In network engineering terms an enclave is a tenant on someone else’s pipe. SpaceX owns the pipe.

Third, SDA itself is being legislated out. The HASC mark folds SDA and the Space Rapid Capabilities Office into the Space Force acquisition structure established under Secretary Hegseth’s 2025 reforms. The dual-hatting was already in motion. Gurpartap Sandhoo, named SDA director on May 19, was simultaneously given the Space Force’s portfolio acquisition role for missile warning and tracking — anticipating the consolidation rather than waiting for it. The committee debates the bill on June 4.

Fourth, the heavy-lift alternative itself went offline. The New Glenn pad explosion takes SLC-36 out of service for what early industry assessments place at roughly a year. Blue Moon, already two years behind its competitor, loses any realistic claim on Artemis III. For National Security Space Launch Phase 3, the practical competitive set narrows from three providers to ULA Vulcan and SpaceX, and Vulcan’s class-A cadence cannot absorb the SpaceX manifest. The Tracking Layer primes who fly to orbit on whatever the Space Force certifies, $1.1 billion of Lockheed Martin and $764 million of Northrop Grumman in Tranche 3 alone, now ride a backbone they do not control on a rocket whose nearest competitor is grounded.

There is a precedent the industry has stopped citing. The Operationally Responsive Space Office was established at Kirtland in 2007 to do exactly what SDA was created to do: fast, cheap, multi-vendor tactical space. It was folded into the Space and Missile Systems Center in 2013, then renamed the Space Rapid Capabilities Office in 2018, which is itself one of the two agencies the HASC mark now dissolves. SDA is the third iteration of the same idea. Each prior iteration was absorbed by the acquisition system it was meant to disrupt.

The strongest counterargument is the one the Space Force itself has made. The service has stated that SpaceX will not be the sole supplier, that an IDIQ pool exists for AMTI follow-ons, and that the SDN is intended as an open architecture with multiple participants over time. The pad loss, the argument continues, is a contained industrial accident Blue Origin will work through.

That argument fails on the schedule. The “open architecture” framing is what every prime backbone procurement has promised before lock-in, and there are two histories that show how the framing collapses. GPS receiver standards converged onto a small set of incumbents despite a published RF interface, because once the integration certification process favored Rockwell Collins, Trimble, and a handful of others, the standard remained nominally open while the supplier set hardened. The F-35’s open mission systems architecture has the same shape. Mission-data files, integration test rigs, and Block 4 software dependencies anchored Lockheed Martin as the de facto integration authority within the first decade of the program, and Congress is now legislating an unlock that has not yet been delivered almost twenty years in. The SDN starts in the same posture, with one difference. SDA had a real multi-prime base when its program began. The SDN does not. The relevant question is not whether the SDN architecture is nominally open. It is whether a second vendor can deliver an operationally interoperable backbone, and a second heavy-lift provider can be recertified for class-A national security payloads, before SpaceX’s 2027 prototype goes live and the Phase 3 manifest reweights toward whoever is flying. Neither is on contract to deliver.

The verdict is already written into the week’s events. The Department of Defense has chosen single-supplier speed over multi-prime competition, and that choice was sealed not by the SDN Backbone award alone but by the loss of New Glenn pad capacity in the same week, because the award would not matter if a second rocket could move the integration schedule. The Tracking Layer primes will become payload providers on a SpaceX-routed network and a SpaceX-launched manifest, with margins to match. Bus demand consolidates behind whichever primes maintain backbone integration certifications. Congress’s June 4 NDAA debate is no longer about whether SDA’s nameplate survives — it is about whether the consolidation stands. The question is not how to compete with Starshield. It is what to do about a stack that is now end-to-end SpaceX. The disruption has been dismantled.

---

About the Author

A storyteller at heart, Nick David covers space policy, satellite markets, defense, and the technologies reshaping how humanity operates beyond Earth. With a background in creative direction, brand strategy, and editorial storytelling, he brings a modern lens to complex subjects and a relentless curiosity about what comes next.