The Pentagon's Eye in the Sky Goes to Orbit
A $4.16B SpaceX deal aims to put air-tracking sensors in low Earth orbit by 2028—years ahead of schedule and decades ahead of doctrine.
By The Defense Desk
For decades, the job of watching the skies for hostile aircraft fell to a handful of large, slow, expensive jets circling at altitude—visible to adversaries, vulnerable to missiles, and finite in number. That logic is being dismantled from above. The U.S. Space Force has awarded SpaceX a $4.16 billion agreement to build a Space-Based Airborne Moving Target Indicator (SB-AMTI) constellation, with an early operational capability targeted for 2028 [1]—a timeline that, until recently, no one in uniform was seriously discussing.
The previous consensus placed space-based AMTI capability somewhere in the 2030s. Officials spoke about it the way defense planners speak about most transformational programs: cautiously, conditionally, with generous timelines padded for political and technical uncertainty [1]. The new SpaceX deal collapses that window by the better part of a decade, and it arrives not as a quiet procurement action but as a statement of intent about where the U.S. military believes the future of air surveillance actually lives.
The Space Force framed the urgency plainly. "The long-standing method of military airborne platforms to track moving targets faces continued challenges as adversaries develop increasingly sophisticated anti-access/area-denial (A2/AD) systems," the service said in its announcement [1]. The answer it is pursuing: "a persistent, global capability to sense and track airborne targets from space" [1]—one that does not require an aircraft, a runway, or an aircrew anywhere near the threat.
The deal is structured as a "competitive Other Transaction Authority (OTA) agreement" rather than a traditional contract, routed through the Portfolio Acquisition Executive for Space-Based Sensing and Targeting [1]. That mechanism allows the Space Force to move faster than standard procurement rules permit, and to bring in non-traditional vendors alongside established defense contractors. The agency has been explicit that SpaceX will not be the sole provider: it has established a broader vendor pool, with Space Force Col. Ryan Frazier, the acting PAE SBST, stating that the service will partner with "a highly diversified pool of traditional and non-traditional vendors, each bringing various capabilities to support the SB-AMTI architecture" [1].
Even so, the gravitational pull toward one company is hard to ignore. When a single provider commands so much of the available launch capacity, the architecture of national security starts to rhyme with the architecture of a private enterprise.
The Space Force has acknowledged the imbalance directly: at least currently, no other company has the same capacity to provide the U.S. military with the kind of reliable access to space at the required cadence, and within budget constraints [1]. SpaceX's role also extends beyond the sensor satellites themselves. Its Starlink and Starshield networks are already playing an increasingly central role in the communications architecture that would move targeting data from orbit to the warfighter [1]. Laser-based communications relays are identified as another key supporting capability [1].
The SB-AMTI program is already a significant line item. The Space Force is asking for more than $7 billion in additional funds to procure additional elements of the system in its fiscal year 2027 budget request [1]. That figure sits alongside whatever classified expenditure has already supported prototype on-orbit sensor testing—work that has been underway for at least a year, possibly much longer, but has been heavily classified [1]. The National Reconnaissance Office, America's main remote sensing intelligence arm, is a key player in this ecosystem, though the details of its contributions remain shrouded in secrecy [1].
The technical gap between tracking things on the ground and tracking things in the air is not trivial. Chief of Space Operations Gen. Chance Saltzman addressed this directly during a press briefing on the sidelines of a conference in December 2025, according to Breaking Defense. "So GMTI and AMTI sound like they're really close, just because one little letter that is all you changed, [but it] turns out they're pretty different," he said. "Things on the ground move slower than things on the air, so [they] require different levels of fidelity tracks" [1]. A spokesperson for the NRO told Breaking Defense earlier this year that the AMTI data warfighters need presents "a multi-phenomenology challenge that requires automated orchestration of the NRO's collectors, low-latency data transport, and rapid data fusion" [1].
Speed and fidelity are not interchangeable virtues in sensing systems. Getting both simultaneously, from orbit, against fast-moving aerial targets, is the engineering problem that has kept this capability theoretical for so long.
What a mature SB-AMTI constellation could actually do to the operational landscape is significant.
The resilience argument is equally compelling from a force-design perspective. Unlike a single surveillance aircraft orbiting in a known pattern, a constellation made up of a large number of individual satellites is inherently harder to suppress [1]. Shooting down one satellite does not blind the network. Jamming one node does not silence the architecture. The distributed nature of the system is itself a form of survivability that no airborne platform can replicate at equivalent cost.
Machine learning and artificial intelligence are woven into the vision for what this network becomes at scale. Greater collaborative capabilities enabled by AI technologies could help find targets of interest and anomalies far faster than current methods allow, and could open a path toward more autonomous collection, tasking, and retasking of satellites, without the need for human deconfliction and even direct operator direction [1]. Areas of interest that need seamless coverage could have extra satellites retasked to the necessary orbit in order to do so automatically. The phrase used in planning documents is "panoptic or near panoptic targeting and surveillance from space" [1].
The political subplot running through all of this involves the E-7 Wedgetail, the Boeing aircraft the Air Force selected to replace its aging E-3 Sentry AWACS jets. In the past year, plans for SB-AMTI were directly tied to an attempt to cancel E-7 purchases entirely—an argument that space could do the job sooner and cheaper. Congress intervened, the cancellation was abandoned, and the E-7 program is now moving forward again [1]. Then-Air Force Maj. Gen. Christopher Niemi, who has since been promoted to lieutenant general and is currently Deputy Chief of Staff of the Air Force for Force Modernization and the service's Chief Modernization Officer, offered a hint about why the space advocates were so confident. "The capabilities that are happening in space are far exceeding our expectations," he said at a hearing earlier this year, declining to elaborate publicly [1].
The E-7 and SB-AMTI are now framed as complementary rather than competing—a layered architecture in which orbital sensors backstop airborne platforms during the transition period, and eventually carry the bulk of the mission load once the constellation matures. The Space Force's own language describes the goal as a "layered, highly resilient tracking architecture" [1]. Traditional aerial AMTI capabilities are expected to remain an important element of U.S. military operations for the foreseeable future [1], even as the longer arc bends toward orbit.
The SB-AMTI program also connects directly to the Golden Dome missile defense initiative, which has already drawn on programs under the PAE SBST umbrella to accelerate fielding of an initial capability layer [1]. The logic is cumulative: sensors that can track fast-moving aerial targets are also sensors that can support missile defense. As those functions converge in orbit, the distinction between air surveillance and missile defense infrastructure starts to blur.
The bureaucratic lines drawn between services, between mission sets, between offense and defense, were always artifacts of the atmosphere. In orbit, those distinctions become harder to maintain—and harder to explain to an adversary.
The $4.16 billion deal with SpaceX is not a bet that space-based air tracking will work. The prototype testing already underway, the classified NRO programs, the AI-driven fusion architectures—these suggest the bet has already been placed. What the new agreement actually signals is that the Space Force believes 2028 is achievable, that the technology is ready enough to commit to at scale, and that the era of the airborne radar jet as the primary eye of the joint force is entering its final act. Whether the constellation delivers on its early capability promise in two years or slips toward the old 2030s timeline will determine whether orbit truly becomes the nervous system of American air power—or remains, for a while longer, a very expensive aspiration.