SDA Misses Orbital Mesh Network Deadline, Pushing AI-Enabled Defence Architecture Back a Year

The Space Development Agency has missed its self-imposed deadline for completing an orbital mesh network, pushing the critical satellite-to-satellite data relay system at least a year behind schedule. SDA Director G.P. Sandhoo revealed the slip at the Air Force Association's 2026 Air, Space and Cyber conference in National Harbor, Maryland, marking a significant setback for the Pentagon's ability to field resilient, AI-ready space architecture on schedule.

The Deadline Miss and Current Status

Sandhoo told reporters that the SDA has not yet begun construction on the orbital mesh network, despite a self-imposed deadline that was set to expire within days of his announcement. Just over six months prior, at the Satellite 2026 conference in Washington, D.C., Sandhoo had stated the SDA was "about three months behind" and hoped to start building the network within six months. That timeline has now shifted dramatically.

"We have not started yet," Sandhoo said bluntly. "We are behind on that. We should be there already." He indicated the SDA does not expect to begin construction for another three to four months, with that window now serving as the agency's updated priority. "I push on the team every week," he continued. "The next step is to get that done. In the next three to four months, that's the number one priority for us."

The total delay now stands at approximately one year from the original self-imposed target, representing a substantial slip in a programme designed to underpin next-generation military operations.

Why the Network Matters to Defence Architecture

The orbital mesh network is not merely a communications upgrade—it is foundational infrastructure for the Pentagon's AI-enabled multi-domain warfare strategy. The network is designed to streamline data transfers across the Proliferated Warfighter Space Architecture (PWSA), which encompasses hundreds of satellites providing missile warning, tracking, and fire control capabilities.

The architecture works by allowing satellites to transfer data directly to each other using laser beams rather than radio waves, eliminating the need to relay signals through ground stations. This approach dramatically reduces latency and increases resilience. For military autonomous systems—including autonomous targeting platforms, sensor-fusion algorithms, and machine-speed command-and-control systems—this low-latency, redundant orbital connectivity is essential. Every month the network slips is a month that AI-driven defence concepts remain theoretical rather than operational.

The longer-term vision for the PWSA includes adding edge computing and artificial intelligence to the satellites themselves, enabling them to digest vast amounts of data in orbit and conserve bandwidth for space-to-Earth transmission while accelerating the flow of priority data. However, Sandhoo made clear that the immediate focus must remain on launching the basic satellite constellation. "We plan to leverage all of that as we get into this," he said. "But we are not going to start with transformative AI. We've still got to get the satellites up first."

The Supply Chain Bottleneck

The primary cause of the delay is a supply chain constraint affecting optical communication terminals (OCTs)—the specialized components that enable satellite-to-satellite laser communication. These optical heads must be engineered with extreme precision to send and receive clear signals while remaining robust enough to survive launch and orbital operations.

For the initial group of PWSA satellites, known as Tranche 1, the SDA awarded all firm fixed-price contracts for optical communication terminals to a single supplier in an effort to manage costs. However, that approach has created a critical vulnerability. "That vendor is having issues with throughput and not able to deliver," Sandhoo explained. "I have not seen the improvement that I wanted to see in Tranche 1."

For Tranche 2 satellites, the SDA has implemented risk mitigation by creating a parallel procurement path, allowing the agency to source OCTs from a second vendor if needed and maintain design flexibility. This approach aims to avoid repeating the single-vendor dependency that has constrained Tranche 1 delivery, though it does not immediately resolve the current delays.

Implications for Autonomous Systems Integration

The slip in the orbital mesh network timeline has direct consequences for the deployment of AI-enabled autonomous weapons and sensor platforms across all military domains. Autonomous targeting systems, whether deployed on unmanned aerial vehicles, naval platforms, or ground-based systems, increasingly depend on the same real-time data connectivity and AI navigation principles that commercial drones use—but at orbital scale and with far higher latency requirements and resilience demands.

Military planners designed the Proliferated Warfighter Space Architecture specifically to provide the redundant, low-Earth-orbit data layer that next-generation autonomous weapons and sensors require to operate at machine speed. The orbital mesh network was conceived as the "nervous system" binding these satellites together. A one-year slip in that connectivity layer translates directly into a one-year delay in the operational readiness of any autonomous system that depends on it—including multi-domain command-and-control nodes that fuse sensor data in real time to support autonomous decision-making at tactical and operational levels.

The delay also creates a window of vulnerability in contested orbital scenarios. As adversaries field their own maneuvering satellites and anti-satellite capabilities, any extended gap in U.S. satellite-to-satellite connectivity represents a strategic disadvantage. The Pentagon's doctrine for AI-enabled operations assumes persistent, resilient orbital data infrastructure; the longer that infrastructure remains incomplete, the longer the force posture remains asymmetrical relative to emerging threats in space.

Broader Context and the Path Forward

Sandhoo's acknowledgment of the slip, delivered at a major defence industry conference, signals that the SDA is attempting to reset expectations and communicate realistic timelines to stakeholders. The candour about supply chain constraints and vendor performance also highlights a structural challenge facing the defence space industrial base: the complexity of designing and manufacturing components that must meet both precision and robustness standards simultaneously, while operating within procurement timelines driven by geopolitical urgency.

The revised three-to-four-month timeline for starting construction means the orbital mesh network will not be operational until late 2027 or 2028 at the earliest—substantially later than original planning assumptions. Any further slips would cascade into downstream autonomous systems integration schedules and the operational timelines for multi-domain AI-enabled command-and-control capabilities that depend on this infrastructure.