Lockheed Martin is marketing its Next Generation Interceptor (NGI) as the most capable and cost-effective homeland missile interceptor yet designed, but as of October 14, 2026, the weapon meant to carry the U.S. Ground-Based Midcourse Defense (GMD) system into the 2030s has not completed its Critical Design Review (CDR), has no public cost and schedule baseline, and has never flown. The company now targets fielding by 2030, later than the initial operational capability in the fourth quarter of fiscal 2028 that the Missile Defense Agency (MDA) set when it selected Lockheed Martin on April 15, 2024.
The gap matters because NGI is the ground-based, exo-atmospheric strategic layer of the Golden Dome for America architecture, and the only part of it with silos, radars, and fire control already in place. MDA Director Gen. Heath Collins, who first flagged the 18-month delay to House lawmakers in April 2025, told the Senate Armed Services Committee on April 30, 2026, that the program had slipped 18 months after problems emerged with the solid rocket motor, the inertial measurement unit, and onboard sensors as the effort moved toward manufacturing. The Government Accountability Office (GAO) left NGI out of its 2026 annual weapon systems assessment entirely, citing the lack of long-term cost and schedule baselines to measure progress against. Lockheed Martin’s April 6, 2026, feature on five lesser-known NGI attributes therefore reads as a design pitch rather than a status report.
How the Next Generation Interceptor Replaces the GBI Single Kill Vehicle
The system NGI replaces is the Ground-Based Interceptor (GBI), a three-stage booster topped by a single Raytheon Exoatmospheric Kill Vehicle (EKV). Publicly available information indicates a fleet of 44 GBIs, 40 at Fort Greely, Alaska, and four at Vandenberg Space Force Base, California, built around a design dating to the early 2000s. Each round carries one kill vehicle and can engage one object, which is why GMD planning has long relied on firing several interceptors at a single incoming warhead. MDA’s first attempt to fix that limitation, the Redesigned Kill Vehicle (RKV), was canceled in 2019 after roughly $1.2 billion had been spent, and NGI emerged as a clean-sheet replacement with a new booster and a new payload.
That payload is the core of the program. NGI carries multiple kill vehicles on a single booster, each destroying its target through kinetic impact rather than an explosive warhead, the hit-to-kill principle Lockheed Martin describes as a bullet hitting a bullet. The number of kill vehicles per interceptor has not been disclosed. According to the company, onboard sensors and processing allow the payload to distinguish a real reentry vehicle from decoys autonomously while drawing cues from ground-based radars and command-and-control networks.
The operational consequence is a change in the shot arithmetic. A single NGI that can assign separate kill vehicles to a warhead and its most credible decoys reduces the number of rounds expended per threat, and it shifts part of the discrimination burden from external sensors onto the interceptor itself. The initial plan calls for 20 NGIs to be emplaced in silos already built at Fort Greely. The Pentagon’s Cost Assessment and Program Evaluation (CAPE) office, in its 2021 independent estimate, assumed 21 operational interceptors within a total buy of 31 rounds. With a fixed silo count, capacity gains have to come from inside each missile, and multiple kill vehicles are the mechanism MDA chose to deliver them.
L3Harris Propulsion Stack and the 18-Month Schedule Slip
The propulsion work also shows where the program concentrates risk. Around August 5, 2026, Lockheed Martin completed a burst test of the Stage 2 motor case, pressurizing a water-filled carbon-fiber case beyond expected launch loads until it failed. On August 10, 2026, the Stage 2 solid rocket motor completed a static fire inside a high-vacuum chamber that replicates low Earth orbit conditions, confirming thrust, chamber pressure, and combustion stability under the thermal and pressure stresses of an intercept mission. Christopher Jewell, Lockheed Martin’s NGI vice president, said the results keep the interceptor on track for fielding by 2030.
L3Harris Technologies, which absorbed Aerojet Rocketdyne in 2023, supplies the Stage 1 and Stage 2 motors, the Attitude Control System, and the Divert and Attitude Control System that steers the kill vehicles. One supplier therefore owns effectively the entire propulsion and control stack. That also places NGI inside a solid rocket motor industrial base already absorbing the Pentagon’s push to expand tactical interceptor output, with the fiscal 2027 request seeking roughly $23 billion for missile defense interceptor production.
The two August tests validated one propulsion element, not an integrated interceptor. Collins named the solid rocket motor first among the issues behind the 18-month slip. CDR, originally planned for the third quarter of fiscal 2025, is now expected by the end of 2026. Integration also runs beyond Lockheed Martin: Northrop Grumman, which lost the NGI competition in 2024, holds the GMD Weapon System work that modernizes legacy fire control software and integrates the new interceptor fleet into the broader system.
Digital Twin and In-Silo Upgrades Target GMD Sustainment Costs
Three of the five attributes Lockheed Martin highlighted concern how NGI will be maintained rather than how it will intercept. The interceptor was developed as a born-digital design, with a digital twin mirroring the physical missile in detail. The company says it will deliver an NGI simulator, known as NGISim, for the ground segment to model system performance for years after fielding. Every component is traceable with photographs, measurements, and material data, allowing engineers to run as-built analyses throughout the interceptor’s service life.
The in-silo design follows the same logic. Lockheed Martin says power systems, thermal control, and a durable airframe will keep NGI mission-ready underground for decades, and that engineers can adjust and upgrade components without removing the interceptor from its silo. For a fleet that must stay on alert continuously, that matters: CAPE’s 2021 estimate put the program’s total at about $17.7 billion, including $13.1 billion for development and $2.3 billion for fielding, with the remainder covering operation and sustainment of the interceptors. Upgrades performed in place reduce handling, transport, and the number of rounds off alert at any given time.
The modular architecture is meant to let NGI plug into new sensors and command networks without a redesign, from ground-based discrimination radars to the space-based tracking layers Golden Dome is funding. The limit of the digital approach is equally clear. A model can narrow the test matrix, but MDA leaders have repeatedly said NGI must complete an intercept flight test, and a salvo test, before full production. Simulation fidelity only becomes credible once flight data exists to anchor it, and NGI has none yet.
Analysis: NGI is the least visible and most structurally grounded layer of Golden Dome. Space-based interceptors draw most of the debate, with the Congressional Budget Office (CBO) projecting about $1.2 trillion for an architecture built around them, while Golden Dome program manager Gen. Michael Guetlein has said his office received $22.5 billion from Congress, 95 percent of it already committed to contracts, and warned that the program faces a funding impasse for fiscal 2027. The slip from fiscal 2028 to 2030 means the legacy GBI fleet must hold the homeland mission through the end of the decade while North Korea fields solid-fuel Hwasong-18 and Hwasong-19 intercontinental ballistic missiles. Lockheed Martin’s emphasis on sustainment and digital traceability sells lifecycle value before any intercept has been demonstrated; the decisive evidence will come from flight testing, not the digital twin.
The near-term milestone is CDR closure, expected before December 31, 2026; no completion had been announced as of publication. After that come the production and emplacement decision, the establishment of a program baseline that GAO can track, and a first flight test whose date has not been made public, followed by intercept and salvo testing. The number of kill vehicles per round, the updated unit cost, and the first emplacement date at Fort Greely all remain undisclosed.