DOD's hypersonic missiles are late, and the test margin is gone
Summary
The Pentagon is racing to field offensive hypersonic weapons -- missiles that fly at least five times the speed of sound on unpredictable flight paths -- across the Army, Navy, and Air Force. In July 2024, GAO reviewed all six of DOD's offensive hypersonic efforts against leading practices for product development and found most were skipping the ones meant to catch problems early: four of six weren't soliciting direct feedback from actual users, and four of six weren't using digital engineering tools like digital twins. GAO made 10 recommendations; DOD concurred with all of them. About two years later, GAO's 2026 annual weapons assessment reports on what happened next: the Army's LRHW program is fielding its first missile battery more than two years behind its original goal, the Navy's CPS program had its testing phase extended a full year, and the Air Force's HACM program has already cut its own flight-test plan and says its schedule has effectively zero margin left.
Two practices meant to catch problems early -- and most programs skipped them
The first practice is soliciting direct feedback from the people who will actually operate the weapon -- not generic stakeholder input, but feedback from users who physically interact with the system. found only two of the six efforts, the Army's Long-Range Hypersonic Weapon (LRHW) and the Navy's Conventional Prompt Strike (CPS), were doing this. The other four -- the Navy's Standard Missile-6 Block IB and Hypersonic Air-Launched Offensive anti-surface warfare (HALO) efforts, and the Air Force's Hypersonic Attack Cruise Missile (HACM) and Air-launched Rapid Response Weapon (ARRW) -- were not, and had no plans to.
View data as table
| Soliciting user feedback (LRHW, CPS) | 2 |
|---|---|
| Not soliciting user feedback (SM-6 Block IB, HACM, HALO, ARRW) | 4 |
The second practice is using digital engineering tools -- virtual models, including 'digital twins,' that simulate a system's performance before it's built, so design flaws surface in software instead of on a test range. found only two efforts, HALO and CPS, were using or planning to use these tools. The other four -- LRHW, SM-6 Block IB, ARRW, and HACM -- were not, or were using them only in a limited way with no plans to expand. made 10 recommendations addressing these gaps, plus 's cost estimating and enterprise-level risk reporting. concurred with all 10.
View data as table
| Using or planning digital engineering (HALO, CPS) | 2 |
|---|---|
| Not using digital engineering (LRHW, SM-6 Block IB, ARRW, HACM) | 4 |
Two years later: the Army's missile battery, over two years late
's 2026 assessment reports that the Army will not field its first LRHW battery -- including missiles -- until at least the second quarter of fiscal year 2026: more than two years later than the program's initial goal, and nine months later than 's own prior assessment. The Army did resolve the launcher and missile quality issues driving earlier delays, completing a successful end-to-end flight test in the first quarter of fiscal year 2025. But the program's second battery is also behind, delayed to the second quarter of fiscal year 2027 by missile delivery delays of at least six months. On the digital engineering recommendation specifically, the Army did complete the assessment asked for -- and concluded in December 2024 that digital twinning wasn't worth building for a program that only plans to produce three batteries.
The Navy's missile: a year behind on testing, production still bottlenecked
CPS's rapid-prototyping phase was extended a full year, from October 2024 to October 2025, because of testing issues, before the program hit its main milestone -- a successful end-to-end flight test from a Navy cold-gas launcher -- in April 2025. The program's own officials point to missile availability as a continuing risk: earlier flight-test failures forced a hold on missile production until the weapon was proven, and the resulting stop-and-start production schedule has made it hard for the contractor to keep the same staff on the line between runs, limiting the kind of learning that speeds up manufacturing.
The Air Force's missile: fewer tests planned, and no room for a bad one
HACM is still on track for its first flight test in the second quarter of fiscal year 2026, but the program has already cut its own rapid-prototyping flight-test plan from seven tests to five -- and officials told there is effectively zero margin left in the schedule. If a single significant test fails, the program likely won't be able to complete even the reduced five-test plan within its five-year window. Officials said completing at least the first three tests is critical to the Air Force's decision on whether to move HACM into full production in fiscal year 2027. One piece of good news: unlike in 2024, HACM has since begun incorporating user feedback into its design.
View data as table
| Originally planned flight tests | 7 |
|---|---|
| Flight tests in the descoped plan (as of 2026) | 5 |
The takeaway
- The practices flagged as missing are exactly the ones meant to prevent this. Two-thirds of 's hypersonic efforts skipped direct user feedback, and two-thirds skipped digital engineering tools -- both meant to surface design and production problems before they hit a test range or a fielding date.
- All three programs checked back on are running into the kind of trouble those practices are supposed to catch. The Army's missile battery is over two years late, the Navy's testing phase ran a year over, and the Air Force's program has already given up two of its planned flight tests.
- said yes to every fix in 2024, and the follow-through has been uneven. The Army did complete the digital-engineering assessment asked for -- and concluded against building the tool. The Air Force's HACM program has since added user feedback to its process. But none of that has kept schedules from slipping.
Findings on 's six hypersonic efforts, their 2024 leading-practices gaps, and 's 10 recommendations are from -24-106792, 'Hypersonic Weapons: Could Reduce Cost and Schedule Risks by Following Leading Practices' (July 29, 2024), read directly in full via an archived copy after the current gao.gov asset URL blocked direct access. Findings on the LRHW, CPS, and HACM programs' 2026 status are from -26-108457, 'Weapon Systems Annual Assessment: Requiring Mature Technologies Could Enable Shift to Rapid Delivery' (July 2, 2026), also read directly in full via an archived copy; that report also contains a much broader -wide portfolio finding covered separately by another BlackLeaf article, and only its hypersonic program-level assessments are used here. The two reports examine the same six-program hypersonic effort about two years apart -- the first flagging skipped practices as an avoidable risk, the second documenting the schedule and production problems that followed for three of those programs.
Sources(2) ▾
- U.S. Government Accountability Office, Hypersonic Weapons: DOD Could Reduce Cost and Schedule Risks by Following Leading Practices (2024-07-29) — -24-106792, a report to congressional committees assessing 's six offensive hypersonic weapon efforts against leading practices for product development, cost estimating, and risk management -- the warning facet. Distinct from doc-gao26-108457 (the confirmed-outcome facet, published about 2 years later). Direct gao.gov PDF fetch returns HTTP 403; fetched in full via an archived copy. gao.gov · original document
- U.S. Government Accountability Office, Weapon Systems Annual Assessment: Requiring Mature Technologies Could Enable Shift to Rapid Delivery (2026-07-02) — -26-108457, 's annual omnibus assessment of weapon programs, containing individual program-level assessments (as of January 2026) for the Army's LRHW, the Navy's CPS, and the Air Force's HACM hypersonic programs -- the confirmed-outcome facet, published about 2 years after doc-gao24-106792's leading-practices warning. This report also contains a much broader, -wide portfolio finding (aggregate ~$2.4 trillion cost growth across 104 programs) that is outside the scope of this piece and already covered by a separate BlackLeaf article; only the report's LRHW/CPS/HACM program-level assessments are used here. Direct gao.gov PDF fetch returns HTTP 403; fetched in full via an archived copy. gao.gov · original document
Comments
Always open. Logged-in readers can annotate paragraphs in place.
Hypersonic weapons fly at least five times the speed of sound on flight paths that are hard to predict or intercept, and has six offensive hypersonic efforts underway across the Army, Navy, and Air Force. In July 2024⧉, assessed all six against the practices that leading product-development organizations use to catch problems before they become expensive. In July 2026⧉ -- about two years later -- 's annual weapons assessment reported on how three of those programs were actually doing.