The expected time for major defense acquisition programs to deliver an initial capability rose by 18 months in the past year, to almost 12 years from a program's start, per GAO's Weapon Systems Annual Assessment of June 11, 2025. Combined estimates for 30 major programs rose $49.3 billion, with the Sentinel missile program accounting for over $36 billion.
What does the delivery record actually show?
GAO's annual assessment is the public scorecard for the Pentagon's costliest programs, and the 2025 edition describes a portfolio moving in the wrong direction on time. The Department of Defense plans to invest nearly $2.4 trillion to develop and acquire its costliest weapon programs, while the expected time for major defense acquisition programs (MDAPs) to provide an initial capability increased to almost 12 years, per GAO-25-107569, published June 11, 2025. Program development delays and inflation, among other factors, contributed to cost growth across the portfolio.
The pattern GAO describes is structural rather than episodic: programs begin with optimistic baselines, encounter technology immaturity, and pay for it later in schedule and cost. Defense weapon systems acquisition has sat on GAO's High Risk List for decades precisely because this cycle repeats across administrations, budgets, and reform initiatives. The Navy's John Lewis-class fleet replenishment oiler program — a support ship, not a first-in-class combatant — reported a $941 million increase since GAO's prior report, evidence that overruns are not confined to exotic technology.
How does a schedule slip, concretely?
The Navy's Constellation-class frigate is the current textbook case of a delivery date eroding in steps. The lead ship was contracted with a delivery date that the Navy now expects to miss by three years: the service plans to deliver the lead frigate in April 2029, three years later than the contracted date, and the ship's design is further behind schedule than previously realized, per RINA's June 16, 2025 report on the GAO assessment.
The documented sequence shows how announced dates diverge from reality one revision at a time:
- August 2023: the program reports functional design 92 percent complete.
- May 2024: after a GAO recommendation, the program restructures its functional design metrics to align with actual design progress.
- December 2024: under the revised metrics, functional design is reported at 70 percent complete — and the ship is 759 tonnes, nearly 13 percent, heavier than anticipated.
- June 2025: lead ship delivery is planned for April 2029, three years past contract, with construction stalled by persistent design instability.
More than two years after lead ship construction began, GAO found that the persistent lack of design stability had stalled construction of the lead ship and posed the same risk to the initial follow-on ships. The shipbuilder filed multiple equitable adjustment requests — the contractual mechanism for claiming reimbursement when the government's requirements change — and the Navy continued working through revised basic design documents, including the ship's general arrangement drawings.
Why do announced dates diverge from delivery dates?
GAO's findings point to a consistent mechanism: programs enter development with low levels of technology maturity. Even programs on the middle tier of acquisition (MTA) pathway — the streamlined process intended specifically for speed — spent development time on efforts with low maturity, resulting in lengthy development instead of the speed for which the pathway was designed, per the 2025 assessment. Of seven former MTA programs GAO examined, none were ready for production or fielding when the effort ended.
The frigate adds a second lesson about measurement. When the program restructured its functional design metrics under GAO's recommendation, measured progress dropped from 92 percent to 70 percent. Nothing in the ship had changed — only the honesty of the ruler. Programs that report progress against loose metrics produce schedules that look healthy until someone tightens the definition, which is why GAO's recommendations so often target how progress is counted before they target the schedule itself.
Can the gap be closed?
GAO's central recommendation is to ensure that new programs include leading product development practices at their earliest stages, which could prevent them from getting locked into rigid requirements and development approaches that cause delays. The emphasis on the earliest stages is the operative point: by the time a program has a public delivery date, its schedule risk is largely already priced in, and the date functions as a planning anchor that only moves through formal rebaselines.
There is also a reading discipline for anyone who follows the budget. A program's announced date is a commitment made at the maturity the program had on announcement day; the delivery date is a fact produced by the maturity the program actually achieved. GAO's annual assessments are the only public documents that track both over time, and the frigate's record — 92 percent to 70 percent on a single metric revision — shows how much daylight can sit between them.
The pattern holds across reform eras because its causes are constant: immature technology at start, unstable design, optimistic baselines, and metrics that flatter. Until those change, the documented rule of thumb stands: treat the announced date as the opening bid, and the GAO record as the schedule history that does not move.
What does the cost record add?
Schedule and cost are the same story told twice. The $49.3 billion increase across 30 major defense acquisition programs in a single reporting year, with Sentinel alone accounting for over $36 billion — roughly 73 percent of the growth — shows how one program's delay can dominate a portfolio's arithmetic. Inflation contributed, GAO found, but development delays were the persistent driver.
The scale of the stake explains the attention. The department plans to invest nearly $2.4 trillion across its costliest programs — a figure large enough that a single year of portfolio-wide cost growth approaches the annual budget of a mid-sized federal agency. The oiler program's $941 million increase, small by comparison, matters as a control case: a mature design, a support mission, and still a nine-figure overrun.
The incentive problem runs in both directions. Programs that report progress honestly absorb near-term criticism for visible slips; programs that report loosely defer the reckoning to a metric revision, as the frigate's 92-to-70 percent correction showed. GAO's recommendation — leading product development practices from a program's earliest stages — targets the front end because that is where the announced date is set, and where the eventual delivery date is quietly determined.
For Congress, the annual assessment is the checklist for authorization and appropriations decisions; for the public, it is the only annually updated, program-by-program record of whether the equipment the budget buys arrives when the announcement said it would. On the current evidence — 12 years to an initial capability and rising — the answer, more often than not, is later.
The frigate's case closes the argument. At contract award, the announced schedule promised a lead ship years sooner than April 2029; the documented record — design metrics restated, weight growth booked, delivery date re-baselined — is the difference between what was promised and what the evidence supported. Any reader of a program announcement can apply the same test: ask for the maturity of the technology, the stability of the design, and the metric by which progress is reported. The answers predict the delivery date better than the press release does, because they are the same factors GAO weighs when it counts the months.
