What Does GAO’s 2026 Assessment Reveal About NASA’s Major Projects?

Key Takeaways

• NASA’s portfolio appears stable, but the Artemis reset leaves substantial costs under review.
• Orion drives most cumulative overruns, yet its departure will improve future portfolio totals.
• Workforce losses and funding uncertainty threaten NASA’s ability to oversee complex projects.

Why the 2026 NASA Major Projects Assessment Looks Better Than It Feels

NASA’s 36 major projects carry at least $70 billion in estimated life-cycle costs. Eighteen have progressed far enough into development for the U.S. Government Accountability Office (GAO) to measure them against approved cost and schedule baselines. The remaining portfolio includes 17 projects in formulation and the Commercial Crew Program, which operates under tailored management requirements without a comparable project baseline.

GAO released its 2026 assessment of NASA’s major projects on July 23, 2026. The document is the agency’s 18th annual examination of NASA’s largest acquisitions. New Space Economy’s guide to GAO space oversight places the assessment within the longer record of congressional scrutiny covering NASA, national-security space programs, launch infrastructure, and commercial services.

The headline finding appears reassuring. Only two development projects reported new schedule delays, totaling two months, and three reported new cost overruns, totaling $501.4 million. NASA’s cumulative development cost overrun grew from $4.4 billion to almost $4.7 billion. Cumulative delays increased from 13.1 years to 14 years.

Those numbers do not describe a settled portfolio. NASA changed the Artemis III, IV, and V mission sequence in early 2026, paused work on several large systems, revised its surface-transportation plans, and began reconsidering cost and schedule commitments attached to existing contracts. Five of the six Artemis-related projects with measurable baselines were under review when GAO completed its analysis.

A project can show no variance against its baseline because NASA has not completed a replacement estimate. A canceled or paused project can disappear from future portfolio totals without recovering the money already spent. A technically difficult project may remain formally on schedule by consuming reserves that managers had intended to preserve for later problems.

GAO’s figures consequently provide a snapshot of approved baselines, recorded variances, and incomplete replanning. They do not prove that NASA’s largest projects face little risk. Much of that risk now resides in architecture decisions, contract changes, workforce gaps, replacement systems, and mission assumptions that have not yet been converted into stable commitments.

The successful Artemis II mission provides an operational achievement against this unsettled management picture. Artemis II launched on April 1, 2026, carried four astronauts around the Moon, and returned safely on April 10. The flight tested the crewed Orion spacecraft, life-support equipment, navigation, communications, ground operations, recovery procedures, and the Space Launch System rocket.

Operational success and acquisition performance measure different things. Artemis II demonstrated that NASA could fly a crew around the Moon with its existing hardware. GAO examined whether NASA can afford, staff, modify, and repeatedly operate the larger collection of systems needed for lunar landings, surface mobility, science, transportation, and sustained operations.

What the Portfolio Numbers Actually Show

Most development projects reported no new annual cost growth or schedule delay. That finding deserves recognition because spacecraft commonly encounter design changes, test failures, supplier delays, launch-vehicle problems, software defects, or integration conflicts.

NASA completed development of Orion, the Interstellar Mapping and Acceleration Probe, the Low Boom Flight Demonstrator, and the NASA-Indian Space Research Organisation Synthetic Aperture Radar mission during the assessment period.

Two projects reported minor annual schedule changes. The Low Boom Flight Demonstrator added one month before completing its initial flight in October 2025. The flight still occurred more than three and a half years later than its original January 2022 baseline.

The NASA-Indian Space Research Organisation Synthetic Aperture Radar mission added one month and $11 million before launching in July 2025. That launch occurred more than two and a half years after its original September 2022 target. A launch failure involving another Indian mission contributed to the delay because the Indian Space Research Organisation needed to confirm that the launch vehicle assigned to the radar mission was not exposed to the same problem.

The Interstellar Mapping and Acceleration Probe offered a favorable counterexample. It launched in September 2025, three months ahead of its committed readiness date, and transferred $22.2 million in unused development reserves into operations. Part of that funding covered inflation and rate adjustments, with the remainder strengthening operational reserves.

Orion dominates the less favorable side of the ledger. Its annual overrun reached $261 million, connected to the seven-month movement of Artemis II from September 2025 to April 2026. Orion accounted for about 52% of the portfolio’s annual cost growth and almost 75% of cumulative development overruns.

Orion’s original development baseline was approximately $6.8 billion. Cumulative development cost growth reached about $3.5 billion. GAO’s calculation uses the original baseline rather than the rebaselined commitment NASA established in 2021 after scope changes, pandemic effects, program growth, and delays involving the European Service Module.

Space Launch System Block 1B produced another $229.4 million in projected annual cost growth before NASA discontinued the upgrade. Delayed deliveries affected the Exploration Upper Stage, Universal Stage Adapter, payload adapter, and related hardware. Managers used schedule reserves to remain within the approved completion date, but only one month of reserve remained.

Orion and Block 1B accounted for about 98% of the annual cost overruns identified by GAO. The rest of the development portfolio contributed little to the annual increase. That concentration means broad portfolio statistics can make the performance of a few expensive programs appear representative of the entire agency.

Orion, the Low Boom Flight Demonstrator, the radar mission, and Block 1B are expected to leave GAO’s development analysis. Their removal could eliminate about 9.5 years of recorded delay and more than $4 billion in accumulated cost overruns from the portfolio calculation. Orion alone will remove approximately $3.5 billion, or 73% of the present cumulative overrun.

Future totals may consequently improve even if no continuing project becomes cheaper or faster. GAO cautions that changes in portfolio membership must be separated from improvements in project execution.

An acquisition can leave the calculation because it succeeded, was canceled, changed scope, or reached the end of development after exceeding its baseline. A falling aggregate overrun does not recover earlier spending, remove termination costs, or pay for replacement capabilities.

Congress and NASA need project-level explanations alongside portfolio totals. Without that detail, a cleaner chart can conceal abandoned hardware, contract closeout expenses, delayed capabilities, and successor systems that have yet to receive formal baselines.

How the Artemis Reset Rewrites Existing Contracts and Baselines

NASA’s revised Artemis III mission is planned as a low-Earth-orbit demonstration in 2027. Orion conducts rendezvous and docking activities with one or both commercial Human Landing System providers, SpaceX and Blue Origin. The mission may also test lander interfaces, life-support connections, communications, propulsion systems, and spacesuit-related equipment.

The revised Artemis III orbital test keeps the crew closer to Earth than a lunar mission would. It also moves the need for representative lander hardware and interfaces earlier. SpaceX and Blue Origin must provide systems for the 2027 demonstration before completing every capability needed for lunar descent and ascent.

NASA plans for Artemis IV to attempt the next crewed lunar landing in early 2028. Artemis V is planned for later in 2028 and is also expected to include a surface mission. NASA has described a goal of conducting at least one lunar landing per year after the revised sequence begins.

The NASA Office of Inspector General’s Human Landing System audit identifies the contractual and technical pressures facing the lander program. Both providers have encountered delays, development problems, integration demands, and unresolved crew-safety matters. NASA is working with both companies to accelerate lander development toward a 2028 surface mission.

New Space Economy’s examination of Human Landing System contract management explains why a firm-fixed-price contract does not make government-directed scope changes free. NASA’s approved baseline for the SpaceX initial lander capability covered a lunar landing mission. It did not include every activity now planned for the 2027 demonstration or the acceleration work NASA requested from both providers.

Government-directed changes may lead to higher costs, modified contract terms, new acceptance requirements, additional tests, and schedule movement. The commercial provider retains responsibility for delivering the contracted system, but NASA remains responsible for defining requirements and integrating the lander with Orion, ground operations, spacesuits, mission control, and crew procedures.

Gateway faces a more fundamental change. NASA paused the small lunar-orbit station in its existing form and shifted attention toward surface infrastructure. The agency is studying whether hardware developed for Gateway can support other missions.

The Gateway Power and Propulsion Element could contribute to a nuclear electric propulsion demonstration. Components from the Habitation and Logistics Outpost may be considered for surface modules or another exploration application.

Repurposing has practical limits. Hardware designed for lunar orbit may need new structures, thermal controls, radiation protection, software, interfaces, qualification tests, transportation arrangements, and operations concepts before it can function on the surface or travel to another destination.

The Office of Inspector General’s review of canceled or repurposed Artemis systems examined the Exploration Upper Stage, Universal Stage Adapter, Mobile Launcher 2, and Gateway habitation hardware. New Space Economy’s analysis of these Artemis canceled systems describes the resulting asset-management problem.

A technically possible reuse proposal is not automatically an economical mission. NASA must identify the destination, requirements, integration authority, transportation method, additional development cost, operating concept, and scientific or exploration purpose before preserved hardware can be treated as a useful asset.

Space Launch System Block 1B and Mobile Launcher 2 present the same issue at a larger scale. NASA no longer plans to introduce the Exploration Upper Stage for the next Artemis missions. It intends to standardize the rocket configuration and pursue a modified United Launch Alliance Centaur V upper stage for later flights.

The revised approach requires changes involving the rocket, Orion, launch tower, fueling connections, software, ground systems, and certification work. United Launch Alliance retired the production line for the Interim Cryogenic Propulsion Stage used by earlier Artemis missions, leaving NASA with one remaining unit for Artemis IV.

Mobile Launcher 2 was designed for Block 1B. NASA paused the launcher project and began evaluating its partially completed hardware, software, workforce, and facilities. Exploration Ground Systems must also determine how to modify existing launch infrastructure for the Centaur V approach.

The Orion program faces compressed production requirements. NASA now needs an Artemis V capsule approximately two years earlier than planned under the prior sequence. The European Space Agency must support faster delivery of a service module, and Orion must prepare for direct docking with commercial landers without relying on Gateway as an intermediate destination.

The Artemis reset may remove systems that were late, expensive, or poorly aligned with the revised surface strategy. It also creates new integration demands among systems developed under different contracts, assumptions, schedules, and technical authorities.

The financial result will depend on how quickly NASA defines the replacement architecture, updates contracts, assigns responsibilities, and establishes auditable commitments for each revised capability.

Why Workforce Losses Have Become a Project Risk

NASA lost more than 4,000 civil servants between March 2025 and January 2026, reducing its civil-service workforce by approximately 22%. Twenty-five of the 36 major projects told GAO that staffing reductions had affected their work.

Remaining employees often absorbed duties previously held by departing personnel. That response preserved near-term activity, but it also increased workloads, reduced backup capacity, and made project offices more dependent on a smaller number of experienced employees.

The reductions reached engineering, procurement, project management, science, information technology, cybersecurity, manufacturing oversight, safety, and mission operations. The loss of one experienced engineer can affect several projects when that person reviews shared hardware, approves waivers, supports independent assessments, or holds knowledge that has not been fully documented.

The Space Launch System program reported losing almost 20% of its civil-service workforce. Orion lost about 10% and faced difficulty filling several positions. Near-Earth Object Surveyor experienced disruption through workforce changes at the Jet Propulsion Laboratory, including the departure of employees in leadership and hardware-development positions.

The Deep Atmosphere Venus Investigation of Noble Gases, Chemistry, and Imaging mission reported losses among engineering and program-management personnel. Some vacancies could not be filled during the 2025 hiring restrictions.

Such departures are harder to correct than a simple head-count figure suggests. Experienced personnel understand design history, contractor performance, test anomalies, earlier trade studies, accepted risks, and the reasons behind technical requirements.

Supply-chain problems compound staffing pressure. Projects such as the Multi-slit Solar Explorer reported late custom mechanical parts, avionics delays, communications-system problems, and supplier-capacity conflicts.

New Space Economy’s discussion of single-source suppliers explains how a small group of specialized vendors can affect several missions at the same time. Fewer NASA employees leave less capacity for factory visits, technical reviews, alternate-source qualification, schedule recovery, and contractor surveillance.

NASA changed direction after receiving its fiscal year 2026 appropriation. The agency announced plans to rebuild selected technical skills, improve onboarding, strengthen training, expand mentorship, and reconsider which functions should remain inside government.

The NASA Force hiring initiative opened applications in April 2026. Developed with the U.S. Office of Personnel Management, the initiative offers one-year to two-year appointments for engineers, technologists, and other specialists assigned to mission-focused work.

NASA Administrator Jared Isaacman’s workforce message and the agency’s May 2026 realignment described center specialization, workforce rebalancing, science-process changes, lunar-surface integration, space nuclear power, low-Earth-orbit transition, and aviation-program management.

New Space Economy’s review of the NASA workforce realignment frames the policy as an attempt to retain direct government control over selected technical functions while buying commercial services where working markets exist.

That distinction can support better management, but execution requires stable funding, clearly assigned government responsibilities, timely hiring, competitive compensation, security processing, and managers with authority to fill identified gaps.

Budget uncertainty may obstruct that rebuilding effort. NASA’s fiscal year 2027 budget request seeks $18.8 billion, compared with an enacted fiscal year 2026 level of about $24.4 billion. The request proposes substantial reductions in science funding alongside continued investment in Artemis and lunar infrastructure.

Workforce risk seldom appears immediately as a launch delay. Teams may defer documentation, combine review duties, postpone lower-priority analysis, extend working hours, or rely more heavily on contractors. The visible consequence can arrive months later as a missed defect, weak estimate, incomplete test, delayed approval, or loss of competition within a specialized supply chain.

Which Science and Infrastructure Projects Face the Most Pressure

NASA’s project portfolio extends far beyond Artemis. It includes astrophysics observatories, Earth-science satellites, planetary missions, solar instruments, aeronautics demonstrations, space-station transportation, and the vehicle intended to deorbit the International Space Station.

The relatively stable annual cost and schedule totals conceal different conditions within these programs.

The Nancy Grace Roman Space Telescope is one of the strongest-performing large projects. GAO’s assessment recorded a formal launch-readiness commitment of May 2027 and an earlier project target in 2026. NASA subsequently scheduled Roman to launch no earlier than August 30, 2026, aboard a SpaceX Falcon Heavy rocket.

Roman reached Kennedy Space Center in June 2026 and entered final launch processing. The observatory is scheduled to launch about eight months ahead of its formal commitment, assuming final spacecraft work, fueling, integration, range availability, and launch conditions proceed as planned.

Roman’s cumulative development estimate remained above its original baseline in GAO’s calculation. NASA had submitted a lower estimate in January 2026, but the agency had not finalized it when GAO completed the portfolio analysis.

The Dragonfly mission to Saturn’s moon Titan remained within its approved baseline and entered rotorcraft integration and testing in 2026. NASA plans to launch Dragonfly in July 2028 aboard a Falcon Heavy rocket, followed by a journey lasting about six years.

The mission depends on a Multi-Mission Radioisotope Thermoelectric Generator, specialized scientific instruments, planetary-protection controls, and equipment that must survive launch, interplanetary travel, atmospheric entry, landing, and repeated flights in Titan’s environment.

The Near-Earth Object Surveyor remained within its cost and schedule commitment. The infrared observatory is scheduled to launch no earlier than September 2027 and conduct a five-year survey for asteroids and comets that may pose an impact hazard to Earth.

Workforce changes at the Jet Propulsion Laboratory affected leadership and hardware-development teams associated with the project. The mission remained on schedule as of July 23, 2026, but the staffing losses reduced organizational margin during assembly, integration, testing, and software development.

The Compton Spectrometer and Imager was operating within its approved estimates and remained scheduled for launch in 2027. The gamma-ray telescope will investigate antimatter, the formation of chemical elements, gamma-ray polarization, and energetic events associated with gravitational-wave sources.

The project encountered clearance and cooling problems involving its cryostat. Testing indicated that the flight system could achieve acceptable detector temperatures, and the project retained reserves for the remaining work.

The Multi-slit Solar Explorer remained planned for launch in 2027. Supplier delays affected instruments, avionics, communications equipment, spacecraft systems, and custom mechanical parts. Project managers considered extended work periods to recover time, a step that can increase labor costs and employee fatigue.

The United States Deorbit Vehicle entered GAO’s portfolio with a development baseline of about $1.04 billion. SpaceX is developing the spacecraft to guide the International Space Station through a controlled disposal after the station’s operating life ends.

NASA will take ownership of the vehicle after development and operate it during the deorbit mission. Its performance will become increasingly important as NASA balances station safety, commercial destination development, research continuity, crew transportation, and the risk of a gap in American crew-capable orbital facilities.

New Space Economy’s article on the United States Deorbit Vehicle examines the unusual propulsion, reliability, docking, and control demands created by the need to move the station safely out of orbit.

Mars Sample Return illustrates how scientific value can collide with affordability. NASA began closing out elements of the earlier architecture after studying government and commercial approaches intended to lower costs and return samples during the 2030s.

NASA has continued evaluating possible landing, ascent, containment, and return methods. The agency previously indicated that it expected to confirm a revised design in the second half of 2026. Until that decision is made, Mars Sample Return remains a scientific objective without a fully approved replacement architecture.

The global mission schedule for 2026 and 2027 shows that NASA’s projects operate within an international system. European, Japanese, Chinese, Indian, Canadian, and commercial missions compete for launch capacity, instruments, ground networks, specialized components, and scientific personnel.

Delaying a NASA mission can affect international agreements, shared instruments, launch opportunities, data plans, and joint research campaigns. Budget proposals can also influence projects before Congress makes a final funding decision. Teams may slow procurement, avoid hiring, prepare cancellation plans, or postpone reviews when a mission’s future appears uncertain.

What GAO’s Findings Say About NASA Acquisition Management

GAO’s central management concern is not that every NASA project is failing. The concern is that NASA repeatedly begins or modifies expensive campaigns without complete life-cycle estimates, integrated schedules, or enforceable controls covering every capability required to accomplish the mission.

The Artemis restructuring increases the need for those controls because a decision affecting one project can alter several others.

GAO has asked NASA to develop a transparent life-cycle cost estimate covering the initial Artemis lunar landing since December 2019. NASA had not completed that estimate as of May 2026.

Individual project baselines exist for several Artemis components, but those estimates do not automatically reveal the complete cost of combining rockets, spacecraft, landers, spacesuits, launch infrastructure, logistics, mission operations, integration work, and surface equipment into a completed lunar landing.

A campaign estimate does more than produce a single total. It shows when funding will be required, which costs depend on mission frequency, where reserves are held, how replacement systems affect prior investments, and whether annual appropriations can support the planned sequence.

Without that view, NASA can approve each component separately even when the combined campaign exceeds a realistic funding profile.

GAO has also pressed NASA to establish separate cost and schedule baselines for capability upgrades. NASA revised its project-management policy and created distinct commitments for systems such as Space Launch System Block 1B and Mobile Launcher 2.

Those baselines exposed cost growth that might otherwise have remained inside larger program accounts. Canceling a baselined project does not make the baseline unnecessary. It permits Congress to compare what NASA approved, what contractors delivered, how much was spent, and what useful hardware remains.

Schedule controls matter at the campaign level as well. NASA wants to move from the interval between Artemis I and Artemis II toward a much faster launch schedule after Artemis III.

That increase requires synchronized delivery of Space Launch System hardware, Orion capsules, European Service Modules, commercial landers, spacesuits, upper stages, launch infrastructure, software, crews, recovery forces, suppliers, and mission-control capacity.

The revised Artemis plan removes some earlier dependencies, but it leaves substantial lander-development pressure in place. It also introduces new work through the 2027 demonstration, the replacement upper stage, Gateway changes, Orion production acceleration, and international-partner adjustments.

NASA’s confidence in a 2027 Artemis III flight depends on successful progress by SpaceX, Blue Origin, United Launch Alliance, Orion contractors, spacesuit providers, Exploration Ground Systems, and NASA’s own integration teams.

GAO’s preferred management discipline does not require NASA to avoid changing plans. Space programs must respond to test results, contractor performance, budgets, policy decisions, safety findings, and new technologies.

The discipline requires NASA to define each change, estimate its consequences, assign decision authority, preserve reserves, update contracts, and provide Congress with a traceable account of what changed.

A faster Artemis cadence may justify simplifying the rocket configuration and removing Gateway from near-term landing missions. It does not make replacement capabilities cost-free or schedule-neutral. Simplification in one project can transfer cost, work, or technical risk to another contractor, center, partner, or system.

What Congress, NASA, and Contractors Must Decide Next

NASA must turn broad Artemis announcements into project definitions, contract modifications, technical requirements, budgets, and integrated schedules before uncertainty consumes more time and funding.

Human Landing System scope needs early resolution. NASA must decide what SpaceX and Blue Origin provides for Artemis III, which provider will support Artemis IV, how acceleration work will be funded, and which demonstrations must occur before crewed lunar operations.

Those decisions affect lander development, Orion docking equipment, spacesuit interfaces, launch campaigns, simulations, training, life-support connections, surface procedures, and mission-control planning.

NASA also needs a documented future for Gateway hardware. Repurposing should proceed only when the destination, requirements, transportation plan, integration authority, and estimated cost are credible.

Preserving hardware without an approved use can produce continuing storage, engineering, maintenance, and contract expenses without advancing an operational mission.

The Centaur V proposal requires similar management discipline. NASA must define the modified stage, human-spaceflight certification work, structural interfaces, ground changes, software, procurement terms, fueling procedures, and schedule relationship with Artemis IV and V.

Commercial heritage can reduce development work, but integration into a crewed lunar transportation system remains a government responsibility.

Congress must determine whether fiscal year 2027 funding supports the agency’s stated program. The administration’s $18.8 billion request is about $5.6 billion below the fiscal year 2026 appropriation and directs a greater share of the budget toward Artemis and lunar infrastructure.

New Space Economy’s NASA budget analysis explains why top-line funding alone cannot settle the issue. An appropriation can preserve NASA’s total budget but still leave individual missions underfunded, restrict cancellations, or require the agency to continue projects that no longer fit its preferred architecture.

Contractors face decisions of their own. SpaceX must demonstrate Starship reliability, long-duration propellant management, orbital transfer, repeated launch operations, and lunar-lander performance.

Blue Origin must advance Blue Moon hardware, New Glenn availability, fueling operations, flight testing, and crew-system integration. Boeing, Lockheed Martin, Northrop Grumman, United Launch Alliance, Axiom Space, Bechtel, and specialized suppliers must respond to changed quantities, interfaces, delivery dates, and contract scope.

NASA’s broader civil space technology shortfalls connect the project assessment to surface power, communications, autonomy, construction, thermal control, resource use, mobility, propulsion, life support, ground facilities, and supplier capacity.

A sustained lunar base cannot be created by rescheduling transportation projects alone. Long-term operations require infrastructure that survives extended exposure, works with more than one provider, supports maintenance, and can operate without constant replacement from Earth.

GAO’s assessment gives NASA qualified credit for limited annual overruns. It also shows that the agency is entering a period in which traditional portfolio metrics provides incomplete guidance.

Rebaselines, canceled projects, transferred hardware, new contracts, workforce rebuilding, and replacement architectures will change what the portfolio contains.

A future assessment may report lower cumulative overruns because Orion and other delayed projects have departed. The more meaningful test will be whether NASA has established credible commitments for the revised Artemis campaign, retained enough government expertise to oversee contractors, protected viable science missions, and explained the full cost of moving from demonstrations to repeatable lunar operations.

Summary

GAO’s 2026 assessment presents two valid pictures of NASA. One shows a development portfolio in which most projects recorded no new annual delay or cost growth. The other shows an agency rewriting its most expensive exploration campaign after reducing its civil-service workforce by about 22%.

Orion and Space Launch System Block 1B dominate the present overrun figures. Their removal will make future portfolio totals appear healthier, but it will not recover prior expenditures or settle the cost of replacement systems.

The Artemis reset transfers attention from Gateway, Block 1B, and Mobile Launcher 2 toward standardized launch configurations, commercial landers, direct docking, lunar-surface infrastructure, and a faster mission sequence.

NASA now needs revised project baselines, an integrated Artemis cost estimate, realistic schedules, defined contractor responsibilities, and enough government expertise to evaluate technical and financial claims independently.

Artemis II proved that NASA and its partners could send a crew around the Moon and return them safely. The next stage asks whether that achievement can become repeatable, affordable, and compatible with NASA’s science, aeronautics, low-Earth-orbit, planetary-defense, and infrastructure obligations.

GAO’s findings leave that question unresolved as of July 23, 2026.

Appendix: Useful Books Available on Amazon

• Escaping Gravity
• The Space Barons
• The Moon: A History for the Future
• Rise of the Rocket Girls
• Apollo: The Race to the Moon

Appendix: Top Questions Answered in This Article

How Many Major NASA Projects Did GAO Examine in 2026?

GAO examined a portfolio of 36 projects with estimated life-cycle costs above $250 million. Eighteen were in development with approved cost and schedule baselines, 17 remained in formulation, and the Commercial Crew Program operated under tailored requirements without a comparable baseline.

How Much Is NASA’s Major-Project Portfolio Expected to Cost?

NASA estimates that the 36 projects will cost at least $70 billion over their life cycles. That figure includes development, operations, and other approved phases. Artemis restructuring, contract changes, cancellations, and replacement systems may change the eventual total.

Did Most NASA Projects Experience New Overruns in 2026?

No. Most development projects reported no new annual cost growth or schedule movement. Three projects reported new cost overruns totaling $501.4 million, and two reported schedule delays totaling two months.

Which Project Accounts for Most of NASA’s Cumulative Overruns?

Orion accounts for almost 75% of cumulative development cost overruns in the present portfolio. Its development cost grew by approximately $3.5 billion against its original baseline, and its schedule moved by about three years.

Why Might NASA’s Portfolio Statistics Improve in a Future Assessment?

Orion and several delayed projects are leaving the development portfolio after completing missions or development work. Space Launch System Block 1B is also being discontinued. Removing these projects will lower aggregate delays and overruns without necessarily showing better performance among continuing projects.

What Changed About Artemis III?

Artemis III is planned as a 2027 low-Earth-orbit demonstration involving Orion and one or both commercial lunar landers. The mission is intended to test rendezvous, docking, interfaces, life support, communications, and related systems before a lunar landing attempt under Artemis IV.

What Happened to the Gateway Lunar Station?

NASA paused Gateway in its existing form and shifted attention toward lunar-surface infrastructure. The agency is evaluating whether completed or partially completed Gateway hardware can support surface modules, a propulsion demonstration, or another mission.

How Have NASA Workforce Reductions Affected Projects?

NASA lost more than 4,000 civil servants, approximately 22% of its government workforce. Twenty-five major projects reported effects that included vacancies, reduced engineering capacity, increased workloads, loss of institutional knowledge, and difficulty replacing specialized personnel.

Why Does GAO Want a Complete Artemis Cost Estimate?

Separate project estimates do not show the complete expense of conducting a lunar landing campaign. A campaign-level estimate would connect rockets, Orion, landers, spacesuits, ground systems, operations, logistics, reserves, infrastructure, and integration costs to an executable funding plan.

Does Artemis II’s Success Remove the Risks Identified by GAO?

No. Artemis II demonstrated the Space Launch System rocket, Orion spacecraft, ground systems, life support, navigation, and mission operations during crewed lunar flight. Later missions require new landers, docking arrangements, spacesuits, upper-stage work, faster production, surface systems, and revised contracts that Artemis II did not fully test.

Appendix: Glossary of Key Terms

Agency Baseline Commitment

NASA’s approved cost and schedule commitment for a project. It includes project-level or agency-level reserves and provides the standard against which GAO measures cost growth and schedule delay.

Artemis

NASA’s campaign to conduct crewed lunar missions, establish sustained activity near and on the Moon, develop commercial and international partnerships, and prepare technologies and operating experience for later human missions to Mars.

Cost Reserve

Funding held for risks or expected work that has not yet been assigned to a specific project activity. NASA refers to these funds as unallocated future expenses.

Development

The portion of project implementation covering detailed design, fabrication, assembly, integration, testing, and launch preparation. GAO measures development projects against approved cost and schedule baselines.

Formulation

The period during which a NASA project studies concepts, develops requirements, matures technology, creates preliminary designs, and prepares estimates before receiving approval to begin full development.

Human Landing System

The commercial lunar-lander capability intended to transport astronauts between lunar orbit and the Moon’s surface. SpaceX is developing a Starship-derived system, and Blue Origin is developing the Blue Moon architecture.

Life-Cycle Cost

The estimated cost of a project across formulation, development, launch, operations, and closeout. The phases included depend on the project’s management plan and operating structure.

Rebaseline

A formal change to a project’s approved cost or schedule commitment. NASA may rebaseline when expected cost growth exceeds defined limits or when changed circumstances require a new commitment.

Schedule Reserve

Additional time included within a project schedule to absorb delays, test results, supplier problems, technical defects, or other uncertainty without moving the agency’s committed completion date.

Space Launch System

NASA’s government-developed heavy-lift launch vehicle for crewed Artemis missions. The Block 1 configuration launched Artemis I and Artemis II, and NASA plans to use a standardized configuration for upcoming flights.