GrayMatter Robotics uses its Factory SuperIntelligence AI architecture across industries, environments, materials, geometries, and applications. | Source: GrayMatter Robotics
Workforce shortfall and attrition in defense manufacturing are measurable, and their effects are showing up in readiness data. GrayMatter Robotics said that its autonomous surface-finishing systems represent one structural response to the trades shortage driving that attrition.
According to the Government Accountability Office’s (GAO) March 2025 military readiness report, the U.S. military missed its aircraft readiness goals on 42 of 45 fleets in 2024, largely due to a shortage of trained maintenance workers. Surface preparation and finishing work that precedes depot-level repair sits on the critical path of those workflows.
With the U.S. Navy’s 2024 industrial base review identifying a 174,000-worker shortfall, the readiness shortage is an industrial-capacity problem.
“Depot facilities have requirements that most automation platforms weren’t designed around: no external data routing, no reprogram cycles between parts, and full traceability on every surface the system touches,” said Ariyan Kabir, co-founder and CEO of GrayMatter Robotics. “Our edge-deployed physical AI architecture was built around those constraints from Day 1.”
Aging depot workforce creates surface preparation bottleneck
Depot-level maintenance is specialized work in the defense industrial base, said GrayMatter Robotics. A technician overhauling fighter aircraft landing gear or preparing naval vessel surfaces for protective coatings has spent considerable time acquiring that expertise.
For many technicians in the field, apprenticeship began early in life, and those same workers may now be reaching retirement age across major defense depots. Because it takes four to six months to train new hires before they acquire proficiency, the current personnel pipelines cannot replace them at the rate they are leaving, said the Carson, Calif.-based company.
Surface preparation is the hidden constraint within this workflow. Before components get new systems installed or before corrosion-resistant coatings go on, the surfaces must be prepared to specification. For aircraft that spent 20 years in operational environments, that means addressing corrosion and irregularities unique to each platform’s service history, GrayMatter noted.
Shipbuilding’s labor shortage runs deeper than recruiting, says GrayMatter
A separate GAO report on shipbuilding and repair found that the U.S. Navy‘s own 45-day review projects a need for 174,000 new workers over the next decade. According to a senior Navy civilian official, 50% to 60% of first-year shipbuilding workers leave before completing their first year on the job. At that rate, hiring programs struggle to make up the difference between workforce demand and available labor.
Physical AI finishing systems can add production capacity without extending the training timeline, asserted GrayMatter Robotics.
Key facts:
- GrayMatter Robotics and HII (Huntington Ingalls Industries) agreed in April 2026 to integrate physical AI into manned and unmanned shipbuilding programs
- The HYPR (High-Yield Production Robotics) program, a joint initiative with HII and Path Robotics, was established in April 2026 to build autonomous assembly lines for ship and submarine construction
Why traditional automation couldn’t touch depot work
Unlike factory finishing, where parts arrive in standard configurations, depot work offers significant geometric variation. A corroded landing gear strut looks different every time, and hull preparation for a 40-year-old destroyer presents unique surface conditions on every visit, noted GrayMatter Robotics.
Traditional robotic systems required preprogrammed paths, an approach that works for commodity production but breaks down when no two jobs are alike.
According to the CIRP Annals research paper, finishing complex surfaces has historically depended on manual labor from skilled workers. The central challenge of automating it is maintaining consistent material removal across variable geometry and surface conditions, it said.
“Every part coming through a depot has its own surface history that includes corrosion patterns, coating buildup, and prior repair work,” said Kabir. “The geometry changes with each unit, and so does the finishing challenge. Systems trained on millions of real surface interactions handle that variability as a matter of course. That accumulated process knowledge is what makes geometry-agnostic finishing practical at depot scale.”
Active DoW and Navy programs signal a procurement shift
The AFWERX SBIR Phase II program selected GrayMatter Robotics to develop autonomous systems for defense manufacturing. The Navy’s depot maintenance efficiency challenge named 12 finalists from a competitive applicant pool, among them GrayMatter Robotics, HII, and Path Robotics, the same organizations that established the HYPR program.
These selections represent active U.S. Department of War procurement responses to throughput losses that are already showing up in readiness data.
Readiness is a deployment speed problem, GrayMatter asserts
Experienced depot technicians retire with decades of accumulated process knowledge that a four-to-six-month training cycle only partially transfers, said GrayMatter.
Depot maintenance contractors are responding by deploying autonomous surface-finishing systems at the front of the workflow with surface preparation, where the labor constraint hits first and where consistent, repeatable output has the most downstream leverage.
GrayMatter shares FAQ about manufacturing readiness
GrayMatter Robotics provided replies to frequently asked questions about autonomous surface finishing:
How are defense manufacturers automating surface preparation and coating?
Defense manufacturers are deploying robotic systems that perform sanding, blasting, coating preparation, and inspection within air-gapped facilities. These systems operate without external network connectivity, meeting data-sovereignty requirements for classified platforms while processing parts with variable geometry and corrosion conditions unique to each maintenance cycle.
What physical AI capabilities does GrayMatter bring to defense surface finishing?
The company said it deploys AI-powered finishing systems purpose-built for defense manufacturing environments. GrayMatter’s air-gapped, edge-deployed architecture meets data sovereignty requirements for classified facilities.
Meanwhile Process Intelligence, the learned understanding of how tools, media, and workpiece materials co-evolve during process execution, enables geometry-agnostic processing across armored vehicle platforms without part-specific programming, it said.
How does adaptive sanding autonomy handle variable surface conditions in defense MRO (manufacturing, repair, and overhaul)?
Adaptive sanding systems combine vision-based surface scanning with active force control to adjust tool pressure and path in real time. This allows the system to address corrosion and coating buildup unique to each vehicle without manual setup between parts.
Submit your session idea for the 2026 RoboBusinessThe post Defense manufacturing readiness hinges on autonomous finishing, says GrayMatter Robotics appeared first on The Robot Report.
