The research infrastructure behind America's defense technology edge is "deteriorating," according to a Department of Defense report released Wednesday, and one driver is familiar to any R&D manager: money budgeted for facilities keeps getting diverted to operations.

The study, produced by the Office of the Under Secretary of Defense for Research and Engineering, warns that the Pentagon's "research, development, test, and evaluation (RDT&E) infrastructure is deteriorating and weakening the Department's ability to maintain a technically advanced warfighting capability." Authorized major military construction (MILCON) projects meant to modernize critical joint-mission RDT&E infrastructure "continually slip due to the services' reprioritizing of scarce MILCON funds toward other operationally relevant priorities."

The scope of the review was unusually broad. Investigators visited 30 sites — roughly one-third of the total — spanning government laboratories, federally funded think tanks and university research centers. The report describes the data collection effort as "unprecedented." That sample size matters when weighing the findings: two-thirds of facilities went unvisited, so site-level conclusions rest on a substantial but partial inspection base.

The verdict is mixed. The defense research enterprise, or DRE, is "fundamentally sound," the report finds. But it "needs to rapidly adapt to a new environment moving at the accelerating pace of commercial technology and driven by a broader set of global security threats." In practical terms, the reviewers identified a set of workflow and budget constraints that will sound familiar to lab directors outside government too.

Three bottlenecks stand out. Backlogged security clearances slow access to classified work. Limited funds to build or refurbish labs compound the MILCON slippage. And a slow, difficult hiring process discourages younger skilled personnel — a talent pipeline problem that directly constrains throughput at the bench level.

The Pentagon also lacks a clear picture of its own footprint. The report states DoD does not have a comprehensive inventory of specialized facilities, and it cites a concrete example: the Triaxial Earthquake and Shock Simulator in Illinois. For an enterprise that runs on portfolio decisions, the absence of a full facility register is a governance gap in itself.

Even the taxonomy is inconsistent. Pentagon research facilities "don't even speak a common language," the report states. "For example, a single technical domain may be variously labeled as 'Human-Machine Teaming,' 'Autonomy and Teaming,' or 'AI Agent Development' depending on the reporting entity. This semantic variability ensures that high-level titles alone are insufficient for determining the true depth or specificity of the work being performed." Anyone who has tried to consolidate R&D portfolios across business units will recognize the problem: without shared definitions, overlap analysis and gap analysis both fail.

Commercialization is the other weak link. The report finds that DoD's "vast intellectual property remains underutilized due to passive marketing and the lack of a centralized discovery mechanism." The administrative burden cuts deepest where it matters most: "Current administrative burdens for transition agreements often exceed the funding timelines of startups, hindering innovation and collaboration." For venture-backed companies that plan in 18-month runway increments, a licensing process slower than their capital clock is effectively a closed door.

The consequence is delay in getting technology to the field. "The Department's ability to rapidly transition technologies from knowledge producers to capability fielders is severely degraded by structural and cultural barriers, specifically bureaucratic stovepipes, fragmented funding streams, and misaligned mechanisms, authorities, and incentives," the report says.

Notably, the study explicitly rejects the consolidation option. "The evidence in this report does not support consolidating or eliminating institutions: Most overlap is driven by mission need," it states. "The findings suggest that the way to reform the DRE is to fix the system around its institutions — how authority, money, and decisions flow, and how the institutions are funded, measured, and governed." That framing shifts the reform agenda from portfolio rationalization — which often means closures — to process and funding-mechanism redesign.

The benchmark the report invokes is China. "China has solidified its civil-military fusion model and is investing on a scale and at a pace that requires the United States to develop a new Government-industry paradigm — one that advances and protects critical defense research while extending the technical reach of the DRE through industrial and academic partnerships the DRE itself does not own." That last clause signals a preference for partnership structures over acquisition — extending reach without adding owned infrastructure the department would then have to maintain.

The recommendations are concrete and operational: ease budgetary limits on lab refurbishment, use AI to speed up security clearances, and create a searchable database for DoD intellectual property. Each targets a specific bottleneck identified in the findings — facility funding, access clearance, and IP discovery respectively.

The report closes with the stakes stated plainly. "Without a robust, agile, and properly resourced DRE, the department cannot rapidly onramp innovative capabilities or deliver the advanced capabilities the warfighter requires to deter and, if necessary, defeat emerging threats in an era of intense strategic competition." With the recommendations now on record, the next data point to watch is whether Congress and the services adjust MILCON prioritization and clearance funding in the next budget cycle.