Commercial and VC-Led or federally funded R&D: a debate on optimal defense innovation

A crucial distinction in this debate is between defense-unique technologies and true dual-use technologies. Defense needs often involve capabilities that have no natural commercial counterpart. Systems built for needs that involve contested, denied, or classified environments, such as hardened communications, secure logistics networks, and electromagnetic spectrum dominance. By contrast, true dual-use technologies are commercially viable in civilian markets and can be adapted for defense use with minimal modification – solutions such as AI-enabled supply-chain analytics or satellite-based sensing.

The lines between these requirements seem to be blurred in the public discourse, and we would point out that not every military requirement can or should be met by the private sector, and not every commercial innovation meaningfully advances national defense. Having a defense solution easily accessible in the commercial market makes it less of a discriminator over time if you are relying solely on tactical vs. technological barriers for access by adversaries.

The following is a debate for and against these positions, synthesized by the authors based on research and participant observation. We note that one may take for and against positions on one or more of these arguments without being fully for or against them all.

This article is presented to stoke conversation and external consideration and debate. It should not be construed as an official position by the authors or their organizations.

Efficient and effective innovation

Argument: commercial firms innovate faster and more efficiently

Advocates for commercial-led defense innovation argue that venture-backed firms operate under pure market competition, rewarding rapid iteration and efficiency. Venture funding models demand short development cycles and measurable outcomes, qualities often absent in government programs. The Federal Acquisition Streamlining Act of 1994 was driven by this logic: to leverage competition and market pricing to access the latest technologies faster and cheaper.

Proponents point to companies like SpaceX, which has achieved commercial scale in satellite launches and internet services. Although, it is worth noting that this success followed substantial federal development funding. They also point to other defense-tech ventures like Anduril and Palantir as champions of the prowess of the commercial sector. For policymakers, these examples suggest that commercial enterprise can bring cutting-edge innovation to the warfighter faster and at lower cost than legacy programs weighed down by oversight, compliance, and long acquisition cycles. The promise is a more responsive defense industrial base that evolves with technology rather than bureaucracy.

Further, development costs are borne by more users. Commercial products are built for a wide user base and backed by strong subject expertise, aimed at large total addressable markets. For buyers, that means the cost of development is spread across thousands or millions of users. The future cost of sustaining and improving those products is shared across that same base and influenced by performance expectations from a large user community, leading to lower cost and better capability. For sellers, this model supports taking on more development risk upfront, with the potential for very high returns once the product gains broad adoption. This dynamic is what makes dual-use technologies so attractive to defense leaders. They appear to combine commercial scale with defense utility, allowing the government to capture the benefits of broad cost-sharing, faster iteration, and private investment without fully funding development.

Counter: evidence shows federal R&D makes firms more innovative

Knott, Josephson, and Lee (2024) test the prevailing assumption directly. Using firm-level data from 2001 to 2021 and a dynamic difference-in-differences model, they find that receipt of a federal R&D contract leads to a sustained ten percent increase in R&D productivity, as measured by Research Quotient. The effect strengthens as the proportion of federal R&D rises.

This effect isn't a statistical artifact. When federal spending was reduced under the 2011 Budget Control Act, contractor R&D productivity increased rather than fell. That finding disproves the idea that government contracts inflate performance by guaranteeing follow-on procurement. Federal R&D appears to catalyze, not crowd out, private innovation, particularly when awarded through competitive mechanisms (Lichtenberg, 1988). This is consistent with findings by Moretti et al. (2021) that defense R&D creates technological spillovers and covers fixed costs that enable additional private R&D projects, and with Pallante et al. (2023) who found significant crowding-in effects from defense R&D and by Belenzon and Cioaca (2022), who demonstrate that federal R&D contracts tied to guaranteed public demand increase firms’ own R&D investment and scientific output, reinforcing the role of procurement as a demand-side innovation policy.

These types of firms often become the “darlings of D.C.” that struggle with fielding working, integrated, and sustainable solutions to downrange units. Recent reporting underscores the point: a 2025 U.S. Army memo described the joint Anduril–Palantir battlefield communications system as “very high risk,” citing fundamental security flaws, uncontrolled data access, and hundreds of code vulnerabilities (Reuters, 2025). The same system was said to lack user-action logging and proper security vetting for third-party applications, reinforcing concerns about readiness and operational trust. Palantir’s stock dropped following the disclosure, and both companies disputed the findings as outdated, but the episode highlights persistent challenges in moving commercial-tech prototypes into the field.

The requirement, not the trend, should drive the approach. Dual-use works when a defense need genuinely aligns with a commercial one, allowing the government to benefit from shared development costs and rapid private-sector iteration. However, when the requirement is uniquely military and demands classified operations, hardened systems, longer sustainment periods or extreme reliability, forcing it into a dual-use frame undermines both cost and capability. As defense-specific features are added, the product diverges from its commercial base and collapses the total addressable market that made it efficient in the first place. Calling something dual-use when it is not does not make it more innovative. It makes it more expensive and less effective.

Meeting the needs of the mission

Argument: government contracting inhibits innovation

Scholars and practitioners have long claimed that federal procurement processes discourage creativity. Josephson et al. (2019) describe how firms adapt to bureaucratic environments by cultivating compliance and lobbying capabilities rather than technological ones, trading innovation capability for influence and sustainment capability. The Defense Innovation Board (2024) similarly warned that tangled federal systems hinder rapid adoption and implementation of innovative solutions.

Counter: commercial innovation optimizes for what sells, not what is needed

Venture-backed innovation follows market incentives, not national security priorities. As Mariana Mazzucato (2013) argues, markets will not finance what cannot be appropriated and scaled for profit. This means foundational or mission-specific technologies with no commercial return are neglected. Defense challenges often fall into this category.

Normally, defense needs like secure communications, resilient logistics, and electromagnetic spectrum dominance lack consumer markets. Take, for instance, research into smaller modular nuclear reactors used on U.S. Navy submarines, there is not a sufficient consumer market for this technology to be developed at scale for investors to see sufficient margins to make meaningful returns to justify their initial investment and risk portfolio. Yet, the need for these reactors is paramount to the Navy across its entire portfolio. Without research investments from DARPA or the Office of Naval Research, many critically vital and mission-centric technologies would fail to naturally exist and mature.

Knott et al. (2024) show that federally funded firms pursue more expansive research trajectories, diverging from their previous technological domains. This exploratory behavior is critical in defense innovation, where solutions often emerge from pushing beyond market-ready ideas. Commercial firms, by contrast, tend to focus on incremental improvements to products that already sell.

The proposed National Defense Authorization Act changes to non-traditional defense contractor (NTDC) status amplify this concern. By incentivizing firms to minimize direct Independent Research and Development (IR&D) billing to government, the policy effectively rewards companies that pursue privately financed research agendas rather than government-directed ones. This creates a perverse outcome where the firms granted preferential access to defense contracts are precisely those least aligned with federal R&D priorities. When companies optimize their IR&D accounting to achieve NTDC status, they're likely optimizing away from mission-critical research that lacks commercial viability.

Visibility, control, and risk management

Argument: private capital de-risks government portfolios

Supporters of venture-based models contend that shifting early-stage risk to private investors saves taxpayer dollars. The Defense Innovation Unit and AFWERX have both argued that commercial prototypes can be fielded faster and at lower cost because venture funding covers pre-prototype development. In this view, the government becomes a late-stage buyer rather than a speculative investor, a fiscally conservative strategy aligned with market logic.

Counter: loss of federal insight reduces strategic control

When the DoD funds R&D directly, it retains visibility into research direction, technical progress, and potential dual-use applications. When innovation is privately financed, the government loses that insight. This "line-of-sight gap" undermines the ability to anticipate dependencies or align development with mission needs.

As Singer (2014) documents, nearly every transformative technology underpinning U.S. military superiority (the Internet, GPS, semiconductors, and lasers) originated from federally funded research. None emerged purely from commercial demand. Once that visibility is lost, defense buyers become dependent on private capital cycles and investor sentiment. When venture priorities shift, access to critical capabilities can evaporate.

Additional counter: financial instability creates strategic vulnerability

Recent patterns in tech investment reveal how circular valuation dynamics create national security risks. In September 2025, Nvidia announced plans to invest up to $100 billion in OpenAI to fund data centers, while OpenAI simultaneously pledged to purchase millions of Nvidia chips for those same facilities. Bloomberg described this as "an increasingly complex and interconnected web of business transactions" that resembles the circular financing arrangements of the late 1990s dot-com bubble (Carvão, 2025).

For defense innovation, this creates a critical vulnerability. Systems built on financially fabricated foundations face catastrophic failure modes when valuations collapse. When overvalued defense tech companies enter financial distress, their intellectual property, source code, and sensitive operational data become liquidation assets sold to the highest bidder. Bankruptcy courts prioritize creditor recovery over national security, enabling Foreign Ownership, Control, or Influence risks as adversaries or their proxies acquire critical technologies through distressed asset purchases.

Proven capabilities vs. new potential: traditional and non-traditional contractors

Argument: recent policy shifts favor non-traditional contractors

Recent NDAA language proposes relaxing Defense Federal Acquisition Regulation Supplement (DFARS) rules for NTDCs, with eligibility increasingly tied to how firms account for their IR&D spending. Under the emerging framework, firms that bill IR&D directly to the government over a certain threshold would not qualify for NTDC status, while those that reduce or eliminate direct IR&D billing would earn relaxed exemptions to federal regulatory requirements.

Advocates argue this creates incentives for firms to fund their own R&D, reducing government expenditure and encouraging entrepreneurial risk-taking. The policy assumes that privately funded innovation will naturally align with defense needs through market forces, making federal oversight of R&D priorities unnecessary or even counterproductive.

Counter: national innovation capacity declines when federal R&D contracts do

Federal funding for industrial R&D has dropped 81 percent as a share of GDP since 1972. During the same period, U.S. research productivity declined 71 percent (Knott et al., 2024). Their analysis suggests these parallel declines are causally related: federal R&D contracts increase firms' innovation productivity by approximately 10 percent, indicating that the decline in federal funding likely contributed significantly to the broader decline in U.S. R&D productivity.

Belenzon and Cioaca (2025) attribute part of this decline to a strategic policy shift in the late 1980s and 1990s, when the Department of Defense and DARPA began emphasizing dual-use and commercial-first innovation. The logic was that if civilian markets could sustain R&D, government could reduce its own funding burden. But this shift reoriented federal innovation policy away from mission-specific demand, treating government as one customer among many rather than the anchor client driving the national R&D pipeline. The government's retreat from direct R&D participation narrows the defense technology pipeline, increasing dependence on short-horizon commercial investors whose incentives often end before defense integration begins.

It is also worth asking the following question: what significant innovation has occurred solely by commercial and/or VC funding? The majority, if not all, the major technologies that shape our daily lives today, from advances in electronics, computing, AI, networks, etc. all came from some form of federal funding. In other words, the government, due to its long-term mission and orientation, is able to take more risks and fund more opportunities, than VCs who need a more immediate timetable for their returns.

Reconciliation: speed and stewardship

The evidence doesn't suggest abandoning commercial innovation. It warns against substituting it for public R&D. Effective innovation policy requires recognizing which side of that line a capability falls on before deciding where government must lead, where it can partner, and where it should simply procure.

Venture capital delivers speed, but federal contracts deliver direction. The two are complementary when properly balanced: government-funded R&D de-risks fundamental science and shapes national technology priorities, while commercial players accelerate prototyping and production. The agility debate around alleviating acquisition process friction isn’t really about who the firm is; it’s about what the requirement demands. Government shouldn’t relax oversight because a firm looks agile, but because a requirement truly calls for agility.

Federal participation should be seen not as a bureaucratic burden but as a strategic steering function that keeps innovation aligned with national needs rather than transient market trends. Commercial innovation is optimized for market viability. Federal R&D prioritizes mission requirements. These objectives sometimes align, but often diverge.

Both models contribute essential capabilities to the defense ecosystem. The challenge for acquisition leaders is determining the appropriate balance between market-driven efficiency and government-directed alignment, recognizing that the optimal mix likely varies by technology domain, development stage, and strategic priority.

Our U.S. innovation ecosystem achieved its historical strength through complementary public and private investment, suggesting that either approach alone may prove insufficient for future defense challenges. Both sides of the argument have good reasons for taking their stance, and a public discussion is warranted. Let the public debate begin…

‍