When the world's at stake,
go beyond the headlines.

Wars are not won by speed and better decisions. They are won when weapons find their targets and destroy them. Yet in the race to build faster, cheaper, and smarter systems, the Pentagon risks forgetting the key source of that reliability: hard, specific intelligence on adversary military systems. Decision-makers who understand their opponent’s machines — not just their moves — can turn the tide of a battle and a war. U.S. leaders ought to make sure intelligence is fueling military innovation and adaptation. Otherwise, America’s foundries will produce weapons that break on contact with reality.

I’ve worked with other dedicated professionals to turn intelligence into lethal advantage for the U.S. Navy. I’m going to describe how intelligence has fueled past breakthroughs, examine today’s intelligence shortfalls across U.S. acquisition programs, and show how current policies leave fast-track weapons vulnerable to failure. I then offer recommendations on how policies can be improved to maximize the use of intelligence, and how the intelligence community can better prepare to fight the next big war.

The Limits of Decision Superiority

Decision superiority, or the ability to make faster and better decisions than the enemy, is the north star guiding American military modernization. Artificial intelligence will play an undisputable role in analyzing the threat data needed for the Joint Warfighting Concept, the Pentagon’s vision of great power war. But a myopic reliance on data-centric decisions may cause leaders to over-prioritize first mover advantage, and to gloss over the many other decisions — and weapons — that need to be made in a protracted war.

There is also a tendency to simplify intelligence as just information. This attitude overlooks the fact that good intelligence is the result of extremely technical and risky work. During World War II, Allied codebreakers identified Japanese intentions before the Battle of Midway, and collected deep insights on German military capabilities, which informed Allied invasion plans for Normandy. According to some assessments, actionable intelligence on German U-boat operations in the North Atlantic shortened the war by two to four years.

These anecdotes are worthy of inspiration, but they can also overstate the importance of decision superiority. Effective military innovations and adaptations are just as important as stealing the enemy’s war plans. No matter how good a military decision is, weapons still have to damage or destroy their targets.

Intelligence’s Role in Innovation and Adaptation

Historians observe that victory depends on how well militaries prepare themselves before a war, and how well they change after the first shots are fired. Innovation occurs during peacetime, when militaries have the luxury of time to carefully study the contours of a future conflict.

Case studies from scientific and technical intelligence, the discipline that studies how foreign weapon systems work, show how intelligence supports military innovation. During the Cold War, a dissident Soviet scientist named Adolf Tolkachev clandestinely photographed technical documents related to Soviet radar systems for the Central Intelligence Agency. Tolkachev’s insights helped the Pentagon innovate new electronic warfare capabilities, such as radar jammers and anti-radiation missiles. These capabilities suppressed the Soviet-built air defense systems of Libya, Iraq, and Serbia in the 1980s and 1990s, resulting in American air superiority in each of those respective military campaigns. In 1985, the tragic arrest and execution of Tolkachev and other Soviet informants shows the human cost of such insights.

In contrast, adaptation occurs during war when reality forces military commanders to gut check their original assumptions. As the scholar Williamson Murray observed, “one of the foremost attributes of military effectiveness must lie in the ability of armies, navies, or air forces to recognize and adapt to the actual conditions of combat.”

After the Battle of Midway, American forces captured a Japanese A6M Zero fighter that crashed in Alaska shows how intelligence supports military adaptation. The Americans repaired and flew the Zero to study its performance and used these insights to train new fighter pilots. Intelligence on the Zero also informed changes to the U.S. Navy’s F6F Hellcat fighter, which achieved an impressive 19-to-1 kill ratio against Japanese aircraft. Similar wartime exploitation of German ordinance during World War II and of improvised explosive devices during Operation Iraqi Freedom further shows the historic value of scientific and technical intelligence.

Acquisition Intelligence

Today, the intelligence community’s acquisition intelligence cadre — which I am a part of — plays an important role in creating new weapons. Serialized policies such as Department of Defense Directive 5000.01 and Department of Defense Instruction 5000.86 are some of the dizzying number of arcane documents that govern how intelligence is used to create new military hardware.

In a nutshell, these instructions require the U.S. military to make sure acquisition programs are continuously informed by and evaluated against the intelligence record. This process influences design requirements for new weapons or modifications to currently fielded systems and helps confer advantage to U.S. warfighters. To support this goal, acquisition intelligence analysts create a variety of reports that describe the threats program managers have to mitigate, such as missiles, satellites, hackers, and even supply chain vulnerabilities.

But intelligence costs money. Intelligence agencies need enough spies, sensors, and analysts looking at the right things to produce the data needed for those weapons to work. Vice versa, the Pentagon also has to make sure it’s not buying things that the intelligence community can’t feed. Acquisition intelligence plays a key role in managing this balance, which is codified in the soon to be defunct Joint Capabilities Integration and Development System.

In typical bureaucratic fashion, this doesn’t happen perfectly. A 2021 RAND report that studied U.S. Air Force acquisition intelligence comprehensively shows where this process breaks down. And in my recent experience, many of these issues broadly persist today.

In general, acquisition intelligence has a chicken-or-the-egg problem. There is a shortage of intelligence in acquisitions because there aren’t enough intelligence analysts supporting acquisitions. This scarcity results in hit-or-miss collaboration between intelligence and acquisition personnel. Where successful collaboration does occur, tight-knit professional relationships — or “bro-nets” — are more often used to fuse intelligence into acquisition decisions, instead of the formal step-by-step processes prescribed by policy. This alternative practice — while more dynamic and effective — sends muddled signals to the intelligence community to produce more analysis for acquisition customers.

Another problem is that intelligence support to ongoing military operations and national policymakers both take priority over acquisitions. Acquisition intelligence analysts typically collaborate with other technical and non-technical experts across the intelligence community to produce tailored assessments for acquisition programs. But when these intelligence analysts are forced to shift their limited focus on urgent requests, it puts acquisition stakeholders on a waitlist. This creates a perception that the intelligence community, despite its resources, is too slow.

Feeling the need for speed!

In 2020, the Pentagon published Department of Defense Instruction 5000.02, also known as the Adaptive Acquisition Framework, to fix the cumbersome nature of the Joint Capabilities and Integration Development System. In April 2025, this framework received particular attention in a White House executive order focused on acquisition reform. Accompanying the traditional “Major Capability Acquisition” pathway, the framework created two faster options: an “Urgent Capability” pathway, which fields a new capability within two years, and a “Middle Tier” pathway that seeks to do so in two to five years.

The Pentagon’s need for speed is leaving intelligence on the side of the road. There is no direct mention of intelligence in the above-referenced instruction. Likewise, intelligence only receives cursory discussion in two other serialized policies referenced in the framework. And according to optional intelligence guidance for the framework, one essential intelligence document is “ … required for Major Capability Acquisitions … but can be valuable to Middle Tier and Urgent Capability Pathways if the [Defense Intelligence Enterprise] is able to produce one.”

These inconsistent policies reveal three problems. First, they don’t require intelligence for Urgent and Middle Tier programs as rigorously as Major Capability programs. Guidance stating that program offices are “highly encouraged to formally request [intelligence] support” is not an explicit requirement.

Second, it assumes consistent organizational alignment between the intelligence community and acquisition enterprise, which the RAND report reveals is not always the case. In other words, there aren’t enough intelligence personnel playing zone coverage on an increasing variety of programs. This in turn tacitly accepts the intelligence community’s resource constraints.

Third, it puts the onus on the intelligence community to internally shuffle resources to support acquisition programs. Guidance states the importance of articulating a “demand signal so that the [Defense Intelligence Enterprise] can advocate for more resources.” While this is needed to accurately measure demand for budget purposes, this language overlooks the Pentagon’s heft in the intelligence community. The Pentagon’s Military Intelligence Program is roughly a third of the intelligence community’s budget, and half of the intelligence community is composed of Department of Defense elements. This is untapped potential.

The intelligence community needs to keep up. The previous administration’s Replicator initiative rapidly acquired aerial and naval drones for U.S. Indo-Pacific Command’s Hellscape concept to counter a Chinese invasion of Taiwan by the year 2027. Likely motivated by the same timeline, an August 2025 Pentagon memo directing further streamlining and congressional legislation oriented at similar reforms undoubtedly show that today’s efforts will only gain steam.

The Risks of Going Too Fast

Current acquisition intelligence guidance creates three risks. First, prioritizing Major Capability programs may create intelligence gaps for fast-track programs. The intelligence community is required to continuously report on threats to Major Defense Acquisition Programs, which are Major Capability programs that legally require more scrutiny from military leaders and Congress, such as the Columbia-class ballistic missile submarine. While such vigilance is needed to guarantee taxpayer spending, it skews the intelligence community’s analytic focus.

Second, current policies don’t incentivize the intelligence community to keep up with smaller, innovative programs that address the U.S. military’s most serious operational problems. Instead, the programs that taxpayers pay the most money for are the priority. As Hellscape might demonstrate, Urgent and Middle Tier programs could play a critical role in a Taiwan conflict and impact the war’s outcome in a strategically significant manner. The Replicator initiative spent approximately $1 billion for the Hellscape concept. And while Congress plans to spend $16 billion on more drones, artificial intelligence, and low-cost weapons, it’s unclear what the pathways will be for these systems. The focus on 2027 suggests that many of these programs will fall under Urgent or Middle Tier pathways. This means the Pentagon will spend billions of dollars on strategically significant programs that nominally don’t require intelligence.

Third and most dangerously, weak intelligence requirements may create weapons that fail to work against enemy countermeasures. This steepens the slope of military adaptation and increases the chance of U.S. defeat. And when the Pentagon is motivated by the urgency of 2027, pressure to meet costs and program deadlines could dangerously incentivize short cuts and ignorance of the intelligence record.

Solutions

The Pentagon is just starting to make huge changes to how it buys new capabilities, but it’s a big question how intelligence will factor into these reforms. Acquisition policies in their current form treat intelligence as an afterthought. Even worse, streamlining efforts may treat intelligence as red tape that needs to be cut. These fatal perspectives of intelligence need to be fixed now.

Of utmost importance, acquisition policies should explicitly state intelligence as a hard requirement for Urgent and Middle Tier pathways. Congress should match that mandate with earmarked money in the intelligence budget to make sure the intelligence community is prioritizing warfighter lethality. To prevent neglect, acquisition programs should also be evaluated on how well they dynamically address their threat environments, not just how well they meet cost, schedule, and performance goals. This is because military weapons are for warfighters, not financial auditors.

The intelligence community also should build close ties with an array of new commands and offices focused on rapid capability development and should make sure they have access to authoritative threat intelligence. This ensures intelligence is driving innovation and provides opportunities for these entities to communicate their knowledge gaps to the intelligence community. Such synergy also helps cue intelligence collection against adversary military programs.

Lastly, the intelligence community should study how it would support rapid adaptation during protracted war. Questions should include how to conduct foreign material exploitation in contested environments and how intelligence community resources can be reprioritized and surged to accelerate intelligence cycles and adaptation. U.S. Special Operations Command’s sensitive site exploitation processes during the Global War on Terror, and the U.S. Navy’s adaptation against Houthi missile and drone attacks in the Red Sea provide useful lessons on speed and agility.

Critics may assert that the intelligence community is too slow and that artificial intelligence will play the lead role in adaptation. In fact, lessons from the Russo-Ukrainian war suggests warfighters having direct access to raw intelligence is a best practice. But the increasing sophistication of Chinese weapon systems may exceed the technical expertise of uniformed intelligence analysts. During war, these analysts will be too busy maintaining a “common intelligence picture” for military commanders. Humans will still be needed to refine algorithms and to ensure AI-enabled analysis conforms to strict analytic standards. If there is a continued shortage of intelligence expertise or a lag in surging intelligence support after a war starts, warfighters won’t be able to quickly adapt. The fact that China is also doubling down on artificial intelligence suggests that the side with better intelligence — not more data — is more likely to win.

Critics may also suggest additional resources are unnecessary given the intelligence community’s increasingly large budget, which totaled $106.3 billion in FY-2024. The FY-2026 budget requested $33.6 billion for the Military Intelligence Program, which is a $5.4 billion increase from the previous year. Details below the intelligence community’s budget topline are classified. While a budget increase could mean more money for acquisition intelligence, acquisition policies still need to be updated to make sure intelligence is going where it is needed most.

Conclusion

Military leaders face some tough realities. It will take years for the United States to reinvigorate the defense industrial base. And China’s growing navy and long-range missile inventory will only increase the difficulty of a Taiwan conflict. Given these disadvantages, national leaders need to rediscover how intelligence catalyzes America’s lethality.

No matter when conflict occurs, the U.S. military will need to rely on a combination of better decisions, superior operational concepts, and more effective weapons to offset China’s numerical superiority, technological strengths, and advantageous fighting position. Intelligence will guide the lethal edge of American weapon systems when military commanders decide to strike, but intelligence is needed now to sharpen the nation’s steel.

Michael Borja is an intelligence officer in the U.S. Navy’s Information Warfare Community, and a former Navy Federal Executive Fellow at the Center for a New American Security in Washington, D.C. He is currently managing requirements and funding for Navy intelligence programs at the Pentagon.

The opinions expressed are those of the author and do not reflect the views or policy of the U.S. Department of Defense, the Department of the Navy nor the U.S. government. No federal endorsement is implied or intended.

**Please note, as a matter of house style War on the Rocks will not use a different name for the U.S. Department of Defense until and unless the name is changed by statute by the U.S. Congress.

Image: Midjourney