The U.S. Department of Defense's Adoption of AI for Missile and Space Threat Detection and the Structural Gap in the U.S.-China Military AI Competition

Executive Summary

The U.S. Department of Defense is pursuing the adoption of AI for detecting missile and space threats, but structural bottlenecks—high-speed targets, short response times, and ambiguous sensor data—remain unresolved. This push is driven by concerns over a technological deficit to China, exemplified by remarks from Treasury Secretary Besant. However, internal Chinese military assessments admit that the PLA's use of combat AI is still in an exploratory phase, suggesting a gap between the urgency felt by the U.S. and the actual disparity in capabilities. The most likely future path is a baseline scenario where these bottlenecks are only partially eased, leading to a phased adoption starting with lower-risk areas like space threat identification. A pessimistic scenario, in which political pressure outpaces technical validation, cannot be ruled out, increasing the risk of accidental conflict due to false positives and misjudgments. The case of Japan's Ministry of Defense developing AI-powered satellites shows that the regional space militarization race is already expanding beyond the U.S.-China dynamic. South Korea needs to adopt a phased approach, building its capabilities by simultaneously verifying interoperability and developing its own detection systems, rather than hastily integrating into U.S. standards.

I. Situational Analysis

The Pentagon's Adoption of AI for Missile and Space Threat Detection: A Situational Analysis

1. Background and Developments

The U.S. Department of Defense's missile defense system has long relied on the judgment of human operators. Three constraints have been identified as bottlenecks in real-world response: high-speed targets, short response times, and ambiguous sensor data [1]. Defense News reports that the Pentagon is pushing to adopt AI to overcome these challenges [1]. This is not a new initiative but rather an extension of the U.S. defense authorities' multi-year effort to advance their command and control systems.

This trend is rooted in concerns, widespread in Washington, about a technological deficit to China. At a recent Breitbart event, Treasury Secretary Besant remarked, "If China wins the AI race, nothing else will matter" [15]. He warned that even the vast U.S. defense budget would not be able to offset a disadvantage in AI [15]. While his comments were made in the context of local opposition to data center construction, they also reveal a facet of the political pressure driving the Pentagon to accelerate AI adoption.

China's actions are another variable shaping these developments. The Brookings Institution analyzes that China's People's Liberation Army Strategic Support Force (SSF) is purposefully pursuing military innovation in AI, centered on a command structure that integrates space, cyber, electronic, and information warfare [2]. An EAI report examines this in greater detail, noting that even internal Chinese military researchers have self-assessed that "the PLA's use of AI is still concentrated in auxiliary areas such as intelligent Q&A and content generation" [3]. They admit that the use of offensive combat AI remains in an "exploratory phase" [3]. This asymmetric assessment suggests a potential gap between the urgency felt by the Pentagon and the actual pace of China's progress.

2. Current Situation

The key takeaway at present, as reported by Defense News, is that the bottlenecks remain unresolved [1]. The uncertainty of sensor data and the short response times for high-speed targets are not problems that can be immediately solved by deploying AI algorithms. Instead, the data processing stage itself is emerging as a structural constraint hindering operational deployment.

This sense of urgency is also evident in other Pentagon programs. Defense startups are developing AI to optimize munitions management [5], and the Air Force is exploring rapidly deployable drone swarms for base defense [14]. DARPA has launched a next-generation hypersonic cruise missile program to counter China's air-breathing hypersonic missiles [12]. While these are separate projects, they all reflect the pressure for defense modernization stemming from concerns about the technology gap with China.

Within the region, Japan's moves are also intertwined with this dynamic. According to a report by Nikkei, Japan's Ministry of Defense plans to co-develop AI-equipped satellites with domestic companies for launch starting in the 2030s [9]. These satellites are intended to analyze space-based surveillance data to speed up intercept decisions and enhance counterstrike capabilities [9]. China's state-run CCTV described this as a "serious threat" and criticized it as a cornerstone of Tokyo's space militarization strategy [9]. This shows that the Pentagon's efforts to upgrade its AI-driven command and control are not limited to the bilateral U.S.-China dynamic but are also expanding along an axis of integrating allies' space assets.

On the political level, the Trump administration has reaffirmed its opposition to slowing the pace of AI development. During a visit to Ireland, President Trump characterized calls for stronger AI regulation as being driven by "negative forces," and the Republican Party has warned that strict AI regulations would cede a strategic advantage to China [13][16]. This shares the same political logic as the pressure for early adoption of AI in the defense sector.

3. Key Actors and Positions

U.S. Department of DefenseIts top priority is to secure an operationally deployable AI detection system. The goal is to use AI to supplement human operators in handling data volumes and response times that are beyond their capacity, but it is still grappling with the technical limitation of uncertain sensor data [1].

U.S. Treasury and AdministrationThey frame the AI competition as a matter of national survival. Secretary Besant's remarks reflect a sense of crisis that an expanded defense budget alone cannot overcome a deficit in AI [15]. Based on this reasoning, the Trump administration maintains its stance of easing AI regulations [13][16].

China's Strategic Support Force (SSF)It possesses the organizational strength of integrating space, cyber, electronic, and information warfare [2]. However, internal research acknowledges that the operational use of offensive combat AI is still in its early stages [3]. EAI analysis suggests this caution stems not from a lack of technological capability but from a structural constraint rooted in "political prudence surrounding Party-military relations and the delegation of command authority" [3]. On the other hand, China's low-cost commercial AI ecosystem is seen as having the potential to asymmetrically narrow the military AI gap [3].

Japan's Ministry of DefenseIt is pursuing enhanced intercept and counterstrike capabilities through AI satellites, effectively aligning with the U.S. trend of command and control modernization [9]. China has condemned this as space militarization, making the issue a catalyst that not only fuels the bilateral U.S.-China competition but also heightens Beijing's vigilance toward U.S.-Japan cooperation.

4. Core Issues

First, the data processing bottleneck is a practical barrier to operational application. The key is not the performance of the AI algorithms themselves, but the maturity of the upstream infrastructure for cleaning up signals from high-speed targets and ambiguous sensors in real time [1].

Second is the gap between the actual U.S.-China disparity and the perceived threat. The crisis narrative in Washington portrays China's AI catch-up as imminent, yet internal Chinese military assessments admit that offensive applications are still in their early stages [3][15]. This gap is a variable that will influence future U.S. defense budget allocations and procurement priorities.

Third is the direction of allied integration. Japan's AI satellite development is likely to be integrated into the U.S.-led command and control system, a factor that is intensifying China's backlash [9]. Middle powers, including South Korea, face the challenge of securing their own command, control, and monitoring capabilities within this integration trend [3].

II. In-Depth Analysis

The Pentagon's Adoption of AI for Missile and Space Threat Detection: An In-Depth Analysis

1. Analysis of Root Causes

The surface-level cause of this issue is technical. The combination of three conditions—high-speed targets, short response times, and ambiguous sensor data—pushes the decision-making speed of human operators to its physical limits [1]. However, the essence of the bottleneck lies in data processing itself. The missile defense system receives raw data from numerous radars, satellites, and ground sensors, but this data is noisy and comes in various formats. Introducing AI does not immediately resolve this problem. The task of creating a clean dataset that algorithms can learn from acts as a separate bottleneck in itself [1].

A more fundamental cause is the gap between political urgency and technological maturity. In Washington, policy discourse is dominated by concerns over a technological deficit to China. Treasury Secretary Besant stated, "If China wins this race, nothing else will matter" [15]. He warned that even America's vast defense budget would be powerless against an AI deficit [15]. Such statements translate into pressure on the Pentagon to rush unproven technologies into the field. Meanwhile, assessments suggest that the military AI capabilities of China, the designated adversary, are not as advanced. Even internal Chinese military researchers have admitted that "the PLA's use of AI is still concentrated in auxiliary areas such as intelligent Q&A and content generation" [3]. Their self-assessment is that offensive combat AI remains in an "exploratory phase" [3]. This implies a high probability of a gap between the urgency felt by the Pentagon and the adversary's actual capabilities. This gap itself acts as a political incentive to hastily adopt technologically immature systems.

2. Structural Context

Security Structure: A Shift to Competition in Command and Control

EAI analysis points out that the central axis of U.S.-China military competition is shifting from a contest for superiority in individual weapon systems to a "competition over integrated information and command and control capabilities" [3]. The attempt to introduce AI into missile defense is an American example of this shift. On the Chinese side, the Strategic Support Force (SSF) is pursuing a "systems warfare" concept by integrating space, cyber, electronic, and information warfare under a single command structure [2][3]. However, the EAI report judges that the SSF's integration efforts are hampered less by technical constraints and more by political caution regarding Party-military relations—specifically, the reluctance to delegate command authority to lower echelons or algorithms [3]. This is a structural constraint different from that facing the United States. In essence, the U.S. Department of Defense is stymied by a technical bottleneck (data processing), while the Chinese military is stymied by a political one (delegation of command). The different nature of these two bottlenecks is a key variable in assessing the future pace of the competition.

Economic Structure: Linkage Between the Commercial AI Ecosystem and Defense Procurement

America's adoption of defense AI is not a standalone Pentagon project but proceeds in conjunction with the commercial AI ecosystem. The development of AI for munitions optimization by defense startups [5] and the Air Force's move to procure drone swarms for base defense [14] are both predicated on a structure that absorbs civilian technology into defense systems. This contrasts with China's structure. The EAI report assesses that "China's commercial AI ecosystem has the potential to asymmetrically narrow the military AI gap by leveraging low-cost development models" [3]. In other words, while both countries rely on pathways to transfer civilian AI capabilities to the military sector, an asymmetric structure is forming where the U.S. focuses on high-precision, high-cost systems integration, and China has an advantage in low-cost proliferation.

Political Structure: Pressure for Defense Budgets and Deregulation

The Trump administration has made its opposition to slowing down AI development clear. Republican figures warn that strict AI regulations would cede a strategic advantage to China [13][16]. This sentiment is directly reflected in the Pentagon's race to adopt AI. It signifies a political environment where speed of deployment is prioritized over safety verification. This stems from the same political soil as the previously analyzed acceleration of DARPA's hypersonic missile program [12] and the Air Force's procurement of drone swarms [14], among other defense modernization efforts.

3. Historical Precedents and Comparative Cases

This is not the first attempt to delegate decision-making to automated systems in missile defense. During the Cold War, both the U.S. and the Soviet Union repeatedly experienced the risk of misjudgment from early warning systems. The lesson from that era was that the higher the level of automation, the greater the risk of misjudgment due to false positives, which is why the final launch decision always retained a human in the loop. The AI-based threat identification systems currently under discussion face a similar dilemma. The more decision-making is entrusted to AI to increase data processing speed, the less time there is to reverse a decision in the event of a false positive. The problem of "ambiguous sensor data" pointed out by Defense News [1] is the modern version of this very dilemma.

A regional example is Japan's plan to develop AI satellites. According to a Nikkei report, Japan's Ministry of Defense aims to develop AI-equipped satellites with domestic companies for launch in the 2030s [9]. These satellites are intended to analyze surveillance data to speed up intercept decisions and enhance counterstrike capabilities [9]. China's state-run CCTV criticized this as a "serious threat" and a cornerstone of Tokyo's space militarization strategy [9]. This case shows that the Pentagon's AI adoption is not an isolated event but part of a regional trend also occurring at the allied level. However, Japan's satellite plan is still in the prototype stage [9], so it is on a different timeline from the operational deployment bottlenecks faced by the Pentagon.

Another comparative case is Iran's attack on AWS data centers. The Brookings Institution reports that a drone attack by the Islamic Revolutionary Guard Corps in March caused structural damage to three AWS facilities in the Middle East [10]. This incident demonstrates that AI infrastructure itself can become a target of physical attacks, suggesting a new vulnerability the Pentagon must accept if it builds its missile defense AI to rely on cloud and data centers.

4. Key Variables Shaping Future Developments

The first variable is the maturation speed of data refinement technology. Given that the core of the bottleneck lies in the quality of input data rather than the algorithms themselves [1], the pace of progress in sensor fusion and data standardization will determine the timeline for operational deployment.

The second variable is whether the political constraints on China's SSF will be relaxed. If the reason the PLA's offensive AI application remains in an exploratory phase is not technology but political caution surrounding the delegation of command authority [3], the direction of the U.S.-China gap could change rapidly depending on when this constraint is lifted.

The third variable is budget allocation and procurement speed. Within the Pentagon, various AI programs—such as munitions optimization [5], drone swarms [14], and hypersonic defense [12]—are competing simultaneously. If AI for missile and space threat detection fails to secure priority, its development pace may lag behind the political rhetoric.

The fourth variable is the division of roles with allies. If allies, like Japan with its AI satellite plan [9], share the burden of regional surveillance and intercept decision support, the Pentagon's standalone data processing load could be partially alleviated. However, this comes at the separate cost of provoking a backlash from China and heightening regional tensions [9].