Artificial Intelligence (AI) is decisively shaping the future of warfare. It accelerates decision cycles, extends operational reach, and enables exercised control of the informational, and cognitive dimensions of engagement. The scale, tempo, and precision of AI surpass conventional means, and will be instrumental across command structures, intelligence, surveillance and response (ISR) networks, and both non-kinetic and kinetic force platforms.  We opine that AI is rapidly changing how national security is secured, sustained, and projected across contested domains. A challenge lies in synchronizing investments, managing risk exposures, and maximizing strategic yields across all layers of engagement. The table below presents core imperatives driving this effort:

The Human Weapon System: Neuro-Cognitive Integration

The Department of Defense’s most valuable and costly resource is its personnel. Operational success depends on transforming traditional operators into fully integrated agents of an AI-powered warfighting system; enabling- and enhancing- performance through amplified decision speed, expanded situational awareness, advanced cognitive capability, and precision under pressure.

Such transformative shift requires integrated action via a three-domain functionality of:

1. Enhanced warfighter cognition through neuromodulation, augmented reality interfaces, and adaptive feedback systems that synchronize human perception and actions with the scope and tempo of AI analytics.
2. Cognitive defense through real-time detection and disruption of adversarial influence via the use of AI-driven threat analytics that utilize secured, self-authenticating microchips.
3. Closed-loop human-AI integration, wherein command intent is registered through neural and behavioral signaling, thus compressing decision cycles to milliseconds and enabling machine-speed execution grounded in human judgment.

The infrastructure required to articulate these functions must deliver scalable, high-throughput computing capable of processing ISR data in real time. Such capability is critical for rapid targeting and adaptive maneuvering in dynamic operational environments. Data integrity is paramount to this enterprise; information must be accurate, controlled, secured, and optimized to support precise predictive analytics.

Deus in Machina: Microelectronics

Microelectronics are essential to all of these AI functions and applications. Algorithms operate only as quickly, securely, and as reliably as the silicon components that subserve them. Indeed, microelectronics define tempo, fidelity, and survivability of AI in contested domains. Hence, AI requires domain-specific processors, multi-modal signal chains, and mission-aligned chipsets engineered for battlefield conditions. These components, once fielded, expand operational reach and secure decision-cycle primacy.

Yet, operational yield, while advancing, remains fundamentally constrained by finite resources within a contracting defense budget across the 2027–2031 Program Objective Memorandum cycle. Compounding this reality, start-up costs for AI are substantial, requiring significant upfront capital that intensifies an already pressured fiscal environment. Consequently, channeling resources toward AI will reshape the defense portfolio, mandating explicit trade-offs. Investment in AI will inevitably necessitate some redirection of funding from existing programs, compelling a transparent assessment of capability prioritization. Each allocation toward AI carries a counterbalance—potential reductions in legacy systems, emerging capabilities, or critical support functions—with consequences that can, and likely will reverberate throughout operational readiness and strategic architectures and postures.

We assert that the imperative lies in discerning where AI investments yield decisive advantage and where careful reallocation aligns with overarching national defense objectives. Only through this careful calculus can transparency be achieved, ensuring that AI adoption enhances, rather than undermines, comprehensive mission effectiveness. Without clear and measurable return, AI risks displacing the very foundations of combat effectiveness.

Compromised Components: The Microelectronic Risk Vector

The discovery of undocumented “rogue” communication devices in Chinese-manufactured solar power inverters serves as a vivid reminder of the operational fragility of the global microelectronics supply chain. These compromised components—capable of circumventing firewalls, transmitting data covertly, and/or initiating system shutdowns—underscore a critical vulnerability: foreign-manufactured, mission-relevant microelectronics may serve dual-use functions that are neither transparent nor controllable under peacetime procurement regimes. This is a live vulnerability with infrastructure-wide implications.

In other words, the insertion of tainted components into domestic energy infrastructure effectively constitutes a latent kill switch. In a conflict scenario, the activation of these embedded systems could catalyze blackouts, damage grid synchronization, or degrade continuity of operations across critical defense installations. As AI-enabled systems increasingly rely on complex energy management hierarchies—many of which draw from the civilian grid—any compromise to these nodes jeopardizes the entire decision-execution architecture of the modern military. This challenge cannot be solved through software updates alone; it demands sovereign control of the microelectronic pipeline and battlefield-grade validation of every system component.

In the AI battlespace, trust in the substrate is non-negotiable. The expansion and integration of AI constituents and functions in other military systems proportionally increases potential for sabotage and a variety of levels and scales of manifest effects. Thus, strategic investment must prioritize domestic, secure microelectronic production—not simply for economic competitiveness but for mission capability, effectiveness and survivability. The recent revelations of corrupted microelectronic components validate the need for national-level defense-industrial policy that establish microelectronics as a foundational pillar of national defense.

Concluding Imperative: Maintaining Secure Microelectronics

We assert that it is not a question of if adversaries will exploit microelectronic systems essential to operations of AI and other human-machine interfaces becoming ever more integral to military operations, but when, where, and to what effect(s) such purloinments will occur and incur. The infiltration of unverified components into critical defense architectures poses a direct threat to operational continuity, data integrity, and command-and-control dominance. As AI-enabled systems become central to warfighting functions, the trustworthiness of their hardware becomes a core requirement of national defense policy. Thus, we propose the following recommendations to fortify the integrity and security of these essential components:

First, we advocate the importance and need for on-shoring of microelectronic design, fabrication, and integration for all mission-relevant AI systems. This should be prioritized and resourced as a matter of defense readiness. Domestic control over the end-to-end microelectronics pipeline is essential to ensure traceability, quality assurance, and resistance to embedded compromise.

Second, friend-shoring should be limited to strategically aligned defense partners operating under bilateral inspection frameworks, joint threat assessments, and interoperable security protocols. These arrangements must be structured through formal defense-industrial agreements—rather based on economic convenience or informal trust.

Third, off-shoring of critical microelectronic subsystems used in military technologies should be categorically prohibited. Any component integrated into systems supporting C5ISR, kinetic targeting, autonomous operations, or national infrastructure must be sourced from fully vetted and secure production environments.

Failure to control the origin and integrity of embedded systems creates latent vulnerabilities exploitable at scale in the event of conflict. The adversary’s intent is to disrupt, if not disable key systems preemptively—without warning, detection, attribution, or recourse. We opine that the DoD should iterate programs to both tactically thwart such adversarial efforts, and through such enterprise, sustain U.S. technological and strategic superiority. 

Disclaimer

Elise Annett is the Institutional Research, Assessment, and Accreditation Associate at the Eisenhower School for National Security and Resource Strategy, National Defense University; and is a doctoral candidate at Georgetown University. Her work addresses operational and ethical issues of iteratively autonomous AI systems in military use.

Steven Hanson (Col. U.S. Army, ret.) is an Associate Professor and the National Security Studies Department Chair at the Eisenhower School for National Security and Resource Strategy, National Defense University. 

Dr. James Giordano is the Director of the Center for Disruptive Technology and Future Warfare of the Institute for National Strategic Studies at the National Defense University.
Contact: james.j.giordano.civ@ndu.edu