File
|
AI as a Structural Revolution in Warfare, Not Just a Weapon: The RAND Report’s Four Axes of Competition in Future Warfare and the ROK Armed Forces’ Challenges |
| August 20, 2025 |
-
Kwang Sup JooVisiting Research Fellow, Sejong Institute | myjohj1@naver.com
-
Recent examples such as the Russia–Ukraine war, the Israel–Hamas full-scale conflict, and the U.S. Department of Defense’s “Replicator” program illustrate that AI is not merely a weapons system but is fundamentally transforming the conduct and structure of warfare. A RAND report, An AI Revolution in Military Affairs?, published in July 2025, assesses that AI will fundamentally reshape the character of war by altering four key axes of competition in future conflicts.1)
The report identifies four principal axes of competition in future AI-enabled warfare:
① Quantity vs. Quality
② Concealment vs. Detection
③ Centralization vs. Decentralization
④ Cyber Offense vs. Cyber Defense
These four axes underscore that AI technologies are not simply matters of tools or equipment, but compel a fundamental reconfiguration of military strategy, organizational structures, and operational concepts. The following section examines each axis in greater detail and considers the structural transformations that the Republic of Korea Armed Forces must confront. -
AI is automating “decision-making,” shortening “time,” and dismantling the boundary between “humans” and “machines.” This, in turn, transforms the hierarchical structure of command and control (C2) into a decentralized network. RAND forecasts that in future battlespaces, “humans will determine the objectives, AI will generate methods in real time, and weapons systems will respond autonomously.”
The U.S. Department of Defense’s JADC2 (Joint All-Domain Command and Control), the Israel Defense Forces’ AI-enabled strike network, and Ukraine’s GIS arta system all point in this direction. In particular, following its conflict with Hamas, the IDF fielded an AI-enabled battlefield integration algorithm called “Fire Weaver.” It links sensors and shooters in real time and functions as an automated strike system that minimizes human judgment in the targeting loop.
JADC2 aims to digitally integrate the existing command–decision–execution chain, reducing decision and strike cycles that previously required hours to minutes. Fire Weaver analyzes target information collected by sensors with AI and automatically transmits it to appropriate fire assets, enabling autonomous strike decisions within seconds.
In this context, AI-based command systems are not mere supplementary functions, but are emerging as core combat assets that fundamentally enhance the speed and accuracy of operations. This shift is not a matter of simple technological introduction, but one that is impossible without transformations in operational concepts, force organization, and information management systems. For the Republic of Korea Armed Forces, the time has come to move beyond mere technology adoption and to pursue simultaneous innovation in organizational structures and operational culture through the delegation of command authority, the decentralization of information sharing, and the establishment of automated strike-decision systems.
This restructuring of command and the battlespace is closely connected to the four axes of future competition identified by RAND. The following section will specifically examine the meaning and operating principles of each axis, as well as the structural transformation challenges faced by the Republic of Korea Armed Forces. -
The four axes of future warfare are not independent variables, but interact with one another, such that improvements in one axis can compensate for weaknesses in another or generate new vulnerabilities. The structural reform tasks facing the Republic of Korea Armed Forces must therefore begin from an integrated design that takes into account the interactions among these axes.
1. Competition between Quantity and Quality
In the past, combat power heavily relied on the absolute quantity of personnel and equipment. However, AI simultaneously improves autonomy and cost-efficiency, enabling the mass operation of “good-enough” unmanned systems. This change makes force structures centered on high-cost, elite platforms vulnerable to quantitative competition. Based on Lanchester’s Square Law, RAND analyzes that the square of the force ratio has a decisive impact on combat power in the same battlespace. AI-based networks shorten detection and engagement times, maximizing the concentration effect of quantitative forces. Historically, the Me 262 fighter (a jet fighter) had a 9:1 firepower advantage over the P-51D (a propeller fighter) but lost due to a 3:1 numerical inferiority, while conversely, a 2:1 numerical superiority could secure victory even with half the firepower. The ROK military needs a system for the mass procurement, attrition management, and regeneration of manned–unmanned teaming (MUM-T) forces, scalable bundling of low-cost sensors and small strike systems, and a shift from “premium product” development to standardization, modular and replaceable design, and attribute-based testing and evaluation.
2. Competition between Concealment and Detection
AI strengthens detection superiority through multi-sensor fusion, target recognition, and signal analysis capabilities, but adversaries also respond with advanced deception using AI, such as false heat sources, electronic decoys, and deepfake terrain and electromagnetic signals. In particular, during the Russia–Ukraine war, reports indicated that about half of the drones in the air were operated not for actual strikes but for deception purposes 3) . This indicates that deception is not merely a supporting measure but a core determinant of force survivability and effectiveness in the detection–deception competition. Detection and deception mutually stimulate and co-evolve in a pattern of “action-denial/action-adaptation (AD/A).” The decisive factors are the diversity of data and the sophistication of signal profiles. Accordingly, the ROK military must institutionalize deception capabilities alongside enhanced detection, establish theatre-level multistatic detection networks4) and passive RF and UAV EO/IR surveillance networks, implement adversarial testing and training for AI models, and create a permanent red team structure with continuous red-teaming operations.
3. Competition between Centralization and Decentralization
AI speeds up decision-making and mitigates information lag, but risks from electronic warfare, jamming, and network fragmentation still remain. Therefore, decentralized mission command and ensuring the autonomy of field commanders are essential. However, this mission command should not rely solely on individual commanders’ capabilities; AI and C2 systems must standardize and systematize the process. In other words, a framework is required that supports commanders in making rapid and consistent decisions based on doctrine and procedures by integrating rule-based control, which links central operational intent with autonomous algorithms in the field, and learning-based optimization. The ROK military should establish brigade- and battalion-level distributed C2 experimental units, develop systems for automated Course of Action (COA) generation and evaluation, and institutionalize multi-path communication networks and procedures for “operations under network isolation” to ensure resilience against network fragmentation.
4. Competition between Cyber Offense and Defense
AI automates and sophisticates complex cyberattacks, such as vulnerability detection, exploit generation, tailored phishing, and supply-chain disruption. Future attackers may leverage AI to execute ultra-fast, large-scale assaults and employ adaptive techniques to evade or degrade defensive algorithms on targeted systems. Defenders must also harness AI to strengthen real-time anomaly detection, automate patching, block malicious activity, and improve data recovery capabilities. In particular, proactive vulnerability discovery through attack–defense simulations, end-to-end supply-chain threat monitoring, and verification of AI model provenance and data integrity are essential. The ROK military must establish software supply-chain security and model provenance verification mechanisms, ensure the integrity of training data, and strengthen log retention and audit trails. It should also institutionalize continuous cyber–physical integrated training and red-team penetration exercises involving critical infrastructure, including power grids, navigation, communications, and weapon-system control networks, to master the full cycle of detection, response, and recovery against complex multi-domain threats.
The four axes of competition are not independent elements but an interactive structure that can either reinforce or undermine one another. For instance, detection superiority is rendered ineffective without decentralized command, and a failure in cyber defense can collapse even qualitatively superior combat power. The ROK military should approach this competitive landscape by designing an integrated systems architecture rather than by adopting individual technologies. It is time to move from tactical adaptation to structural innovation. -
1. The Need to Establish a “Korean Version of the Replicator Program”
“Future battlefield dominance will be determined by the deployment of large-scale unmanned systems, and an unprepared military will lose by ‘quantity,’ not ‘quality.’”
In response to China’s proliferation of large-scale unmanned systems, the U.S. Department of Defense has been actively pursuing the Replicator Program since 2023. The program seeks to field thousands of low-cost unmanned systems within a short period to establish a qualitatively superior network of “autonomous combat nodes.” The ROK should also build an AI-enabled drone combat network that integrates command and control, sensors, communications, and fires, enabling it to respond effectively to large-scale drone warfare and unmanned-system disruption operations in the event of a contingency. To achieve this, a national-level comprehensive roadmap for autonomous weapons force modernization is required that connects defense R&D, K-defense technologies, and the startup ecosystem.
2. Establishment of a Governance Framework for Defense AI Ethics and Control
“Without ethical standards, the ROK risks losing leadership in the international competition over AI norms, which could ultimately cripple its own defense innovation in AI (see the U.S. Political Declaration on Responsible Military Use of AI).”
AI-based weapon systems inherently involve ethical controversies such as misidentification, inhumanity, and ambiguity of responsibility. The ROK Ministry of National Defense should establish comprehensive “AI Ethics and Control Standards” and institutionalize procedures for ethical vetting, legal accountability, and operational control from the initial stages of technology development. Furthermore, beyond internal military oversight, the establishment of an “AI Weapons Ethics Committee” that includes civilian experts, legal professionals, and technology ethics specialists should also be considered.
3. Expansion of ROK–U.S. AI-Enabled Joint Training
“Gaps in AI-based command and control and combined engagement procedures translate directly into differences in the operational tempo of the alliance.”
The United States has been accumulating practical training models that bridge the gap between AI operations and field application through operational experimentation programs such as Project Convergence5) and Ghost Fleet Overlord6) , which focus on integrating AI with weapon systems in real-world environments. In particular, the United States has moved beyond the level of simple unmanned weapon systems and secured an overwhelming advantage in C2 across all domains of warfare by leveraging AI within the framework of JADC2. South Korea should participate in such joint experimental exercises or co-develop similar systems to strengthen AI combat interoperability and operational experience within the combined battlespace. It is especially important to jointly establish AI-based combined engagement procedures across land, air, maritime, and cyber domains.
4. Establishment of an AI Experimentation and Acquisition Organization
“When acquisition lags, AI weapons remain ‘cutting-edge’ in the laboratory but become ‘obsolete’ on the battlefield.”
Organizations such as the U.S. Defense Innovation Unit (DIU) and the U.K.’s Defence Science and Technology Laboratory (DSTL) function as platforms that rapidly connect commercial emerging technologies to the military. The ROK should institutionalize a field-validation structure that links technology, operations, and tactics by creating an “AI Weapons Transition Experimental Unit” or a “Rapid Acquisition Test Cell” under the Defense Acquisition Program Administration or the Agency for Defense Development. Through this mechanism, civilian AI technologies should be able to transition quickly into weapons with demonstrated battlefield applicability and survivability.
5. Institutionalization of Budget Frameworks for AI-Driven Defense Modernization
“Only nations that invest in battlespace awareness rather than platforms will survive in the wars of the AI era.”
The current structure of the defense budget remains centered on weapon-platform requirements, leaving areas such as data-driven operational environments, AI algorithm advancement, and battlespace network development relatively neglected. Accordingly, the mid-term defense plan and annual budget framework for force improvement programs should establish dedicated investment categories for AI, data, and network capabilities. A flexible funding mechanism should also be introduced to allow resource allocation based on technological complexity and operational urgency. Such a structure would prevent AI-related programs from being marginalized by competing platform demands and enable stable investment across the entire cycle of development, experimentation, and deployment. Ultimately, the orientation of the defense budget must shift from “platform acquisition” to securing “battlespace awareness superiority.” -
AI is no longer a concept of the future. Even now the battlespace is being reconstituted by artificial intelligence and data-driven algorithms, and battlefield initiative is determined by information-processing speed and shortened decision cycles. The core of AI warfare is not simply whether one possesses the technology but the “structure” that absorbs and operates it. Merely fielding AI weapons will not suffice for future conflict; the imperative is to comprehensively and organically reform existing command systems, operational concepts, organizational structures, and training and education regimes to align with AI.
AI-driven future warfare is not simply a matter of integrating unmanned systems or algorithms, but a challenge of defense system innovation that transforms the entire military into an intelligent, organism-like entity. Achieving this requires a fundamental transformation that transcends traditional barriers of military culture and institutional norms, such as hierarchical decision-making structures, branch-centered operational concepts, and seniority-based personnel management practices.
RAND’s analysis warns that such a transformation is not a matter of choice but a prerequisite for survival. In future warfare, victory will be determined not by technological superiority but by the capacity for structural adaptation. The challenges currently facing the ROK military, including declining manpower, organizational inefficiency, and limited interoperability in combined operations, will become even more pronounced with the introduction of AI.
Accordingly, what the ROK military needs now is not the introduction of AI equipment, but a fundamental restructuring focused on “operations architecture centered on the battlespace,” “innovation in command systems to shorten decision cycles,” and “personnel redesign for AI operations.” RAND’s recommendations can serve as a compass for navigating the strategic dilemmas currently confronting the ROK armed forces. Ultimately, the Korean military must transform its structure before its technology, while there is still time to act.
| Introduction: The Structure of War Transformed by AI
1) RAND Corporation, An AI Revolution in Military Affairs?, WRA4004-1, July 2025, p.5.
| AI Reshapes Command, Reconfigures the Battlespace
2) https://www.spnews.co.kr/news/articleView.html?idxno=80971 軍, AI 기반 사격지휘체계 구축한다.
| The Four Axes of Future Warfare: Structural Reconfiguration Driven by AI
3) Associated Press, "Russian factory built decoy drones to fool Ukraine’s defenses — and make money," 2024년 12월 12일.
4) A surveillance network that disperses transmitters and receivers across different locations to broaden detection angles and signal processing coverage compared with a single monostatic radar, thereby increasing the likelihood of detecting concealed targets.
| Policy Recommendations: Reconfiguring the Triangle of Technology, Organization, and Institutions
5) A U.S. Army–led project aimed at building an integrated battlespace network across land, sea, air, space, and cyber domains by leveraging AI, cloud computing, and automation technologies.
6) A U.S. Navy–led experimental program designed to test the operational feasibility of unmanned surface vessels (USVs) in real-world maritime environments.
| Conclusion
※ The contents published on 'Sejong Focus' are personal opinions of the author and do not represent the official views of Sejong Institue
