Will Artificial Intelligence Change Warfare?
-Future battlefield of mixed expectations and concerns

Introduction.

The invention of gunpowder contributed to the improvement of people's lives through the development of mines and social infrastructure, while also fundamentally changing the nature of warfare. And now in the 21st century, artificial intelligence (AI)[1] has been called "The Third Revolution in Warfare"[2] and is attracting attention not only for its potential to introduce new conveniences into society, but also transform war itself.[3] In fact, some 70 years after Alan Turing proposed the basic concepts of AI in the 1950s, the possibility that AI-powered autonomous unmanned systems and robots will play a leading role in the new battlefield is becoming a reality.[4]

The United States and NATO are rushing to adapt to Emerging and Disruptive Technologies (EDTs) to maintain military superiority, as they expect powerful AI and autonomous weapons to dominate future wars.[5] This is due to the recent development of information and communication technologies (ICTs) such as high-speed networks and big data, many of which can be converted to military use, as well as the rapid progress of so-called informatization in the military context. For example, drone technology capable of autonomous navigation, image recognition technology essential for automatic driving, and big data processing technology utilizing advanced compression technology are being put to practical use in the civilian sector, but are also subject to military diversion (spin-on). China, in particular, is promoting a policy of “Military-Civil Fusion” as a national strategy, and is working to prepare for an integrated human-machine combat system by efficiently converting civilian technologies to military use.[6]

In light of this situation, this paper will summarize expectations and concerns regarding the military use of AI and point out the ideal response for the nation to take.

What will be Changed and Expected?

Military use of AI enables rapid decision-making based on precise targeting information obtained from drones and reconnaissance satellites that utilize civilian technology, and innovative attack cycles that destroy targets using highly precise missiles and other guided weapons. A good example is how AI-powered attack drones and targeting systems are already being used in actual combat spaces in Ukraine and Israel.[7] The use of AI is also expected to greatly improve command and control capabilities through the rapid and accurate analysis of vast amounts of information data, as well as the provision of the results of this analysis to commanders and soldiers in a visualized form. To meet the growing demand for such AI-equipped weapon systems, many high-tech companies and defense industries continue to compete in their development.[8]

Currently, the rapid evolution and spread of generative AI, such as ChatGPT in the U.S. and DeepSeek in China, is attracting attention as its active use in the security and military domain is rapidly advancing.[9] In particular, Multimodal generative AI, which repeatedly collects and utilizes vast amounts of data such as text, images, and voice, when combined with edge computing,[10] will enable rapid information processing and battle situation analysis. This will contribute to improved operational performance and soldier survivability.

On the other hand, the spread of disinformation using enemy-generated AI is causing social unrest and deteriorating public security. In the case of cognitive warfare on the battlefield, along with cyber attacks, it has been pointed out that manipulation of information by adversaries may cause confusion and anxiety in command posts and units and may seriously affect operations.[11] One possible countermeasure is the introduction of predictive analysis and early detection of AI-generated disinformation.[12]

Thus, AI technology is playing an increasingly important role in the military field, both offensively and defensively. And as this collaboration with AI continues, it is expected to secure military superiority on the battlefield by enabling commanders and soldiers to focus their limited time and resources on creative and productive tasks.[13]

What are the concerns and Issues?

While the military use of AI is rapidly advancing, many issues remain unresolved. These are the lack of transparency and accountability, the risk of misjudgment, and ethical issues.

In general, the decision-making process of AI is opaque and its accountability unclear, which has led to an urgent need for "Responsible AI" with accountability that guarantees constant monitoring and intervention by humans.[14] This means that, as the military use of AI becomes more common, the military use of Responsible AI should be promoted in Western countries to strengthen the identification of and defense against malicious use.[15]

In addition, if the algorithm of the AI in question or the data to be learned is biased, it may lead to erroneous judgments and choices, allowing the military misuse of AI. This could lead to unexpected military escalation, including accidental bombings on civilian facilities, disturbance due to false information, and prolonged battles.[16]

Furthermore, ethical issues associated with the military use of AI have been noted. Currently, the Group of Governmental Experts (GGE) of the Convention on Certain Conventional Weapons (CCW), which prohibits or restricts the use of certain conventional weapons that have inhumane effects, is working on regulation for Lethal Autonomous Weapon Systems (LAWS), which may carry out attacks automatically without human judgment or intervention, to align their use with humanitarian and international law.[17]

However, discussions in this international framework have not been able to agree on even the definition of LAWS, partly because LAWS are not equipment that actually exist[18], and it is expected that the future of UN-level negotiations will see significant gridlock. On the other hand, there is no time to spare in responding to the rapid evolution of AI technology, which is accelerating internationally. As part of efforts to overcome this situation, the "Responsible Artificial Intelligence in the Military Domain (REAIM)" summit in 2023[19] was held to promote the cooperative governance of military use of AI among relevant countries that share a common awareness of the issues.[20] While it is gratifying to see the international discussion on the military use of AI gaining momentum, is there any problem in the actual development, implementation, and operation of AI?

Building AI Interoperability

Currently, each country is making national efforts to develop and utilize AI as a key to economic development. However, if each country adopts its own development path and does not pay attention to guaranteeing the compatibility of AI systems, there is a risk of creating systems that lack interoperability in the security field.[21] Even among allied nations that are supposed to conduct joint military operations, there is concern that differences in the scale and speed of implementation of advanced technologies in their respective societies may lead to a division and polarization of capabilities in joint AI operations among member countries.[22]

In order for allies and partner nations to conduct joint operations as effectively in the future as they have in the past, a common framework for the joint use of military AI must be established[23] and, specifically, interoperability of AI in the military[24] must be ensured. In fact, the U.S. National Security Commission on Artificial Intelligence (NSCAI), while warning that advanced technologies could divide allies in Europe and the Indo-Pacific region and others, emphasizes the need to promote the early adoption of coordinated AI military technologies.[25] On this point, it should be noted that the intelligence alliance "Five Eyes"[26] is coordinating on AI, cyber, semiconductors, and other advanced technologies, including civilian ones,[27] to ensure interoperability for military use of AI.

Conclusion

Japan adheres to a defense strategy that makes its alliance with the U.S. the cornerstone of its security policy, and has affirmed that it will cooperate with maintaining and strengthening the Japan-U.S. alliance. In the future, as countries with shared values trend toward the coordinated military use of AI amongst themselves, the vision of developing interoperable defense capabilities using AI should be clearly stated and shared between Japan and the United States. And it is necessary to deepen interoperability with the U.S. not only through joint development of cutting-edge AI and cooperation in implementing equipment, but also through the formulation of security strategies and doctrines, including responsible AI sharing.

Such efforts could contribute to actively creating long-term stability in regional affairs through responsible sharing of AI and eliminating the possibility of malicious use of AI by the U.S.-Japan alliance, as the military use of AI is expected to grow rapidly in East Asia.

In addition, it is essential for Japan to continue to be involved in creating rules for AI’s military use in the international arena, including the UN, to resolve ethical issues.[28] This is because, as the U.S. will now tend to advance its “America First” policy, Japan is expected more than ever to play a role in improving global solidarity with the EU and Asian democracies, with whom it shares values, and to take the initiative in shaping an inclusive and stable international order.

(2025/04/02)