Technological Evolution on the Battlefield

Technological Evolution on the Battlefield

Photo: Pavlo Bahmut / Ukrinform/Future Publishing/Getty Images

Innovation

The war in Ukraine has established a blistering cycle of measures and countermeasures, with both sides rapidly innovating to stay ahead of enemy advancements. According to Nick Reynolds of the Royal United Services Institute, current technology has a "six week learning cycle on the battlefield."¹⁷ This dynamic is most acute in electronic warfare. As each side has found new ways to jam drone signals, the other side has found ways to get around that jamming. Ukraine has significantly improved its radio jamming capabilities and can now disrupt the communications link between Russia's satellite-guided KAB and UMPK glide bombs and its GLONASS satellite constellation (Russia's equivalent of GPS), causing Russian glide bombs to veer off course by up to a kilometer and detonate harmlessly in open fields.¹⁸ Further, Ukraine has developed AI-enabled drones that can lock onto pre-identified targets in the final phase of flight-an innovation designed to counter Russian jamming.¹⁹ Both sides have reverted to operating drones using fiber-optic cables, which keep drones tethered but can reach up to a 10 km range and are impervious to jamming. Additionally, Russia has made advancements in directed energy weapons, providing Moscow with a more cost-effective way to counter Kyiv's inexpensive FPV drones.²⁰ Ukraine has also adopted a more decentralized communications model, using multiple dispersed radio nodes and smaller radios to reduce the detectability of its signals and make it harder for Russian forces to track and jam their communications.²¹

Innovation does not necessarily mean high-tech solutions. As AI-enabled drones have gotten good at locating tanks and armored vehicles, the Russian military has switched to donkeys and horses for moving troops and delivering supplies. "It's better if a donkey gets killed than two men in a car carrying the things necessary for battle and sustenance," said Russian Lieutenant General Viktor Sobolev.²² Meanwhile, the Ukrainian military has employed hand carts for the same purpose.²³

Information Warfare

Information warfare has leapt ahead in the Ukraine conflict, with Moscow focusing on spreading mis- and disinformation at home and abroad and Kyiv using facial recognition to identify Russian soldiers and lost children. Both sides have capitalized on AI. In March 2022, for example, Russia released a deepfake video of Ukrainian President Volodymyr Zelensky surrendering.²⁴ AI has also played a crucial role in documenting and verifying facts on an unprecedented scale. Ukraine has used AI-powered facial recognition software to identify over a quarter of a million Russian soldiers in the country.²⁵ The software was also instrumental in locating 198 missing Ukrainian children who were kidnapped and taken to Russia early in the war and in enabling Ukraine's Prosecutor's Office to identify the people in Russia who "adopted" them.²⁶ AI has further been used to counter Russian propaganda: While Moscow often downplays or conceals its battlefield losses, Ukrainian authorities have used the same AI to create an online database of identified Russian soldiers killed on the battlefield to notify their families in Russia.²⁷ Additionally, AI software has been used to collect evidence of war crimes, clear land mines, assist in refugee resettlement, and even combat corruption.²⁸

Cyber War

Both Russia and Ukraine have used cyberattacks as additive to their war efforts. Even before the full-scale invasion, Russian-affiliated cyber actors targeted Ukrainian oil and gas companies, Ukraine's largest commercial bank, and the Ministry of Defence's websites.²⁹ These attacks were likely aimed at undermining the Ukrainian public's trust in the military, disrupting their access to money and fuel needed for evacuation, and trapping them in the line of fire, further dampening hope. As the conflict escalated into open war, Russia shifted its cyber focus to government institutions, communication networks, power grids, and media.³⁰ As of April 2024, technology companies on the ground in Ukraine reported an ongoing onslaught of Russian attacks, in particular directed at the power grid and banks. Ukraine, too, has expanded its own cyber operations, primarily through its "IT Army"-a volunteer force of thousands of hackers conducting offensive cyber campaigns against Russian financial systems, state services, and media (to counter Russian disinformation and propaganda campaigns and conduct propaganda campaigns of their own).³¹ Ukraine in particular has proven highly resilient to these attacks-the result of more than a decade of preparation and enduring Russian cyber operations.

AI-Enabled Information Processing

The speed of information processing and decisionmaking is also rapidly changing, with both sides using ISR drones to collect vast amounts of data and AI to exploit the data for usable insight. AI plays an increasing role in Ukraine's targeting operations. Ukrainian UAS units now use AI to automate drone takeoffs and landings and assist in target identification (albeit with human oversight), sometimes reducing the time from detection to destruction to just over 30 seconds.³² The Ukrainian Ministry of Defence is using AI software to "analyze satellite imagery, open-source data, drone footage, and reports from the ground" and provide Ukrainian commanders with lists of potential military targets.³³

Incorporating Commercial Tech

Recognizing the advantages of commercial technology, Ukraine has begun restructuring its military acquisition system away from traditional state-owned research and development (R&D) models in favor of the commercial sector, a shift driven in part by the battlefield successes of commercial technology and cost advantages.³⁴ Ukraine has also streamlined its procurement processes by offering economic incentives to private companies and eased its restrictions on AI defense development. Kyiv has also significantly reduced administrative barriers for acquiring unmanned systems (cutting procurement timelines from months or years to just weeks) and increased the adoption of off-the-shelf commercial technology, allowing military units to bypass long wait times for custom-developed systems and quickly acquire new technology.³⁵

Lessons from Israel

As in Ukraine, emerging technologies have played a critical role on the battlefield in Gaza and are reshaping the conduct of the war. However, the Gaza war has yielded fewer insights on the impact of emerging technologies, as Israel was already a global leader in incorporating technological solutions. Further, the conflict was less evenly matched-Israel's quick bursts of activity and Hamas's lack of technological prowess have not resulted in the leapfrogging technological achievements that have featured in Ukraine.

Among the technologies that have been publicly acknowledged, the most notable is the AI-enabled decision-support systems (AI-DSS) that the Israel Defense Force (IDF) uses for targeting, which has dramatically accelerated the processing and analysis of battlefield information. AI tools have also supported Israeli forces in tracking the movements of suspected Hamas operatives at checkpoints across Gaza and the West Bank. To support these systems, Israel has significantly increased its demand for data storage and cloud computing, drawing heavily on commercial providers as well as its startup ecosystem to rapidly field new technology for the battlefield.

High-Speed Information Processing

The ongoing conflict in Gaza is rapidly transforming how information is processed and used in warfare. For instance, Israel has increasingly relied on AI-DSS such as Gospel for its targeting.³⁶ Gospel is a decision support tool used by the IDF that aggregates vast amounts of intelligence data, including "cell phone messages, satellite imagery, drone footage and . . . seismic sensors" to identify potential Hamas compounds, bases, and homes for targeting.³⁷ Gospel is capable of generating significantly more targets than traditional intelligence teams. Previously, IDF officers could manually identify 50 targets per year; Gospel can generate more than 100 per day.³⁸ These AI-generated recommendations are then reviewed by human analysts, who relay approved targets to the Israeli Air force, Navy, and Ground Forces through an app called Pillar of Fire.³⁹

Israel has also increasingly relied on AI-driven facial recognition at security checkpoints. The IDF uses these systems to scan the faces of passing individuals and to detain those flagged as having ties to Hamas.⁴⁰ While facial recognition technology has been used in the region for over a decade, Israel has significantly expanded its use during the current war, using tools developed by Corsight, a private Israeli company, to scan and cross-reference the faces of Palestinian residents against a "wanted persons" database. If the algorithm identifies a match, they are detained for questioning.⁴¹

Commercial Technology

As in Ukraine, the conflict in Gaza has witnessed a sharp rise in the use of commercial technology on the battlefield. Like the Ukrainian Armed Forces, the IDF is capitalizing on the strengths of smaller, more agile companies capable of rapidly fielding new and innovative designs. For example, as of late 2024, the Israeli Defense Ministry enacted a "green path" program for certain startups to fast-track their licensing processes.⁴² Between October 2023 and December 2024, the ministry also awarded 101 contracts collectively worth ILS 782 million ($219 million) to startups and small firms, many of which sprang up out of skills gained in service with the IDF.⁴³ According to military expert Isaac Ben-Israel, startups in particular excel in this environment, as they are often "a group of few people that can do something in weeks" rather than months or years.⁴⁴ This shift has been particularly evident in the development of anti-drone technology-a critical need as the IDF faces a constant barrage of varied drones and hardware launched from Gaza, Lebanon, Iran, and Yemen. In response, startups have provided approximately 50 percent of the anti-drone systems deployed by the IDF during the conflict so far.⁴⁵

Ravenous Need for Data Storage

The ongoing war in Gaza has highlighted the growing wartime demand for large-capacity data storage and cloud computing-capabilities now largely provided by commercial vendors. As the IDF's use of AI has expanded during the conflict, so too did its need for supporting cloud infrastructure.⁴⁶ However, by the early months of the war, Israel's domestic server capacity had come under strain, possibly due to the large mobilization of reservists from the technology sector and preexisting declines in foreign direct investment. In response, the IDF significantly increased its reliance on overseas cloud providers.⁴⁷ According to the Associated Press, the amount of IDF data stored on Microsoft servers more than doubled between March and July 2024, surpassing 13.6 petabytes-the equivalent of roughly 14 billion printed books.⁴⁸ Beyond direct military applications, Israel has also relied on foreign cloud providers to support systems such as Rolling Stone-a tool developed by Israeli security forces to manage certain population registries in the West Bank and Gaza.⁴⁹ (It remains unclear whether this is part of the same system used at border checkpoints mentioned earlier.) As the war continues, the demand for cloud computing and expansive data storage is only expected to grow.

Danger of Overreliance on Technology

The ongoing war in Gaza has revealed the ethical and practical risks of overreliance on technology. The pace of Israel's offensive against Gaza led the IDF to lean into AI to ease the burden on operators. According to The Guardian, the IDF's policy of targeting all individuals with ties to Hamas, including those of junior rank, significantly expanded the scope and volume of potential targets and quickly overwhelmed human operators.⁵⁰ Some analysts admitted there was insufficient time to carefully "incriminate every target" while another admitted to spending just 20 seconds per target, processing dozens each day, and contributing "zero added-value as a human, apart from being a stamp of approval."⁵¹ Further, during the early months of the Gaza conflict, IDF commanders pushed their analysts to "bring [them] more targets," which caused human analysts to increasingly defer to and trust AI's recommendations.⁵² Border guards fear making errors and tend to assume that the AI is more accurate than they could be, leading to false positives.⁵³ While the Israeli firm behind the facial recognition system claims its technology can accurately recognize a face even if 50 percent is obscured, anonymous IDF officers told the New York Times that the software still struggles with partially covered faces and grainy drone footage and noted that the system occasionally misidentifies individuals as being connected to Hamas.⁵⁴ Israel also leaned heavily on technology to secure the border between Gaza and Israel before October 7, with disastrous effects. Hamas knew how to dismantle static, automated systems, rendering defenses largely useless. (For more on this dynamic, see Chapter 7: Intelligence in a Transparent World.)

Implications for the Future of Warfare

The last three years of warfare have prompted a leap forward in technology on the battlefield, and the near future points to several continuing trend lines. First, the stealthy maneuver of large land and sea forces will become increasingly rare, thanks to small aerial drones using flexible, cheap commercial overhead imagery, crowd-sourced intelligence, and camera-guided one-way USVs at sea. Logistics chains will be in constant peril for many of the same reasons and will increasingly depend on disposable UGVs to perform tasks too dangerous for humans.

Defenders who can dig in with effective counter-drone measures will have a significant advantage, and those who must cross open land will be at a significant disadvantage. Future commanders will have access to a massive amount of information from a multitude of sources, demanding that they operate on a rapid-spin OODA (observe, orient, decide, act) loop. There is great promise for AI to make this process easier. It has already appeared on the battlefield in limited ways, but it is poised to rapidly expand as warriors get comfortable with the technology and iterate on its use.

When both sides have AI-enabled targeting and processing, the incentive will lean heavily toward deferring as much as possible to the AI's capability. For now, that surely will lead to mistakes.

With that ease comes the peril of using AI as a crutch: When both sides have AI-enabled targeting and processing, the incentive will lean heavily toward deferring as much as possible to the AI's capability. For now, that surely will lead to mistakes. AI is not trained or tested enough to take on such life-and-death responsibility. Finally, this entire ecosystem of rapid innovation and speedy decisionmaking will require the heavy involvement of industry, not only back at factories and in labs, but at or near the front lines, to receive rapid feedback and anticipate the next adaptation.

Ready to Fight in Full View

UAS units will prove a quick, lethal tool in future conflict. Drone teams will be able to detect enemy armaments and movement on supply lines within minutes and take rapid action to destroy targets. Units must prepare to fight in full view of the enemy, rendering large-scale combined arms maneuvers less common. Troop transport will be dangerous, necessitating movement in small numbers and likely under cover of some other purpose. Meanwhile, valuable military targets will need to be outfitted with increasingly advanced electronic warfare systems and constantly patrolled by interceptor drones to counter enemy UASs and glide bombs. The air domain will be increasingly contested, with UASs engaging in aerial combat for temporary control of the skies. The sea domain will also change dramatically: If a $500 drone can destroy a multimillion-dollar tank, so, too, can a USV swarm destroy a fleet of ships. Counter-drone solutions will be decisive on the future battlefield.

Highly Empowered Individual Units

The future of warfare is shifting toward smaller, highly mobile, adaptable units, where field commanders are empowered to make decisions about cover, concealment, and tactics while minimizing their communications signature. These units will rely on technology designed for hit-and-run attacks and ambushes, such as quiet ISR drones followed by autonomous swarm attacks, which might provide a distraction from a precise sabotage operation that must be conducted by a human. Rather than large-scale mechanized maneuver-such as the tank battles of Desert Storm-guerrilla-style tactics, hit-and-run operations, and sabotage will define the front lines of battle.⁵⁵

Troubled Logistics Tails

Supplying the front lines will be more challenging-and likely deadlier-than ever. Modern armies need to anticipate that only a fraction of supplies will get through, given adversaries' ability to identify and eliminate targets quickly. As a result, armies will seek to minimize deliveries to the front lines. Where possible, drone deliveries of goods will be preferable to protect the lives of pilots, sailors, and cargo drivers. Additionally, militaries will increasingly turn to inexpensive, expendable UGVs for mine clearing, mine laying, and frontline reconnaissance.

In response, militaries will adopt technologies that increase self-sufficiency. For instance, advancements in 3D printing could enable the on-site production of specialized replacement parts, reducing the need to use long and vulnerable supply routes.⁵⁶ Similarly, bacteria-based biofuels could allow units to generate their own energy, increasing mobility and reducing reliance on traditional fuel supply lines.⁵⁷ Autonomous fuel-delivery vehicles are also expected to play a greater role in resupplying frontline units, minimizing the need for manned convoys that remain vulnerable to enemy drone attacks.⁵⁸

Highly Networked Forces

Future militaries will be equipped with cutting-edge, integrated technologies that form highly advanced battle networks. These include quantum positioning systems, autonomous drones, and autonomous or semi-autonomous support robots. These systems will require two enablers: (1) dependable communications to network the effort and (2) fuel to keep it running. First, secure, redundant communication systems are essential to making this highly networked form of war possible. Units will need resilient mesh networks to ensure continuous communication even if multiple nodes go dark.⁵⁹ Second, highly mobile units will need mobile fuel. Ideally, they will not struggle under the weight of heavy rucksacks loaded with old, bulky batteries and instead can use compact, efficient, newer forms, alongside readily dependable alternative fuel, like high-efficiency solar and biofuels.

Rapid Adaptability Cycles

Future wars will likely feature extremely fast cycles of innovation and adaptation, as seen in Ukraine. With a more widespread battlefield, the front lines may evolve at different paces or in divergent ways, necessitating central nodes to facilitate sharing lessons learned. To keep up with the pace of innovation, militaries must adopt a more streamlined procurement process, allowing commercial vendors to rapidly iterate. Additionally, the speed of adaptation means creating bespoke equipment will likely be too slow; units must make do with off-the-shelf commercial technologies. Central commands can help push these commercial products in the right direction by telegraphing anticipated needs, giving industry a strong demand signal and a head start on the next iteration. This approach bypasses the time-intensive process of developing custom technologies and could reduce procurement timelines from years to weeks.

No Front Lines

The future of warfare will further blur the line between combatants and noncombatants. As the commercial sector takes on a greater role in military operations, civilian service providers-such as those supplying power, cloud storage, and internet connectivity to warring nations-could increasingly be seen as legitimate military targets.⁶⁰ Future militaries will need to develop clear policies about defense of private sector assets. For instance, companies deploying personnel and equipment to the front lines may warrant greater military protection than those providing services from thousands of miles away. As militaries become increasingly reliant on commercial technology, they will have to navigate the influence of corporate leadership, many of whom have their own (sometimes conflicting) sets of priorities-such as revenue growth, shareholder interests, or company reputational risk. This misalignment is likely to become a recurring challenge in military operations.

Humans and AI

Future militaries will likely depend heavily on AI for targeting, with AI systems able to autonomously identify and eliminate targets with minimal or no human intervention, making combat faster.⁶¹ Militaries already have autonomous systems for defense, in particular missile defense; a shift to offense is likely to take place first as matched targeting-for example, AI drone swarms attacking AI drone swarms. Later usages will include autonomous "find, fix, finish" of clear military targets, like ships and tanks. The last frontier will be strikes on particular human targets. However, increasing reliance on AI in life-and-death decisions raises serious ethical concerns. Human error in war is already too common today, but if a human operator allows an AI system to mistakenly target a school bus instead of an enemy tank, who bears responsibility? What are the accountability mechanisms? If an AI system is the cause of a friendly-fire incident, who is to blame?

The Next Decade of Warfare

The future holds other uncertainties for warfare in the next 10 years. Adversaries are already employing AI for deepfakes designed to sow doubt and confusion. The next iteration could entail deepfake military orders or highly realistic denial and deception operations designed to sow tactical chaos. In the next few years, sophisticated adversaries will probably find ways to combine AI and cyberattacks, with offense likely outstripping defense at first. AI agents will be able to find vulnerabilities and exploit them, perhaps in series, without phoning home and raising alarms, which will allow for persistence on networks in a way never seen before. A quantum computer able to defeat military-grade encryption is likely 5 to 10 years out, and the first state to use it will hold an unparalleled advantage. A combination of AI and powerful computing power, quantum or traditional, will allow for leaps ahead in bioengineering, including new chemical combinations and edited viruses for biowarfare; alternatively, these advancements could create biofuels or medicine tailored to a soldier's specific needs. Each of these adaptations could change the way of warfare all over again.

Conclusion

The conflicts in Ukraine and the Middle East have shown how emerging technologies-particularly unmanned vehicles, AI, and information warfare-are reshaping combat, forcing militaries to adapt or risk obsolescence. Modern conflicts are increasingly defined by speed, adaptability, and innovation. Ultimately, the next generation of warfare will not be defined solely by who possesses the most advanced technology, but by who can integrate, adapt, and counter it the fastest.

This trend line will challenge most political systems based on capitalism and democracy. The market will take time to catch up to need and respond to demand, whereas a centrally planned system will shortcut those steps. The United States in particular, despite excelling at invention and problem solving, is tragically slow at purchasing and integrating that new technology. To compete in this iterative form of warfighting, Washington needs to shift away from the fear of corruption and the reams of regulations designed to squash it. Policymakers must recognize a more pressing fear: that the United States is forced into a hot war with last-generation technology while its adversaries sprint ahead-a position it has not experienced since World War I.

Still, future militaries will need to incorporate technology without depending on it. A force multiplier is a high-priority target for an opponent, and militaries must be ready to lose those tools and keep fighting. Using technology as a crutch happens today-overdependence on signals intelligence at the expense of human intelligence contributed to a critical intelligence failure before the attacks on October 7.⁶² Against a highly capable adversary, however, the extent of the failure could be far worse. For example, militaries depending entirely on GPS for precision navigation and timing of weapons systems could find themselves toothless, should the GPS satellite cluster go dark. Quantum sensing might be a future alternative; in the meantime, the U.S. Naval Academy is teaching coping mechanisms like navigating by the stars.⁶³ A drone swarm accompanying a mobile attack squad can be a force multiplier, but if that swarm is taken down by an electromagnetic pulse or another form of electronic warfare, the squad must be able to fight on. If undersea cables are cut and war in space imperils satellite communications, militaries need a backup plan to continue to coordinate multidomain warfare.

While technology offers significant advantages, it also introduces new vulnerabilities, as adversaries are continuously innovating and developing countermeasures. The rise of AI-enabled decisionmaking, for instance, raises ethical concerns about the trend of human deference to AI recommendations. Moreover, the increasing role of the commercial sector in warfare is blurring the lines between combatants and noncombatants. Militaries must strike a delicate balance: using technology without becoming overly reliant on it and maintaining ethical safeguards and rigorous testing to keep technology safe. The side that strikes this balance will be best positioned to harness the full potential of technology in the ever-changing landscape of warfare.

Please consult the PDFfor references.

Emily Harding is director of the Intelligence, National Security, and Technology Program and vice president of the Defense and Security Department at the Center for International and Security Studies in Washington, D.C. Aosheng Pusztaszeri is a research assistant with the CSIS Intelligence, National Security, and Technology Program.