How AI-enabled weapons, drones, and decision systems are reshaping global power—and putting warfare on autopilot.
One of the leading sectors in AI development is the defense sector. As geopolitical tensions increase globally and modern wars break out, AI is increasingly employed for military and defense purposes. In addition, AI technologies are accelerating in their development, including the race to AGI (Artificial General Intelligence), some referring to its achievement as the new atom bomb of warfare. This race takes place particularly between China and the US, who are both vying for AI dominance. Japanese companies are also increasingly investing in AI development in the defense sector, as tensions increase in the South Pacific between China, Japan, and Taiwan.
Many countries are increasing their defense spending because of the complex global security environment, much of which will go towards developing and producing these technologies. The International Institute for Strategic Studies notes that defense spending globally rose to $2.46 trillion USD in 2024, up from $2.24 trillion USD in 2023. In the United States, “Republicans are pushing for at least $100 billion in additional defense spending during the next decade.” As tariff wars escalate between the US and China, both sides have stated that they are “ready to fight” and are prepared for “any type of war”. US Defense Secretary Pete Hegseth said in a media briefing in Stuttgart, that the US is not likely to reduce its defence spending below 3% of GDP in the near future.
In response to increasing pressures from Russia, and wavering support from the United States, “European defence spending jumped by 11.7%”. Notably, Russia’s defense spending rose by 41.9%, indicating a full-blown war economy. On the border with Russia and with a long history of being threatened and harmed by Russian influence, Poland has stepped up their spending to become the 15th largest defence spender globally in 2024, compared to 20th in 2023.
In addition, the German defense budget increased by 23.2% in real terms, and long-standing policies against taking on government debt were lifted in 2025. On the 18th of March 2025, Germany voted officially to reform the “debt brake”, which prevents the German government from taking on debt. The primary purpose of this was to increase spending on defense in support of Ukraine, and to bolster Germany’s own defense sector. “It sends a very clear message to Europe that Germany is determined to invest massively in defense,” said Ursula von der Leyen, the President of the European Commission. France, plagued by high levels of government debt, plans to increase defense spending from 2.1% to 3%, with numerous political questions about where this money will come from.
China is maintaining their defense spending in turn, particularly with the goal of challenging the influence of the United States in Asia. EuroNews reports that for 2025, China will “keep its defence budget by 7.2% this year as part of its ongoing effort to develop a more expansive and advanced military.” As a result, Japan is also “striving hard to expedite its defense buildup to cope with rising military threats posed by three nuclear-armed neighbors, namely China, North Korea, and Russia.” The Japanese defense budget increased by 9.4% in 2024 among what Japanese military chiefs are calling “the most severe and complex security environment since the end of World War II.”
Some technologies that will be funded by these increases are already being used in the war between Ukraine and Russia, as well as by Israel in Gaza. Other technologies are also in development, although not yet implemented for technical or regulatory reasons. Major technologies already in use include drones, autonomous military robots, hypersonic missile technology, and more.
Here’s what you’ll find in this research:
- AI in Autonomous Weapons Systems: From swarming drone attacks to unmanned combat vehicles, AI is revolutionizing warfare with faster, smarter, and deadlier precision.
- AI in Hypersonic Weapons and Missile Defense: Tracking and intercepting missiles moving at five times the speed of sound? AI is making it possible, reshaping next-gen missile defense systems.
- Defense Cybersecurity & Electronic Warfare: AI-driven cyber defenses can detect, neutralize, and even predict cyber threats before they strike. Meanwhile, electronic warfare is going full AI, disrupting enemy communications and defense networks.
- Military Decision-Making & Logistics: AI isn’t just fighting wars—it’s learning how to run them. From real-time battlefield simulations to supply chain optimization, AI is making military logistics smarter, faster, and more autonomous.
- Venture Capital Investment in Defense AI Startups: AI in defense isn’t just a government game—private investors are flooding the sector. From autonomous robotics to AI-driven surveillance, VCs are betting big on AI-driven military tech.
- Insights from Defense Contractor Investor Calls: What are the biggest players in defense—Lockheed Martin, RTX (Raytheon), Northrop Grumman, and others—saying about AI? We analyze key investor calls to spot emerging trends and investment opportunities.
- Challenges and Barriers: AI-driven warfare sounds like sci-fi, but big challenges remain—from ethical concerns and AI bias to cybersecurity risks and global arms races.
So, what’s next? The rise of fully autonomous warfare, AI in strategic deterrence, and the impact of AI-driven military budgets on global security are all things to look out for going forward. The race for AI military dominance is just beginning, and countries that lag may lose out big on a geopolitical scale. For companies and investors looking to find the next hot technology, the outlook for the industry is strong, and the use of AI in the defense sector is only likely to expand.
AI in Autonomous Weapons Systems
Drones and Unmanned Aerial Vehicles (UAVs)
One of the key areas of development for AI in the defense sector is the increased use of drones and UAVs. A Deloitte survey found that in general, “81% of respondents [in the aerospace and defense industry] … are already using or plan to use artificial intelligence and machine learning technology”, and a large portion of this is the drone market. Deloitte reports that “some estimates valued the 2023 global military drone market as high as US$20.21 billion … and the pace for unmanned systems will likely remain in focus in 2025.”
As you can see in Figure 1 below, the drone market is projected to grow across all regions, particularly in the United States, Europe, and Asia.
Figure 1: Drone Life. Drone Market Growth by Region.
The war in Ukraine has become a testing ground for AI-driven drone warfare, showing how autonomous and semi-autonomous systems are changing modern combat. For example, new loitering munitions (like Iran’s Shahed-136 used by Russia) can autonomously track and strike targets with minimal human intervention. Ukraine has also used AI-assisted drones for targeted strikes. Other types of drones include Turkey’s Baykar TB2 drones, which use AI for navigation, object recognition and reconnaissance. This allows more precise targeting and battlefield intelligence.
You can see in Figure 2 how drones and other AI-enabled technologies analyse and label video or static imagery. These types of tools are being applied for target detection, battlefield analysis, and more.
Figure 2: US Government Accountability Office. Example of Labeled Imagery Data.
A major area of interest is the development of drone swarms. For instance, Ukraine has been experimenting with AI-assisted drone swarm technology, where multiple drones work together to overwhelm enemy defenses. EU and US companies are also developing similar tools, as is China. One example is China’s Zhu Hai Yun AI-powered drone carrier, an unmanned “mother ship”. This mothership can deploy and coordinate a fleet of smaller drones for reconnaissance or combat.
The strategic advantage of drone swarms lies in their ability to act as force multipliers. AI-powered coordination enables these swarms to adapt dynamically to threats, change formations mid-flight, and execute complex missions without continuous human control. These capabilities have significant implications for modern warfare, as swarms can be used for suppression of enemy air defenses (SEAD), electronic warfare, and even autonomous kamikaze-style attacks.
Defense spending on AI-enhanced drones has also skyrocketed, with companies like Anduril, Shield AI, and Palantir benefiting from increased contracts. Unlike past conflicts, the Ukraine war has seen heavy involvement from private AI firms, showing a new investment and supply model where startups play a bigger role in defense innovation.
As well as drones, UAVs are also being heavily invested in. Although, in many cases, both terms are used interchangeably, drones are typically smaller and more versatile than UAVs, DARPA has heavily invested in a number of UAV programs, including the ACE (Air Combat Evolution) program, the OFFSET (Offensive Swarm-Enabled Tactics) program, the GREMLINS program, and the development of PERDIX drones. The ACE project is targeted at improving AI dogfighting, while the OFFSET program focuses on drone swarming strategy. GREMLINS targets the development of reusable drones that can be sent out and recovered later, while PERDIX drones are focused on reconnaissance and electronic warfare.
China has also been aggressively developing AI-powered autonomous aircraft. This includes what are known as “Loyal Wingman UAVs” (which are also developed by the US). These are AI-driven UAVs that work alongside manned fighter jets. The combination of manned fighter jets and wingmen extends the range of jets, allows additional weapons to be carried, and allows the wingman to carry out autonomous side-missions. One example is the FH-97A drone (a Chinese equivalent of the US XQ-58A Valkyrie). This is designed to operate alongside the J-20 stealth fighter, using AI to carry out reconnaissance, electronic warfare, and strike missions.
Beyond combat applications, AI-driven UAVs are also being developed for logistics and battlefield resupply. The U.S. Army and NATO allies are experimenting with AI-controlled transport drones capable of delivering ammunition, medical supplies, and fuel to front-line troops in high-risk combat zones. These UAVs leverage real-time AI decision-making to adjust routes mid-flight, avoid enemy detection, and optimize payload delivery based on battlefield conditions.
Due to threats from China and North Korea, Japan has also increased spending and R&D on AI-powered drones, particularly for defense and monitoring in the Pacific, particularly near the Senkaku Islands. This includes developments on surveillance and reconnaissance drones, as well as AI-powered early warning systems.
Because Russia, China and North Korea are developing hypersonic missiles (which travel so fast, they are incredibly difficult to track and intercept) the US and Japan are also increasingly investing in hypersonic missile tracking systems. For example, in partnership with the US, Japan is working on Glide Phase Interceptor (GPI) technology, a next-generation missile defense system for countering hypersonic weapons. Companies like Raytheon, Lockheed Martin, and Northrop Grumman are heavily involved in these developments. This research examines these companies in more detail later.
Autonomous Ground & Naval Robotics
Alongside drones and UAVs for aerial missions, AI-enabled ground and naval robots are being developed for reconnaissance, logistics, and combat missions. Combining machine learning, sensor fusion, computer vision, and real-time battlefield data, AI-powered robots can take over tasks usually performed by soldiers, including scouting enemy positions, resupplying troops, and engaging hostile forces.
This reduces soldiers’ exposure to dangerous combat zones and potentially improves information about the battlefield and logistical information. These autonomous systems can scout enemy positions, deliver supplies, and even engage hostile forces. This significantly minimizes risk to human beings.
You can see in Figure 3 that over time, more and more countries have adopted the use of armed UAVs. Also note that there is a large gap between possession and use, indicating several challenges preventing the full implementation of these tools. Some of these challenges include regulatory, ethical, and development barriers, which will be examined later in this research.
Figure 3. Friedrich Naumann Foundation for Freedom. Countries with armed UAVs, trend over time.
In the United States, companies such as Boston Dynamics and Ghost Robotics are well-known for their military robots and are leading the development of AI-enabled autonomous quadrupeds. For example, the Boston Dynamics robot “Spot” is being developed for reconnaissance and perimeter security and may be equipped with weapons in the future. Ghost Robotics’ Vision 60 quadrupeds are also being used for similar tasks and include AI-powered navigation and threat detection tools.
In China and Japan, similar systems are being developed. China has invested heavily in AI-driven land and naval warfare systems, such as remote-controlled versions of the Type 59 tank and semi-autonomous tanks. Full autonomy and AI enablement would allow these tanks to be operated independently or with remote controls. China has also developed the JARI-USV, an AI-powered unmanned naval combat vessel. This is only one of the new waves of unmanned, stealth-optimized naval vessels that are typical of modern naval forces.
Japanese developments also include unmanned combat ground vehicles and defensive missile tracking systems for asymmetric warfare. Asymmetric warfare is when smaller forces use technology to counter larger adversaries (in this case, with a focus on China).
AI in Hypersonic Weapons and Missile Defense
Another key area of military technology development is the improvement of hypersonic missiles, and the defense systems capable of stopping them. Hypersonic missiles are so dangerous because they can travel at speeds exceeding Mach 5 (five times the speed of sound). This makes them extremely difficult to detect and intercept, as they can maneuver unpredictably and evade traditional missile defense systems. Most older-generation missile defense systems are based on intercepting ballistic missiles in mid-air, typically during their parabolic flight path. However, hypersonic missiles operate differently—they travel at lower altitudes and can change direction mid-flight, making them significantly harder to track and neutralize. The development of these types of missiles by China and Russia is particularly worrying for the United States.
In Figure 4 below, you can see the lower trajectory of the hypersonic missile, and how much closer to the target it is before it is detected by radar. This makes it much harder to prevent a strike from one of these missiles, when compared to older ballistic missile technologies.
Figure 4: The Economist. Missile trajectories.
To counter these hypersonic missiles, militaries are investing heavily in AI-powered missile defense. AI’s ability to process large amounts of data in real time can significantly improve tracking, predicting, and intercepting hypersonic threats. Traditional missile defense relies on human operators and pre-programmed interception strategies, but AI can analyze incoming threats and adjust interception calculations in real time. The US, China, and Japan are all actively developing AI-driven hypersonic defense technologies, which integrate advanced radar systems, space-based sensors, and machine-learning algorithms.
At the same time, AI is also being used to improve hypersonic missile technology itself. AI plays a crucial role in trajectory prediction, autonomous guidance, and adaptive flight control, enabling hypersonic missiles to dodge interceptors and change their flight paths mid-course. In addition, AI-powered simulations are helping engineers refine hypersonic glide vehicles (HGVs), improving their speed, range, and evasive capabilities.
Lockheed Martin is leading the way in AI-enhanced hypersonic glide vehicle development. They integrate AI into target tracking, flight stability, and countermeasure evasion. One example of their technology is the Hypersonic Air-breathing Weapon Concept (HAWC, pronounced “hawk”) program, developed with DARPA. The HAWC program aims to create and improve hypersonic air-launched cruise missile technology, with the first successful flight taking place in 2023.
China is also investing heavily in AI-assisted hypersonic maneuvering and missile tracking. Defense One describes China as wanting to “own the hypersonic domain”, with the Chinese viewing hypersonic missile use as a new domain of warfare necessitating the control of “near space” (i.e. low altitude). Some reports indicate that Chinese hypersonic systems use AI-powered onboard computing to optimize trajectories and evade interception dynamically. The DF-ZF hypersonic glide vehicle, a core component of China’s arsenal, is believed to incorporate AI to enhance its maneuverability and speed.
Japan is working closely with the US to integrate AI-driven hypersonic missile defense into its existing security framework. AI-powered radar and tracking systems, combined with US missile defense platforms like Aegis, are being upgraded to counter Chinese and North Korean hypersonic threats. To date, Japan has claimed four successful hypersonic missile tests, “designed to neutralize threats early and at long range”, and intended to contribute to Japan’s counter strike abilities.
AI in Defense Cybersecurity & Electronic Warfare
As seen in numerous attacks from China and Russia on EU and US companies and government departments, cyberwarfare and electronic warfare have also rapidly stepped up in recent years. This form of hybrid warfare is likely to increase, as AI enhances cyber threat detection, automates electronic warfare, and allows more sophisticated cyberattacks. The ability to defend military networks, counteract disinformation campaigns, and disrupt enemy communications is becoming a critical aspect of national security strategy.
AI allows real-time threat detection, autonomous responses, and predictive defense strategies. Traditional cybersecurity measures struggle against fast-evolving threats like AI-powered malware, deepfakes and disinformation, as well as zero-day vulnerabilities (vulnerabilities that are built into the software from the beginning). AI-enhanced defense systems can analyze large amounts of data, identify anomalies, and neutralize cyberattacks and threats before they escalate. AI is also being used to develop next-generation encryption methods to resist cyberattacks, including quantum computing-based threats.
Additionally, AI-driven security operations centers (SOCs) are being deployed by military agencies to enhance cyber defense readiness. These SOCs integrate AI-powered intrusion detection systems with autonomous threat response mechanisms, allowing military networks to react instantly to cyber intrusions. Machine learning algorithms continuously refine these defense protocols by learning from past cyberattacks, enhancing resilience against future threats. AI-enhanced deception techniques, such as honeypot networks, are also being used to mislead adversaries, diverting cyber attackers into controlled environments where their tactics can be studied and countered effectively.
The US is a leader in AI-driven cyber defense, with Northrop Grumman, Raytheon, and Lockheed Martin taking a large role in developing AI-enhanced cybersecurity solutions for the Pentagon. For example, Northrop Grumman has developed AI-powered cyber defense platforms that protect military command-and-control networks, as well as machine learning-driven intrusion detection systems (IDS) that analyze network traffic in real time, identifying and neutralizing threats before they infiltrate classified systems. The US is the current leader in investments in electronic warfare technology. Airforce Technology reports that “the US military accounted for the largest share of electronic warfare spending by a significant margin – 45% of global expenditure compared to Russia’s 14% and China’s 13%.”
As shown in Figure 5 below, global electronic warfare spending is dominated by the US, China, and Russia. Russia, China and India’s share of the electronic warfare market is also expected to increase over the coming years.
Figure 5. Airforce Technology. Global EW spending by country, 2021-2033
In China, tools are already being developed for “cyber intrusions”, i.e. to break into foreign networks to steal military and corporate secrets. Deepfake propaganda is also used to spread disinformation, and social engineering attacks are used over social media to carry out phishing and other scam attacks targeting government and military officials. Satellite hacking capabilities are also increasingly being developed by China, with satellite jamming and other tools used to take over or disable satellites.
Japan in turn is developing AI cybersecurity and electronic warfare defensive tools, to counter growing threats from China and North Korea in the Indo-Pacific. For example, the Active Cyber Defense (ACD) Project focuses more on protecting against cyber-attacks as a part of national defense strategy, than in the past. These defenses include rapid detection and response systems, radar jamming tools using AI-driven signal analysis, electronic systems to neutralize enemy communications and GPS, and a maritime self-defense force (JMSDF) to detect cyber threats attacking naval systems.
AI in Military Decision-Making & Logistics
An often under-considered area of AI development is the realm of decision-making and battlefield strategy in warfare. The development of AI-guided decision support tools is a significant breakthrough in military and defense strategy technology, yet this is an area that is often under-reported and under considered. As you can see in Figure 6, battlefield analysis and intelligence are still at low rates of maturity. This includes data fusion, target recognition, and integration across large operational theatres.
Figure 6: PWC. Military AI capability.
Military operations require immense strategic planning, real-time situational awareness, and highly efficient supply chains—all areas where AI can play a pivotal role. AI-supported decision systems can improve battlefield management, logistics optimization, and strategic command, reducing human cognitive overload while accelerating response times in combat scenarios.
Some of these systems are already being used in Ukraine. For example, “some domestically developed systems, such as Kropyva or GIS Arta, have been nicknamed ‘artillery Uber’ because they integrate data from radars, drone footage and other sources to compute and share information about Russian forces’ positions with Ukrainian artillery units in real time.”
Traditionally, military commanders have relied on manual intelligence gathering and experience-based decision-making, but AI can now process satellite imagery, battlefield communications, and sensor data in milliseconds. They can also identify enemy positions, analyze potential threats, and suggest the best countermeasures. They can also model different combat scenarios and predict outcomes based on real-time variables. Machine learning algorithms can also analyze past conflicts, adversary behavior, and terrain data to suggest optimal engagement strategies.
Additionally, AI-driven battle management software can dynamically adjust troop movements, supply chain logistics, and aerial support strategies in response to real-time battlefield changes. This adaptability ensures that forces remain agile and capable of countering evolving threats. Furthermore, AI-assisted combat drones and autonomous vehicles are being integrated into battlefield decision-making frameworks, reducing risks to human soldiers and enhancing reconnaissance operations.
These types of systems are called “Decision Support Systems” (DSS) and some of the delay or hesitance behind their use could be because they are being widely analyzed for their compliance with international human rights laws. Major concerns have been raised by organizations like the Red Cross, specifically with regard to the “potential erosion of human judgement” with the use of these systems.
In Figure 7 below, you can see the points at which AI systems engage with and add to human decision-making processes, such as target system analysis, identification of targets, mapping of capabilities, force planning, and performance measures.
Figure 7. P.J.M. Kerbusch. Roles of AI and Simulation for Military Decision Making.
Despite concerns, these technologies continue to be developed. The US is integrating AI into battlefield management systems, with Raytheon Technologies (RTX) leading efforts to develop AI-driven real-time combat simulations. This includes AI-enhanced wargaming simulations like the Rapid Campaign Analysis and Demonstration Environment (a campaign-level modeling and simulation capability) that allow military strategists to run real-time combat scenarios to test different engagement strategies. Raytheon is also developing AI-powered battlefield situational awareness tools that process drone feeds, satellite imagery, and sensor networks to detect enemy forces and predict their movements, as well as AI-assisted tactical decision-making systems that suggest optimal troop and resource deployment strategies in real time.
In addition to battlefield intelligence, AI is being leveraged for automated threat assessment and countermeasure recommendations. AI models trained on vast datasets of historical combat engagements can suggest optimal troop formations, identify weaknesses in enemy defenses, and even simulate adversary decision-making patterns. This level of insight enables commanders to act with greater precision and confidence.
China has also been developing AI-driven war-gaming simulations and autonomous command systems to enhance its military decision-making processes. The People’s Liberation Army (PLA) has invested heavily in AI to develop AI-powered war simulations that model different battlefield scenarios and predict enemy maneuvers based on real-time data. China’s “Intelligentized Warfare” strategy focuses on using AI for autonomous decision-making, potentially allowing AI to coordinate and execute military operations with minimal human oversight.
Japan has been focusing on AI for military logistics and operational efficiency, ensuring rapid deployment and optimal resource allocation for its Self-Defense Forces (JSDF). This includes AI-powered predictive logistics, troop readiness assessment, and inventory management. Machine learning algorithms analyze supply chain demands, transportation logistics, and battlefield conditions to ensure efficient resupply operations. AI systems can also monitor soldier health, equipment status, and unit availability to maintain optimal operational efficiency. AI can also automate munitions and equipment tracking, ensuring critical supplies are available exactly when needed.
Venture Capital Investment in Defense AI Startups
Numerous defense startups have benefited from the wave of interest in AI-enabled military technology, with some companies benefiting more than others. J.P. Morgan notes that “the number of startups working on defense-oriented technologies is rising, backed by record venture investment.”
In Figure 8 below, you can see the types of technology that are invested in, with advanced computer and software technologies leading the way. This category includes defense startups providing intelligent data-management software, data resilience and multi cloud networking, as well as AI-related data processing. Sensing, connectivity and security come in second, with autonomous systems quickly catching up. For autonomous systems in particular, J.P. Morgan describes these technologies as a “rising priority” for the Defense Innovation Unit of the Department of Defense. For example, the systems. This sector reached a new high in U.S. venture capital funding in 2024, with the AI surge driving a range of use cases, including automated industrial and logistics robots, and unmanned aerial and field intelligence systems.
Figure 8. J.P. Morgan. Cumulative Invested Capital (from 2014) for Select U.S. Defense Tech Categories
Anduril Industries, for example, has been a major beneficiary of this focus on AI defense tech. Anduril focuses on AI-driven drones, such as Bolt, Barracuda, Fury, Roadrunner and Ghost, a selection of autonomous air systems that can work together in tandem. Their work also extends into electronic warfare, where AI plays a role in enhancing defense systems and improving the strategic effectiveness of their defense technologies, using Sentry towers and other connected smart technologies.
Similarly, Shield AI has carved out a niche in autonomous piloting and drones, providing their capabilities to military forces. Their small-unmanned aircraft system (sUAS) Nova was the first to be used in US military history. Their autonomous drones are designed to operate in complex environments, offering tactical advantages in defense and surveillance missions. This innovation highlights the growing role of AI in making defense systems more independent and responsive to emerging threats.
Scale AI, another prominent player, makes artificial intelligence and machine learning solutions for intelligence, surveillance, hardware platform modernization, and autonomous vehicles. They are also working on “integrating artificial intelligence into military planning and wargaming” as part of the Thunderforge initiative, in partnership with Anduril, Microsoft, and the US government.
Additionally, newer startups such as Helsing AI in Europe and Singapore-based ST Engineering have begun pioneering AI-driven military software solutions that combine intelligence processing, real-time data analytics, and autonomous decision-making capabilities. These emerging players aim to enhance battlefield intelligence, automate target acquisition, and streamline military logistics with AI-powered systems. With governments actively seeking AI innovations that improve decision-making, defense tech startups continue to gain traction as key contributors to national security.
As defense contractors look to stay ahead of technological trends, these startups are receiving substantial venture capital investment. This funding not only helps accelerate their development but also positions them to be key players in the future of AI-driven defense systems. Their innovations are transforming military strategies and providing more sophisticated, automated solutions for national security challenges.
Insights from Defense Contractor Earnings Calls
Large defense companies such as Lockheed Martin, Northrop Grumman, and Raytheon, have also revealed additional insights through their earnings calls. As you can see in Figure 9 below, these defense companies are some of the biggest in the US and have large government contracts that help them to develop additional technologies in support of US national security.
Figure 9. Statista. America’s Biggest Defense Contractors.
Lockheed Martin’s Q4 2024 earnings call highlighted several advancements, including the integration of AI in hypersonic weapons systems like the ARRW (Air Launch Rapid Response Weapon) and defense solutions such as the Aegis Guam system. They emphasized the importance of partnerships with tech companies like NVIDIA and IBM for AI, large language models, and cloud solutions. The company is also leveraging AI for mission-critical applications in missile defense, aiming to stay ahead in the evolving defense landscape through digital innovation and military tech integration.
Furthermore, Lockheed Martin is developing AI-driven swarm drone technology that enhances battlefield coordination. These swarming drones are designed to operate as coordinated units, leveraging machine learning to adapt in real time. This technology could redefine air combat strategies, allowing multiple autonomous units to execute surveillance, attack, or reconnaissance missions with unprecedented efficiency.
Raytheon is also advancing its AI-driven capabilities, with notable developments in areas such as product testing, operational efficiency, and defense technologies. They also note that they “also expect continued strength in international demand for many of these products with key partners and allies in NATO, in the Indo-Pacific committed to increasing defense spending.” Raytheon continues to rapidly integrate AI across multiple sectors, from improving avionics software testing to enhancing the development of autonomous and AI-powered defense systems. They also plan to continue to invest in fuel efficiency, resilient networks, directed energy, autonomy, AI and advanced materials.
Raytheon is additionally focusing on AI-enhanced missile defense, using predictive analytics to counter hypersonic threats. The company has been developing AI-powered tracking systems that identify and intercept enemy missile systems more efficiently, reducing response times and increasing interception accuracy. AI is also being integrated into Raytheon’s cyber defense division, bolstering threat detection mechanisms in military communication networks.
BAE Systems is another large player that has been using AI to enhance its technological capabilities, particularly in key areas including “electronic warfare, autonomy, laser-guided weapons, uncrewed air systems (UAS), synthetic training, electrification applications and space solutions”. They aim to improve operational efficiency and capabilities across land, air, sea, and space domains, focusing on next-generation defense technologies such as “precision weaponry, space resilience, hyper-velocity projectiles, autonomous platforms and the development of multi-domain capabilities”.
Based on Northrop Grumman’s recent earnings call, they also highlighted significant advancements in defense capabilities, including AI-powered satellite defense, missile detection, and interceptors for hypersonic weapons. Their investments in AI and automation are central to enhancing operational capabilities and delivering next-generation solutions for defense.
Northrop Grumman has also been pioneering AI-driven satellite defense systems that enhance space-based surveillance and threat detection. These AI-powered systems are designed to detect and neutralize potential threats in space, marking a crucial step in securing satellite communications and orbital assets against adversarial threats.
Challenges and Barriers
While the developments in AI-enabled defense technologies seem rapid, there are also a lot of challenges and barriers that companies face in this sector.
One of the most complex and serious challenges is to consider the ethical questions that lie behind AI-driven lethal decision-making. Fully autonomous AI could make life-or-death decisions without human oversight, raising concerns about accountability and unintended escalation. Questions of legal responsibility also become complex: if an AI system kills civilians by mistake, whose fault is it? The military? The AI developers? The government? Existing laws and human rights conventions around war were written during a time in which AI-enabled warfare did not exist. One major concern is that AI tools could violate current international human rights laws. In addition, flawed algorithms could lead to bad decisions with serious consequences, including the significant loss of human life.
Furthermore, beyond ethical and legal concerns, AI-driven decision-making introduces a fundamental shift in warfare. Unlike human decision-makers, AI systems operate based on predictive models, pattern recognition, and probabilistic reasoning, which may lack the nuance and contextual understanding required in complex, high-stakes combat situations. This shift raises the risk of miscalculations that could escalate conflicts, particularly if multiple adversaries deploy AI-driven military strategies that interact unpredictably. The automation of warfare could reduce human hesitation in launching pre-emptive strikes, increasing the likelihood of unintended conflicts.
In Figure 7 below, you can see some of the Department of Defense’s expected uses for AI, and the possible risk to human life. Some of the potential uses they anticipate include “analyzing intelligence, surveillance, and reconnaissance sources; fusing data to provide a common operating picture on the battlefield; supporting semi-autonomous and autonomous vehicles; and operating lethal autonomous weapon systems.” These uses are anticipated as much riskier. For example, an AI model could fail and target the wrong person.
In addition, there are non-warfighting uses for AI, which are labelled as support and business operations. These include, for instance, “resolving unmatched financial transactions, predicting maintenance needs, vetting security clearances, and analyzing warfighter health screenings.”
Figure 7: US Government Accountability Office. Uses of Artificial Intelligence (AI) at the Department of Defense.
Another issue is the regulatory landscape, which remains unclear. There is an ongoing push to ban or regulate fully autonomous weapons, but no laws or regulations on this issue currently exist. So far, the primary preventative mechanism has been a “human in the loop” approach, in which a human decision-maker must still sign off on all decisions made by AI tools in warfare. Nonetheless, in Israel’s war in Gaza, decisions were made by autonomous systems that were only superficially checked by humans, leading to the killing of an innocent person. This happened because an AI-assisted Israeli airstrike misidentified a target, which was not picked up on by human actors checking the AI’s decision.
Additionally, the fragmented global approach to AI defense regulations adds another layer of complexity. While the U.S., NATO allies, and the European Union are working on AI safety frameworks, other countries, such as China and Russia, are moving forward with AI weaponization programs that do not currently adhere nor will they likely adhere in the future to the same ethical guidelines. This regulatory asymmetry creates the possibility of a technological gap, where nations that impose stricter AI governance may find themselves at a strategic disadvantage against adversaries who prioritize rapid deployment over responsible development. Diplomatic efforts to establish global AI arms control agreements remain in their infancy, and the enforcement of any international regulations will be challenging.
Regulation becomes particularly difficult, as AI technologies develop and pivot in sometimes unpredictable directions, or with timelines that are unexpected. Shashank Joshi, Defence editor at The Economist notes on the China Talk podcast: “Where are the drone swarms we were promised?”, referring to the widespread discussion of possible technologies that have not been put into action as quickly as some believed. Regulating this rapidly-developing field is challenging.
In addition, significant risks can arise from the AI-enablement of military technologies, and their increased vulnerability to hacking and cyberattacks. In 2024 and 2025, it has also become clear that AI-powered cyber tools are already being used to (effectively) spread misinformation and deepfakes, influencing elections and destabilizing societies.
Furthermore, China, Russia, and the US are already deploying AI-driven cyber tools to steal military secrets and critical infrastructure data. Hacking tools will be able to be used increasingly autonomously, such as to find and exploit vulnerabilities. This will create a new world of cyber warfare without human oversight. When these tools are used to exploit each other, it’s not hard to imagine that an AI-enabled military technology could be hacked or compromised by a malicious state or non-state actor.
The risks of AI hacking extend beyond traditional cyberwarfare. Military AI models rely on massive datasets to function effectively, and adversaries could use data poisoning techniques to manipulate these datasets, leading AI systems to make flawed strategic decisions. This presents an unprecedented security threat, as a compromised AI could inadvertently provide false intelligence, misdirect autonomous units, or even disable defense infrastructure. Additionally, AI-enabled electronic warfare could allow state and non-state actors to disrupt enemy communications, rendering entire military operations ineffective.
Future Outlook
One of the major issues to consider in the coming years is how AI will shape the balance of power between global military superpowers. The race for AI military supremacy, its impact on next-generation warfare, and the consequences of AI-driven defense spending will define the future of security and geopolitical stability.
The competition for AI dominance in defense is intensifying, particularly among the United States, and China. Japan and other countries are racing to develop technologies to defend themselves against larger nations, and the possibility that they will have to engage in asymmetric warfare. This race for AI military supremacy will define the next era of defense technology.
This technological arms race also introduces economic challenges, as governments and private companies must decide how much to invest in AI-driven defense relative to other pressing priorities. While AI-enabled military innovations promise significant strategic advantages, excessive spending could divert funds from other critical areas such as infrastructure, healthcare, and education. Defense budgets will need to balance immediate security needs with long-term economic sustainability.
As governments allocate billions of dollars toward AI-driven military applications, AI defense spending is set to reshape global security in multiple ways, including massive increases in defense budgets, AI-driven arms race and escalation risk, and several ethical challenges. The next decade will determine whether AI serves as a stabilizing force or a destabilizing arms race. While AI-driven systems enhance military efficiency, speed, and effectiveness, they also raise serious concerns.
For venture capitalists and defense companies, the opportunities in AI-driven military logistics, cybersecurity, and autonomous warfare systems are immense. However, long-term AI governance, regulation, and ethical safeguards will be necessary to ensure AI remains a tool for security, not unchecked destruction.
Ultimately, the trajectory of AI in defense will be shaped by a combination of technological innovation, regulatory decisions, and geopolitical realities. If the international community fails to establish meaningful AI governance structures, the risk of unintended escalation and destabilization will increase. However, if AI is integrated responsibly—with safeguards that ensure ethical decision-making and minimize risks—its potential to enhance security and reduce human casualties could be realized. The next decade will be critical in determining which path the world takes.
Finally…
AI is revolutionizing defense across multiple domains, enhancing cybersecurity, electronic warfare, military decision-making, and logistics. It enables faster threat detection, more sophisticated cyberattacks, autonomous responses, and real-time strategic analysis. Countries like the US, China, Russia, and Japan are developing AI-driven technologies for both offense and defense, ranging from autonomous drones, hacking tools and deepfakes, advanced encryption, AI-equipped hypersonic missiles, and radar jamming technologies.
Venture capital is driving innovation in AI-powered defense technologies, with startups like Anduril Industries, Shield AI, and Scale AI developing autonomous drones, UAVs, and real-time geospatial intelligence. Major defense contractors such as Lockheed Martin, Raytheon, and Northrop Grumman are also integrating AI into their systems, focusing on areas like missile development, missile defense, cybersecurity, and advanced logistics. Together, these players—both startups and established contractors—are rapidly advancing military technologies.
However, this rapid evolution raises significant ethical and regulatory concerns, particularly regarding lethal autonomous decision-making. While the opportunities for AI in defense are significant, challenges include ensuring AI’s ethical use, addressing vulnerabilities, and managing the geopolitical implications of AI-driven warfare.
