China’s Ai Arsenal
China’s AI Arsenal
The PLA’s Tech Strategy Is Working
Sam Bresnick, Emelia S. Probasco, and Cole McFaul
March 2, 2026
Subscribe to unlock this feature or Sign in.
Print Subscribe to unlock this feature or Sign in.
Close
Article link: https://www.foreignaffairs.com/china/chinas-ai-arsenalhttps://www.foreignaffairs.com/china/chinas-ai-arsenal
This is a subscriber-only feature. Subscribe now or Sign in.
Chicago Cite not available at the moment
MLA Cite not available at the moment
APSA Cite not available at the moment
APA Cite not available at the moment
At China’s Victory Day parade in September 2025, it was not the marching troops or rolling tanks that made headlines, but the next-generation weapons systems on display. Uncrewed ground vehicles, underwater and aerial drones, and collaborative combat aircraft—autonomous jets that fly alongside piloted aircraft to aid missions—were presented as core components of China’s future fighting force. The exhibition of these systems sent a message about how the People’s Liberation Army (PLA) intends to leverage emerging technologies for battlefield advantage. Seen from Washington, the parade underscored Beijing’s ambition to erode the United States’ technological edge.
China has stated that its long-planned military modernization will unfold in three distinct but overlapping phases: mechanization, or the adoption of modern machinery and equipment; informatization, or the incorporation of advanced information technologies and cybernetworks to link military platforms and enable real-time information sharing; and intelligentization, or the application of artificial intelligence to automate operations and support decision-making. So far, China has made significant progress in realizing the first two aims. Its mechanization drive has provided the Chinese military with the ships, tanks, and aircraft it once lacked. Informatization has connected those platforms and sensors through data links and digital communications networks.
As part of our research at Georgetown University’s Center for Security and Emerging Technology, we examined thousands of publicly available PLA procurement requests published over the last three years. These documents reveal that China is urgently pushing the third phase of its modernization. The breadth of its efforts to integrate artificial intelligence into its military and the speed of its experimentation are striking. The PLA is prototyping AI capabilities that can pilot unmanned combat vehicles, detect and respond to cyberattacks, track seaborne vessels, and identify and strike targets on land, at sea, and in space. The Chinese military is also developing systems that ingest, analyze, and augment massive amounts of data to enhance tactical and strategic decision-making, as well as tools that create deepfake images and videos for disinformation campaigns.
In short, the PLA is fostering an ecosystem for rapid AI development that connects novel research with frontline operations. The United States, meanwhile, has declared the AI company Anthropic a supply chain risk, effectively barring a leader in frontier AI from supporting the U.S. government. The U.S. military still holds critical advantages in computing power, technical talent, and operational experience. But to stay ahead of Beijing, Washington will need to carefully shepherd its advantages, prototype with greater urgency, and, perhaps most important, scale the AI systems that give it a battlefield advantage.
SMART THINKING
Much as mechanization transformed warfare during World War II, Chinese strategists argue that AI will lead to a new revolution in military affairs. They believe that future wars will become “system of systems” confrontations in which militaries target the critical nodes of their adversaries’ interconnected systems. Victory will depend on the PLA’s ability to degrade or paralyze an enemy’s command-and-control centers, logistics hubs, and strike capabilities. AI will enable this vision of future conflict because it can help operators identify and target system-level vulnerabilities and improve the speed and efficacy of military decision-making. Accordingly, Beijing believes that whichever military better develops and adopts AI and other emerging technologies will gain a major advantage in future wars.
In some areas, the PLA’s AI ambitions have long been visible. Chinese leaders have stressed the importance of the intelligentization push, and state media outlets publish effusive updates on new uncrewed or autonomous air, ground, and undersea vehicles. Although these reports may oversell the capabilities of such systems, our analysis shows that the PLA is delivering. For instance, it is developing swarms of aerial drones that can identify, track, and coordinate attacks on an adversary. The Chinese military is also asking defense manufacturers and researchers to build a variety of robotics platforms, including hardware and software for robotic dogs and humanoid robots.
Procurement documents reveal Beijing’s growing interest in AI technologies that assist and accelerate decision-making. China’s political and military leaders do not trust the PLA’s chain of command and worry that it will be overmatched in a rapidly evolving conflict. AI decision-support systems can help them understand and anticipate an adversary’s movements, potentially allowing PLA leaders to overcome the disadvantage of their limited battlefield experience.
China is intent on developing AI to gain military advantage over the United States.
The Chinese military’s AI experimentation extends to cyber- and information operations. The PLA is developing AI tools to automate the detection of intrusions into its computer networks, enhance the resilience of its military communications, and augment its cyber-operations. In addition, China has placed a premium on developing cognitive warfare techniques. Several procurement documents request deepfake technologies; the military views these AI-generated images, videos, and audio as potent tools for influencing public opinion and manipulating adversaries’ perceptions and decision-making during conflicts. Meanwhile, PLA officers and soldiers are using AI systems to simulate virtual battlefields and model competitor behavior, which enhances their training for future conflicts.
The PLA is increasingly applying AI to diminish U.S. advantages in space and at sea. It is openly pursuing algorithms for satellite targeting as well as new antisatellite weapons, some of which involve small robots that can latch onto and disable an adversary’s space-based platforms. In the maritime domain, the Chinese military is experimenting with an array of autonomous underwater vehicles. Beijing has already deployed advanced sensors in the oceans and in space to map and monitor undersea activity, with the goal of eventually tracking U.S. submarines around the globe.
These projects, all part of Beijing’s intelligentization drive, represent a whole-of-force transformation rather than a series of isolated acquisition programs. China aims to build a military that leverages advanced technologies to learn, adapt, and make decisions quickly and more accurately across all warfighting domains.
China is not waiting for AI breakthroughs. Instead, it is experimenting with what is currently available and betting that incremental advancements will accrue over time. Many of the documents we reviewed feature short development timelines, which allow for fast-paced and relatively inexpensive experimentation across applications and domains. Beijing is also providing subsidies, tax incentives, and other advantages to encourage domestic technology companies to repurpose their products for defense applications.The integration of civilian products and defense applications allows the PLA to capitalize on China’s commercial technology sector and advanced industrial capabilities, where rapid iteration—the process of updating and refining products—and adaptation have propelled advances in such fields as smart manufacturing, robotics, and battery technologies. This approach could help funnel cutting-edge technologies to the military.
COPYCAT?
A number of the PLA’s ideas for AI adoption resemble major U.S. military programs. The Pentagon is pursuing the development of uncrewed and autonomous vehicles; computer vision models, which analyze terabytes of images and videos collected by satellites, drones, and ground cameras to identify and track objects; and predictive maintenance, which uses AI to anticipate necessary equipment repairs before malfunctions occur. Much as the U.S. Department of Defense plans to acquire thousands of low-cost, expendable drones through the Replicator Initiative, the PLA is requesting large quantities of small, cheap drones for use in future conflicts. We also reviewed requests for proposals that looked strikingly similar to the Pentagon’s Combined Joint All-Domain Command and Control initiative, which uses AI and computer networks to connect the command centers, troops, and weapons of different military branches across a modern battlefield.
Some of these similarities, including parallel experiments with targeting algorithms and cyber-operations, seem poised to lock Washington and Beijing into a rapid cycle of iterative technological changes. The outcome of the competition will thus depend on which force can develop and scale novel capabilities and operational improvements faster. Given how quickly many defense technologies evolve, technological advantages may be hard won but ultimately short-lived.
The PLA may use AI as an alternative to its weak, inexperienced officer corps.
Not all PLA approaches to AI-enabled warfare mimic those of the United States. Many American initiatives related to AI-assisted decision-making focus on planning and force management, such as the Defense Innovation Unit’s Thunderforge project for joint operational planning, Palantir’s Maven Smart System for targeting and battlefield awareness, or the Indo-Pacific Command’s Joint Fires Network for collaborative command and control. In addition to those capabilities, the PLA is developing AI systems to track international news, as well as identify foreign populations’ political views, predict social unrest, and manipulate adversaries’ perceptions, cognition, and decision-making.
How the two militaries balance human and AI decision-making may prove to be the deciding factor in the competition. Whereas the U.S. military requires that AI systems involve “appropriate levels of human judgment” by well-trained and experienced personnel, the PLA may be tempted to use AI decision-support systems as alternatives to its weak, inexperienced officer corps. But an overreliance on computer-generated analyses could lead to misinterpretation of military and diplomatic signalsand to unsound military decisions.
Some of the AI decision-making technologies requested in the PLA procurement documents we reviewed rely on open-source data. Militaries may be incentivized to manipulate the information environment—for example, by flooding social media with false signals or disrupting commercial satellite imagery providers—to supercharge information warfare. These attacks would aim to confuse, deceive, and degrade an adversary’s AI tools and could lead to inadvertent escalation.
COMPUTE TO COMPETE
To be sure, China’s path to intelligentization will not be smooth. The PLA still faces hurdles in integrating AI across its force and into its operations. As the war in Ukraine has demonstrated, it is one thing to develop technologies such as autonomous drones, but another thing entirely to deploy them effectively on contested battlefields. The ability to quickly train drone operators, develop attack plans, and prevent “drone fratricide,” or mistakenly destroying a drone deployed by the same side, matters just as much as the technical capabilities of the drones themselves. The speed of adoption will be a key dimension of the competition. Moreover, training AI systems for military use requires reams of data not readily available on the Internet, such as classified imagery of military platforms or the electromagnetic signatures of various radars and weapons. The PLA has limited combat experience and lacks many of these data sets.
Even if some AI systems fail, the PLA’s rapid prototyping and experimentation will accelerate its learning and improvement. China is positioning itself to quickly and effectively adopt and deploy operational military AI, thus keeping the gap between the U.S. and Chinese militaries narrow. China’s efforts should spur Washington to develop more agile processes for military experimentation, acquisition, and adoption. The defense acquisition process in the United States has long moved at a glacial pace. Congress and the Trump administration have begun taking action to reform it: the 2026 National Defense Authorization Act included a suite of provisions designed to reduce unnecessary delays, expand commercial purchasing pathways, and give defense contractors greater flexibility to pivot when new technologies emerge. The White House has issued a series of executive orders to further expedite acquisitions by reducing duplication and centralizing decision-making. Secretary of Defense Pete Hegseth has directed the Pentagon to take a “wartime approach to blockers” that delay the development and deployment of AI capabilities.
China has placed a premium on developing cognitive warfare techniques.
Acquisition reform alone, however, will be insufficient. There is an urgent need to better educate military operators on the strengths and potential shortfalls of AI, especially AI-assisted decision-making tools. Beyond education, operators need training and guidance on how to effectively use new AI systems, which will be a challenge considering how quickly these systems evolve. At the same time, important work still needs to be done on the systems’ reliability, transparency, and cybersecurity before commanders can trust and implement them consistently. These needs should drive the Pentagon to embrace closer relationships with frontier AI labs, the world’s leading experts on the technology, and contract not only for technology licenses but also field engineers and data scientists to help expedite safe and effective AI adoption. This is, at least in part, why the failure of negotiations with Anthropic is so concerning: ensuring U.S. national security will require bolstering such partnerships, rather than dramatic public spats.
Similarly, leaders across academia, industry, and government should prepare for a future of widespread AI-enabled deception that threatens civilians and militaries a. Washington should prioritize the development of standards and regulations to manage information competition while still promoting democratic norms of free expression. Developing tools for detecting and responding to deepfakes will be essential to ensure that military leaders have access to accurate information when making key decisions. This may require establishing new institutions for research, development, and education on information warfare. In the meantime, diplomatic exchanges and nongovernmental Track II discussions remain crucial channels for shaping norms around the responsible use of military AI.
Beijing’s long-promised third phase of military modernization is well underway, and the PLA is intent on developing AI to gain military advantage over the United States. Much of the Chinese military’s experimentation is unsurprising, but some proposed applications, particularly its complex AI decision-support systems and cognitive warfare tools, heighten the risk of miscalculation and destabilization. As China continues its efforts to modernize its military by adopting emerging technologies, the United States should also commit to rapid AI iteration, responsible experimentation, and smart commercial partnerships to maintain its edge.
Loading…
Please enable JavaScript for this site to function properly.
You are reading a free article
Subscribe to Foreign Affairs to get unlimited access.
- Paywall-free reading of new articles and over a century of archives
- Six issues a year in print and online, plus audio articles
- Unlock access to the Foreign Affairs app for reading on the go
Already a subscriber? Sign In
Sumber Artikel:
Foreignaffairs.com