What are Beijing’s strategic hi-tech priorities? An innovative study goes a long way to decipher them, revealing a set of structural gaps in EU law that leaves Europe vulnerable.
A new study from the NATO Defense College offers the most data-driven picture yet of how China’s People’s Liberation Army (PLA) is weaving artificial intelligence into its war-fighting doctrine. The findings are sobering.
The study, Decoding the Dragon: A Language-Based Analysis of China’s Vision of AI Warfare, came out recently by NATO Defense College. The authors are prominent experts; Anselm Küsters is scientific director for the Nordic countries and head of the Department of Digitalisation and AI at the Centre for European Policy; Marius Strubenhoff works for the same think-tank as an independent expert on foreign and security policy.
Mr Küsters and Mr Strubenhoff applied large language models to a corpus of over 171,000 Chinese documents spanning 2000 to 2025. The result is a rare empirical window into Beijing’s military AI ambitions. It also carries direct implications for EU security policy — and, by extension, for the bloc’s legislation.
The study’s most consequential finding concerns priorities. “Near-term PLA investments appear geared towards achieving a battlespace transparency that supports both information dominance and precision strikes,” the authors write. The goal is to see everything, and to use that knowledge to hit targets with pinpoint accuracy before the other side can react.
Surveillance first, autonomy later
Surveillance-related terms dominated the corpus by a wide margin. Computer vision accounted for 49.4 per cent of the subfield distribution, followed by robotics at 12.5 per cent and natural language processing at 6.2 per cent. AI safety, by contrast, accounted for under two per cent.
This is more than a defensive posture. Under China’s dissipative warfare doctrine, surveillance is an offensive enabler. As the study explains, “the doctrine involves maintaining order within one’s own system while deliberately creating disorder within an adversary’s political, economic and military networks”.
This pattern suggests that ‘meaningful human control’ functioned primarily as diplomatic vocabulary during peak international scrutiny but has been progressively de-emphasised. — Anselm Küsters and Marius Strubenhoff, Centre for European Policy
It also “dissolves the differentiation between the front and the home front, envisioning that the latter will become a prominent theatre in any future war”. European cities, ports, and data networks are, in that framing, fair game.
The practical implication is clear. “The dominance of surveillance applications suggests that near-term crisis instability will stem from information asymmetries rather than from rogue autonomous weapons,” the authors conclude. Europe’s immediate exposure is not a swarm of killer robots. It is the gradual, AI-accelerated erosion of its situational awareness, and the compression of its decision time.
The norm gap
Translated into human English: in a crisis, the real danger is not that a robot weapon will go rogue and start firing on its own. It is that China will simply know far more about what is happening on the battlefield than its adversary does. It may then use that advantage to act faster and strike first.
The study’s most politically significant finding concerns China’s relationship with international norms. Beijing’s governance documents emphasise human control and strategic stability. Its state media tells a different story. References to ‘meaningful human control’ in the People’s Daily peaked in 2021—precisely when UN deliberations on lethal autonomous weapons were at their most intense—and have since declined sharply.
“This pattern suggests that ‘meaningful human control’ functioned primarily as diplomatic vocabulary during peak international scrutiny but has been progressively de-emphasised as China’s autonomous systems capabilities matured and as Western regulatory momentum stalled,” the authors write. References to full autonomy have grown in lockstep.
The language of Chinese procurement reinforces the point. “References to ‘autonomous weapons’ are often displaced by euphemistic labels such as ‘autonomous systems’ or ‘unmanned platforms’,” the study notes. The number of military AI publications grew steadily, from 52 papers in 2016 to 181 in 2024. “The combination of positive state narratives and the decline in ‘meaningful human control’ rhetoric after 2021 signals China’s growing confidence and waning commitment to international norms,” Messrs Küsters and Strubenhoff conclude.
What Brussels has, and has not
Europe’s response to this trajectory runs through a legal framework built largely for civilian purposes. Brussels’ flagship law on the subject, the EU AI Act, explicitly excludes AI systems used exclusively for military, defence, or national-security purposes from its scope. That carve-out leaves a governance vacuum precisely where the NATO study sees China moving fastest.
The Dual-Use Export Control Regulation provides some leverage over outward flows of sensitive AI chips and surveillance software, including a catch-all licence requirement for cyber-surveillance items linked to human-rights risks. But it is export-centric and reactive. It does not yet cover foundation models, reinforcement-learning environments, or the synthetic datasets on which PLA intelligentised warfare depends.
Near-term PLA investments appear geared towards achieving a battlespace transparency that supports both information dominance and precision strikes. — Anselm Küsters, Marius Strubenhoff
The GDPR and the Law Enforcement Directive govern biometric and personal data processing. They are strong on individual rights. They are not designed to counter adversarial exploitation of European data streams by a foreign military. The European Parliament has passed resolutions in 2018 and 2021 urging a ban on lethal autonomous weapons that lack meaningful human control. Those resolutions remain politically significant as well as legally non-binding, i.e., next to useless.
The gap between China’s operational trajectory and Europe’s regulatory posture is, in short, structural. Brussels has built a civilian AI governance architecture. Beijing is building a military one. The two are not yet in conversation.
What Europe can do
The study addresses its recommendations to member states rather than to the EU as an institution, but the implications for Brussels are direct. On resilience, the authors argue that “societal resilience and cyber hardening are assets for deterrence in an information-domain contest, not merely defensive measures”.
That reframing matters for EU law. The Network and Information Security Directive (NIS2) and the Critical Entities Resilience Directive were conceived as civilian instruments. Treating them as deterrence assets requires a shift in both political framing and operational scope.
On countermeasures, the study identifies three investment priorities. “NATO members should invest in adversarial robustness research to develop AI systems that are resilient to data poisoning, model extraction and attacks that can blind, misdirect or compromise military decision-making systems,” the authors write.
They also call for investment in counter-ISR technologies: adaptive camouflage, decoy generation, and emissions management. It requires human-machine teaming architectures; that could demonstrate how “Human-in-the-loop systems can match or surpass autonomous systems in decision speed and accuracy, countering the narrative that human control requires accepting operational inferiority.” Meaning, if we have lost you along the way, that people still matter.
Closing the gap
On verification, the authors are unambiguous: “Effective arms control requires technical measures that are both intrusive and analogous to the inspection protocols of the Chemical Weapons Convention.” That means shared documentation standards, software bills of materials, and cryptographic on-chip governance mechanisms. None of this currently features in EU law.
As to binding commitments, the study calls for treaty obligations that “target autonomous target selection and ‘non-reconstructible’ black box military AI systems, whose battlefield actions cannot be audited for compliance with international humanitarian law”. International law requires that every decision to use lethal force can be reviewed after the fact, to check whether the rules of war were followed.
The doctrine involves maintaining order within one’s own system while deliberately creating disorder within an adversary’s political, economic and military networks. — Anselm Küsters, Marius Strubenhoff
However, some military AI systems make targeting decisions in ways that cannot be traced or explained, even by their own designers. The proposal is to ban, by international treaty, any weapon system that selects its own targets autonomously and whose decision-making is so opaque that no one can ever establish whether it broke the law.
The EU AI Act prohibits certain black-box applications in civilian contexts. No equivalent prohibition exists for military systems procured by member states. Needless to say that if it does, one day, it could severely impact most of future military AI uses.
The legislative agenda
Translating the study’s recommendations into EU law would require action on several fronts. A dedicated EU Defence AI Instrument—either a standalone regulation or a chapter within the forthcoming Defence Industrial Programme—could mandate audit-logging, adversarial testing, and conformity assessments for autonomous targeting systems, irrespective of the AI Act’s military carve-out.
Amending Annex I of the Dual-Use Regulation to cover high-performance AI accelerators and large language models above defined compute thresholds would extend export controls to the tools China’s military research most depends on. A Strategic Data Spaces regime, classifying high-resolution satellite and critical-infrastructure sensor data, would reduce Europe’s exposure to the surveillance-led kill chains the study identifies as Beijing’s near-term priority.
Most ambitiously, a European strategic autonomy framework (a new treaty instrument establishing common operational command and a joint defence AI budget) would require unanimous member-state agreement. We know how well that usually goes.
Messrs Küsters and Strubenhoff are careful not to overstate China’s current capabilities. China, they note, is not quite on the verge of autonomous military dominance. But as the dissipative warfare doctrine makes plain, the trajectory is clear: it wants to get there fast. And Europe’s legislative gap is real; and its home front is already part of the battlefield.
