How quickly is China learning from the Ukraine war’s latest drone tactics?
Beijing’s effort is the most systematic of any outside power, but it faces limits learning from a war it is not fighting
But it usually took a year or more before those lessons showed up publicly in People’s Liberation Army (PLA) training, equipment or force structure, Lilly Min-Chen Lee and Kateryna Bondar wrote in a commentary published on October 7 by the Washington-based Centre for Strategic and International Studies (CSIS).
The lessons drawn by China go beyond immediate tactics. Chinese analysts have also looked at the economics of mass-produced drones, ways to shorten the kill chain, fibre-optic first-person-view (FPV) drones that are harder to jam, and multiple layers of defence against sea and aerial drones.
They have also studied how drone units should be organised, how industry can keep producing drones in large numbers and whether concepts such as large drone swarms work in practice.
To examine how those lessons were being absorbed, Lee, a non-resident fellow at Taiwan’s Research Institute for Democracy, Society and Emerging Technology, and Bondar, a senior fellow at CSIS, drew on more than 90 Chinese-language sources published between July 2022 and August 2026 in 45 outlets, ranging from the Central Military Commission’s newspaper to journals run by China’s missile, shipbuilding and aviation conglomerates.
They found that much of the Chinese research examined the war from the perspective of the side under attack. Designers of air and harbour defences treated Ukraine’s sea-drone strikes on Russian ships and ports and its airfield raids as previews of possible attacks on China.
The researchers found that defensive lessons were the quickest to show up in new equipment.
Other signs that the lessons were being put into practice included a paramilitary People’s Armed Police unit training with FPV drones; counter-drone missile-gun systems, lasers and high-power microwave weapons entering public display; and PLA exercises countering aerial and maritime drones.
But some lessons appeared in PLA training much faster than others.
The fastest case Lee and Bondar identified followed Ukraine’s Operation Spiderweb, in which drones launched from inside Russia struck strategic bomber bases on June 1, 2025.
About four and a half months later, a PLA brigade held an exercise defending against an attack team of more than 20 drone operators. Lee and Bondar linked the two, although the report on the drill by state broadcaster CCTV made no reference to Ukraine.
The earliest PLA Navy training exercise against sea drones identified by the two researchers came in August 2025, when unmanned boats featured as a harassment scenario in a missile-boat drill. That was 35 months after the first Ukrainian sea drone washed ashore near Russia’s Black Sea port of Sevastopol.
Dedicated live-fire training against uncrewed boats by an Eastern Theatre Command frigate flotilla followed in April 2026.
Moreover, Beijing is building institutions at home to absorb lessons from the conflict. Lee and Bondar noted new military universities and degree programmes covering unmanned-systems command and uncrewed equipment engineering.
On September 30, the PLA established the College of Unmanned and Intelligent Engineering within the PLA Army Engineering University.
Lee and Bondar wrote that the PLA was studying how drones could support offensive operations and how inexpensive uncrewed systems could impose disproportionate costs on concentrated forces.
But there is a limit to what China can learn from a war it is not fighting. Lee and Bondar said the harder problem was how the PLA could keep testing and adapting new ideas without troops coming under the pressure of combat.