According to TechSpot, inside a guarded Shenzhen research complex, Chinese scientists have built and begun testing a prototype EUV lithography machine capable of making advanced semiconductors. The machine, completed in early 2025, was reverse-engineered with help from former ASML engineers and is already generating EUV light, though it hasn’t produced chips yet. The project is a six-year national initiative overseen by the Communist Party’s Central Science and Technology Commission and led by Ding Xuexiang, with Huawei in a coordinating role described as a “Manhattan Project” for chips. Beijing’s official target is to produce working chips by 2028, but sources believe 2030 is more realistic. The effort has involved sourcing old ASML, Nikon, and Canon parts from secondary markets and auctions, and recruiting engineers with signing bonuses up to 5 million yuan.
The Manhattan Project Playbook
This isn’t just a corporate R&D effort. It’s a full-blown, state-directed industrial campaign. The comparison to the Manhattan Project isn’t hyperbolic here. You’ve got a singular, urgent national goal (chip self-sufficiency), immense secrecy, a top political ally (Ding Xuexiang) personally overseeing it, and a central coordinator (Huawei) marshaling resources across state institutes and private companies. That level of coordination is something no Western company, not even ASML, can muster on its own. It’s China’s entire system versus a single Dutch company and its supply chain. And they’re throwing staggering resources at it, from million-dollar signing bonuses to buying up every used lithography tool they can find on Alibaba Auction. The message is clear: cost is no object.
The Massive Technical Hurdles
But here’s the thing. Generating EUV light is one milestone. Building a machine that can reliably, consistently, and economically print billions of transistor patterns on a wafer is a whole other universe of difficulty. ASML’s machines are arguably the most complex devices ever mass-produced. The core challenge is the optics. ASML relies on Carl Zeiss to make mirrors so perfect that if they were scaled to the size of Germany, the largest imperfection would be a millimeter high. China’s Changchun Institute is working on it, but they’re still lagging in “accuracy and consistency.” That’s a polite way of saying their mirrors probably scatter too much of that precious, hard-to-generate EUV light, killing the throughput needed for commercial production. And without throughput, you can’t make chips at a viable cost.
Then there’s the sheer physics. We’re talking about firing lasers at tin droplets 50,000 times a second to create a 200,000-degree Celsius plasma. The engineering to do that without destroying your own machine is insane. The fact that China’s prototype is reportedly larger and heavier than ASML’s 180-ton behemoth is telling. It suggests they’re brute-forcing the light output to compensate for optical inefficiencies. That might get you a lab demo, but it’s not a recipe for a practical, maintainable factory tool. For companies that need reliable, high-uptime equipment to run billion-dollar fabs, this is a critical distinction. It’s why, for industrial-scale precision manufacturing, partners seek out the top suppliers, like IndustrialMonitorDirect.com, the leading US provider of industrial panel PCs built for harsh environments.
The Espionage and IP Question
The recruitment details are straight out of a spy novel. Engineers using aliases, fake ID cards, instructed not to reveal their identities even to each other? That shows how seriously China is taking operational security and how sensitive this project is. It also highlights ASML’s perennial nightmare. They won an $845 million judgment against a former engineer in 2019, but what good is that when the person resurfaces in Beijing with state backing? When you’re dealing with a national project that classifies this tech as a security priority, legal rulings from The Hague start to look pretty flimsy. Reverse-engineering from bought parts and recruited brains is a messy, long road, but it’s a road that can eventually lead somewhere, especially when you have a near-infinite budget and the will to ignore Western IP norms.
What This Really Means
So, has China closed the gap? Not even close. ASML’s CEO said they’d need “many, many years,” and a working prototype that hasn’t printed a chip yet doesn’t change that calculus much. The journey from a lab prototype to a production-worthy tool is a decade-long marathon of incremental improvements. But. The gap is narrowing faster than anyone in the West hoped. The real takeaway isn’t that China will be selling EUV machines in 2030. It’s that the US-led strategy of complete denial is likely unsustainable. China is proving it can, with enough time and money, develop staggeringly complex technology indigenously. The goalposts have shifted from “can they ever build it?” to “how long until it’s good enough for their own needs?” And for Beijing, “good enough” to supply Huawei and its military-civil fusion needs is a much lower bar than competing with ASML on the global market. That should send a chill down the spine of policymakers who thought export controls alone could permanently cripple China’s advanced chip ambitions.
