China targets 70% domestic wafer use

Silicon wafers are the polished discs every chip starts on. No wafer, no logic chip, no memory chip, no AI accelerator. That is why this new Chinese target matters more than it sounds: Beijing wants more than 70% of the silicon wafers used by its domestic chipmakers to come from local suppliers by the end of 2026, and the push is being treated inside the industry like an unwritten rule — especially for 12-inch wafers, the standard size for modern chip production. ### Why are wafers such a big deal? A wafer is the base layer chips get built on. Think of it like the slab every circuit has to be carved into. The hard part is not making silicon in the abstract. The hard part is making extremely pure, extremely flat, defect-light wafers at industrial scale, then doing it consistently enough that big fabs trust them for advanced production. That is why this part of the chip stack stayed concentrated in Japan, Taiwan, Germany, and South Korea for so long. ### What exactly is China trying to do? The new goal is narrower than China’s older “70% semiconductor self-sufficiency” slogan. That older target covered the whole stack and was widely missed. This one is about one material input — silicon wafers — and mostly about the 12-inch segment, where the strategic value is highest because those wafers feed logic, memory, and image-sensor lines. That narrower scope is why people in the industry seem to think it is actually reachable. ### Why 12-inch wafers? Because 12-inch, or 300mm, wafers are the workhorse for mainstream advanced manufacturing. China is already much closer to self-sufficiency in 8-inch wafers, which are common in mature-node chips and power devices. The gap is the 12-inch market. That is where foreign suppliers still had room — and where Beijing now seems determined to squeeze that room down fast. ### Who is doing the heavy lifting? Xi’an Eswin Material Technology is the name that keeps coming up. The company is expanding in Xi’an and Wuhan and is targeting about 1.2 million 12-inch wafers per month in capacity by 2026. If that ramp lands, Eswin alone could cover roughly 40% of China’s domestic 12-inch wafer demand and push its global capacity share above 10%. That is not a side story — that is the industrial muscle behind the headline. ### Is this just politics, or is demand real? Both. Beijing wants to harden the supply chain against export controls and external chokepoints. But the market pull is real too. Global silicon wafer shipments rose 13.1% year over year in the first quarter of 2026, helped by AI-driven demand and a broader recovery in chip materials. So China is not localizing into a shrinking market. It is trying to grab more of a growing one. ### What does this threaten for foreign suppliers? Mostly share, not immediate collapse. Global leaders still dominate the top end of wafers, and qualification cycles in semiconductors are slow because nobody wants to risk yield problems. But if Chinese fabs are nudged to buy local wherever performance is “good enough,” foreign wafer makers could get pushed into a smaller slice of China’s demand — basically the toughest advanced nodes and the most risk-sensitive customers. ### What is the catch? A wafer target is not the same thing as full chip independence. China still faces bottlenecks in lithography, EDA tools, some deposition and etch gear, and the highest-end manufacturing know-how. So this is not the moment China “solves semiconductors.” It is more like securing one of the load-bearing materials under the stack. Important, yes — but still only one layer. ### Bottom line This story is really about where China thinks the next realistic win is. Not leading-edge lithography — too hard, too blocked. Not total chip autonomy overnight — too broad. Wafers are different. They are essential, measurable, and now close enough to scale that policy pressure can probably move the market. If Beijing clears 70% here, it will not end the chip war. But it will make China’s supply chain a lot harder to squeeze.