If you've been thinking about upgrading your gaming PC, here's some unwelcome news: processors just became one of the hardest computer components to find. A supply crunch is hitting the PC and server CPU markets, leading to an average price hike of between 10 to 15 per cent, with some CPU model prices spiking even higher.
An executive with a gaming PC company said that the shortage of CPUs is expected to become worse in the April-June quarter, stating that Intel and AMD have prioritised capacity for server CPUs, and the supply for PCs has become less, with much less volume available in Q2 compared to Q1.
What makes this crunch different from previous shortages is not a factory disaster or logistical breakdown. AI workloads are pulling semiconductor capacity away from traditional IT infrastructure; GPUs got squeezed first, memory followed, now CPUs. Every manufacturing slot and every wafer coming off the line goes to the highest bidder, and right now, that's hyperscale cloud companies building AI infrastructure.
The AI industry is shifting from building chatbots to deploying autonomous software agents, a shift that fundamentally changes the ratio of hardware data centres need; agentic AI systems plan, execute multi-step tasks, call APIs, query databases, write and run code, coordinate dozens of sub-processes, and evaluate whether they succeeded before starting over, all of which work happens on CPUs. That's different from how large language models work, where GPUs do most of the heavy lifting.
The supply squeeze is being felt most acutely by those making mid-range systems. Mid-range x86 chips are subject to the most acute supply crunch, with Intel tilting production in favour of high-end chips. Smaller PC vendors and custom system builders are in a weaker negotiating position than the massive OEMs, which means they're getting rationed first.
Intel has warned Chinese customers of delivery lead times of up to six months for certain server CPUs, AMD's lead times have stretched to 8-10 weeks for some products, and Intel server chip prices in China have risen more than 10 per cent. The fact that prices are rising even when supply is being rationed speaks to the desperation among buyers locking in allocations while they can.
Intel and AMD find themselves in different predicaments. Intel has been struggling with manufacturing yields at its own fabrication plants, limiting how many usable chips it can produce from each silicon wafer, and the company is investing in new tools and reallocating capacity from PC chips to server chips, but improvements won't arrive until later this year at the earliest. AMD relies on TSMC in Taiwan, the world's most advanced contract manufacturer, but TSMC is prioritising its most advanced production lines for higher-margin AI accelerators and GPUs, leaving less room for CPU orders.
One gaming PC maker put the situation bluntly: "What we worry about is that even if we pay more we still cannot get more. The CPU shortage is getting more serious day by day, no less than the memory chip situation." That last comparison carries weight. Memory availability has been strangling the entire hardware market for months, driving up prices across consumer electronics, smartphones, and PCs alike. Now the same squeeze is hitting the component most essential to making any computer actually work.
This is not just a cyclical shortage driven by a mismatch in supply and demand, but a potentially permanent strategic reallocation of the world's silicon wafer capacity; the dynamic has inverted where the voracious demand for memory by hyperscalers has forced the biggest memory manufacturers to pivot their limited cleanroom space and capital expenditure towards higher margin enterprise-grade components, meaning every wafer allocated to an HBM stack for an Nvidia GPU is a wafer denied to the LPDDR5X module of a mid-range smartphone or the SSD of a consumer laptop.
From a fiscal perspective, the situation creates a worrying precedent. When capital investment flows entirely toward enterprise infrastructure rather than consumer-facing manufacturing, the invisible hand stops optimising for ordinary people's interests. The memory crisis taught us that lesson already. Now it's being taught again with CPUs. One semiconductor analyst expects the shortage to continue through all of 2026 and deep into 2027.
This raises a genuine tension between competing goods. A more powerful AI infrastructure arguably benefits the broader economy; autonomous agents solving problems at scale could improve productivity and drive innovation. That's a reasonable case, and wealthy data centre operators are making it loudly in boardrooms and to shareholders. But that same abundance of capital for enterprise infrastructure comes directly at the expense of a functioning consumer hardware market. Gaming PCs, student laptops, and small business workstations become secondary concerns.
For now, the advice from hardware retailers and PC builders is straightforward: if you need new components, lock in pricing and inventory now. Global PC shipments are expected to drop 11.3 per cent in 2026 compared to 2025. That's not because people stopped wanting computers. It's because the semiconductor industry has reorganised itself around what's profitable, not what's needed.