DRAM Shortage Could Last to 2030, ADATA Chief Says

Key takeaways

• ADATA’s chairman expects tight memory-chip supply to last deep into this decade.
• AI servers need far more memory than many ordinary computers.
• The forecast is an industry view, not a sure promise about future prices.
• Phone, PC and server makers may face higher costs if supply stays tight.

DRAM shortage is a lack of enough computer memory chips for all buyers. ADATA’s chairman says this squeeze could run until 2030. That would keep pressure on firms building AI systems. It could also affect the cost of everyday devices.

What is a DRAM shortage?

DRAM stands for dynamic random-access memory. It is the short-term workspace inside phones, laptops, game consoles and servers. A server is a powerful computer that delivers websites and apps. When a program needs data quickly, DRAM helps it find that data.

A DRAM shortage happens when chip makers cannot produce enough memory for buyers. Prices can then rise because firms compete for limited stock. This matters most for big buyers, such as cloud companies. They buy huge batches of chips for data centres, which are buildings packed with computers.

The market has three dominant memory makers: Samsung Electronics, SK hynix and Micron. Their factory plans shape supply across the world. But opening a new chip factory takes years and costs billions of dollars.

Could the DRAM shortage really last until 2030?

ADATA chairman Simon Chen told TechRadar that he sees the DRAM shortage lasting for another decade. He also said he does not expect the AI boom to burst before 2030. That is a bold forecast, so readers should treat it as one executive’s view.

Chen’s point is simple. AI firms want more memory chips each year, while new factories arrive slowly. A chip plant cannot work like a bakery that simply adds another oven. It needs clean rooms, rare tools, skilled workers and long testing periods.

The forecast does not mean every memory chip will stay scarce for 10 years. Demand can cool, and producers can change their plans. Still, the warning shows why memory has become a key part of the AI race.

ADATA chairman’s outlookTight DRAM supply forecast20252030Forecast, not a guaranteed market outcome

The chart shows the time span in Chen’s forecast. It covers roughly five years from 2025 to 2030. His wider claim is that pressure could last even longer.

Why does the DRAM shortage matter for AI?

AI models need to hold and move vast amounts of data. That is why they use advanced memory, including HBM. HBM means high-bandwidth memory. It moves data very quickly between a chip and its memory.

HBM is especially useful beside AI graphics processors. Graphics processors are chips that handle many tasks at once. They help train AI models by sorting through large piles of text, pictures or video.

Making more HBM can limit other memory output. A factory may shift tools and workers toward the chips with higher demand. As a result, the DRAM shortage can spread beyond the most advanced AI machines.

Buyer	Why memory matters	Likely risk if supply stays tight
AI data centres	They run large models and store fast-moving data.	Higher server costs and slower expansion.
PC makers	Laptops need DRAM for apps and games.	More expensive parts or fewer upgrades.
Phone makers	Memory keeps apps open and supports AI features.	Pressure on prices for premium models.

This does not mean a laptop will suddenly cost twice as much. Device firms often hold chip stock and sign supply deals early. But a long squeeze could make cheap upgrades harder, especially for high-memory products.

What should buyers and investors watch next?

Watch memory makers’ factory plans and their comments on HBM output. Also watch cloud firms’ spending plans. If they keep ordering AI servers, demand for fast memory should remain strong.

Micron, Samsung and SK hynix publish results and investment updates each quarter. Those reports offer firmer clues than any single prediction. Readers can follow Micron’s investor updates for company-reported supply and demand signals.

India has a stake too. More AI data centres need power, land and imported hardware. For example, HCLTech’s planned Odisha data centre investment shows how quickly local computing capacity is growing.

The clearest takeaway is this: AI demand has made memory chips strategically important. A long DRAM shortage would not stop AI. Instead, it would make the race to build AI systems more costly and more competitive.

FAQs

How does a DRAM shortage affect a laptop?

It can raise the cost of a key part. Brands may then increase prices, cut storage options, or delay some models.

What is HBM memory?

HBM is high-bandwidth memory. It is a fast type of memory used with powerful chips, especially for AI work.

Why are AI companies buying so much memory?

AI systems handle huge files and many tasks at once. They need large amounts of fast memory to work well.

</body_html