Your next computer is going to cost more. Here’s why.
If you’ve been quoted on new laptops or desktops recently, you’ve probably noticed the prices are higher than expected. It’s not your imagination, and it’s not a temporary blip.
The cost of memory (RAM) inside every computer has surged by more than 100% year-on-year. That flows directly into the price of every laptop, desktop, and server your business buys. A standard business laptop that cost $1,800 six months ago might now be quoted at $1,950 or more - and servers with large memory configurations are seeing even bigger jumps.
We’re seeing it firsthand in every hardware quote we put together for clients. The same specs, from the same vendors, are simply more expensive than they were at the start of 2025.
This isn’t a simple supply chain hiccup. It’s a structural shift in how the world’s memory is manufactured, who it’s made for, and where the money is going. Here’s what’s actually happening, when it might ease, and what your business can do about it.
The Short Version
Three things are driving device prices up simultaneously:
- AI infrastructure is consuming an enormous share of global memory production - the same factories that make your laptop’s RAM are now feeding AI data centres
- Only three companies manufacture virtually all of the world’s DRAM - and all three are chasing AI margins
- The industry is transitioning from DDR4 to DDR5 - at the worst possible time
The result: less memory available for business computers, higher prices on every device, and no quick fix on the horizon.
Why Your Next Hardware Purchase Will Cost More
1. AI Is Eating the Memory Supply
The single biggest factor is the explosion in AI infrastructure spending. Microsoft, Google, Meta, and Amazon are building out AI data centres at an unprecedented pace, and those servers need enormous amounts of a specific, premium type of memory called High Bandwidth Memory (HBM).
Here’s the problem: HBM is made on the same production lines as the DDR5 memory that goes into your business laptops and servers. Every wafer allocated to AI memory is a wafer that doesn’t become the RAM in your next computer.
According to IDC’s February 2026 analysis, this reallocation of manufacturing capacity is the primary driver of the current shortage. In 2023, AI memory used about 5% of global DRAM capacity. By 2026, that’s expected to hit 32%. It’s a zero-sum game: the same factories simply can’t serve both markets at once.
Server memory alone is expected to surge over 60% quarter-on-quarter in early 2026, because the big cloud providers are securing supply before it reaches the broader market.
2. Three Companies Control Over 90% of the Market
Nearly all the memory in the world comes from just three manufacturers: Samsung, SK Hynix, and Micron. Together, they control over 95% of global DRAM production.
When all three shift their priorities in the same direction, toward AI and enterprise, there’s nowhere else for the rest of us to turn. Samsung has reportedly doubled the cost of DDR5 RAM and redirected large portions of production toward high-margin server modules. SK Hynix, the current leader in HBM, is ramping AI memory production through 2026, but very little of that is expected to ease pricing for business PCs.
Micron, the only US-based manufacturer, has issued end-of-life notices for several DDR4 products, pushing customers toward DDR5 supply that’s already stretched thin.
Less competition and higher-margin alternatives almost always mean higher prices for everyone else.
3. The DDR4 to DDR5 Transition
The timing couldn’t be worse. Current business platforms from Intel and AMD increasingly require DDR5, sometimes exclusively. If you’re buying new computers in 2026, you’re buying DDR5, whether you like the pricing or not.
At the same time, manufacturers are deliberately phasing out DDR4 production, shifting that capacity toward DDR5 and HBM where margins are better. This creates a squeeze from both ends: DDR4 is becoming scarcer as it approaches end-of-life, while DDR5 is expensive and supply-constrained due to AI demand.
It’s not just RAM, either. NAND flash (used in SSDs) is being pulled toward enterprise workloads too. SSD prices are starting to creep upward after years of decline.
The Tariff Wildcard
As if the supply picture wasn’t complicated enough, there’s a geopolitical dimension. US Commerce Secretary Howard Lutnick has warned Samsung and SK Hynix that they face potential 100% tariffs if they don’t expand chip manufacturing in the US.
A 100% tariff on memory imports would effectively double the cost of RAM on top of the price increases already happening. While this is primarily a US issue, global hardware pricing tends to follow, and Australian businesses buying internationally-priced devices would likely feel the ripple effects.
Whether these tariffs materialise remains to be seen, but the uncertainty itself is contributing to stockpiling behaviour that further tightens supply.
How Much Have Prices Actually Increased?
The numbers are significant:
- DRAM contract prices rose over 100% year-on-year through 2025, with some segments exceeding 150%
- Server DDR5 is projected to climb over 60% quarter-on-quarter in Q1 2026
- PC memory increased 55-60% in a single quarter, even with weaker notebook shipments
- NAND flash (SSD) contract prices jumped 33-38% in Q1 2026
Memory typically represents 15-20% of a business laptop’s component cost. When that component doubles in price, IDC estimates that PC average selling prices could rise 4-8% in 2026. For a business refreshing 20 or 30 machines, that’s thousands of dollars in additional spend.
When Will Prices Come Back Down?
This is the question every business owner wants answered. The honest answer: not quickly.
Industry analysts from TrendForce, IDC, and Gartner are broadly aligned:
- Q1 2026 (right now): Peak pricing. This is likely the most expensive quarter for hardware purchases.
- Q3 2026: Some relief expected as production volumes increase, though prices will still be well above 2024 levels.
- Late 2026 to mid-2027: The most likely window for meaningful price normalisation (base-case scenario).
- Late 2027 to early 2028: Worst-case scenario if AI spending continues to accelerate and new fabrication plants are delayed.
The fundamental constraint is that building new chip factories takes years. New facilities from Micron, Samsung, and SK Hynix won’t contribute meaningful additional capacity until late 2027 at the earliest. You simply can’t fast-track a semiconductor fabrication plant.
What This Means for Your Business
If you’re an Australian business, whether you’re on the Gold Coast, Brisbane, or anywhere else, here’s how this affects you in practical terms:
New laptops and desktops cost more. Every PC contains DRAM. As memory prices rise, so does the sticker price on every device you purchase. The same HP or Lenovo laptop that you quoted six months ago is simply more expensive today.
Hardware refreshes are more expensive. If you’ve been planning a fleet upgrade, your per-device costs are higher than any quotes from late 2025. Get updated pricing before committing to timelines or budgets.
Servers are hit hardest. Servers typically need significantly more RAM than laptops. A database server, virtualisation host, or Remote Desktop server with 256GB or 512GB of DDR5 is where the price impact is most acute.
Upgrade cycles may need to stretch. Where it used to make financial sense to refresh hardware every 3-4 years, some businesses may find it more cost-effective to extend the life of existing equipment until prices stabilise.
What You Can Do Right Now
You can’t control global memory markets, but you can make smarter decisions about how you navigate them:
If you need it, don’t wait. Prices are expected to stay elevated through at least mid-2026. If you have a genuine business need for new hardware, delaying probably won’t save you money in the short term.
Consider staggered rollouts. Instead of refreshing an entire fleet at once, stagger purchases across quarters. This spreads the cost and lets you benefit from any price relief as it arrives.
Check DDR4 availability. If your current systems support DDR4, it may be more cost-effective to add memory to existing machines rather than purchasing entirely new DDR5 systems, but act soon, as DDR4 supply is shrinking.
Talk to your IT provider. A good managed IT partner can help you assess what’s urgent, what can wait, and where you can optimise spending. We’re tracking the memory market closely and can help you make informed decisions about timing and specifications.
The Bottom Line
The computers your business buys are more expensive right now, not because of any single event, but because of a perfect storm: AI is consuming memory production capacity, only three companies control the supply, and the transition from DDR4 to DDR5 is squeezing availability at exactly the wrong moment.
The situation is expected to ease gradually from mid-2026, with more meaningful relief arriving in 2027 as new fabrication capacity comes online. Until then, plan for higher hardware costs and be strategic about when and what you buy.
If you’re navigating hardware purchases and want help making sense of the timing, get in touch. We help businesses across the Gold Coast and South-East Queensland make smart technology decisions, even when the market is working against them.
