NorTech Innovations & Solutions

Remember the Great GPU Shortage of 2020 and 2021, when finding a graphics card felt like searching for a unicorn in a haystack? Well, brace yourself, because the tech world is currently navigating a new kind of turbulence. This time, it isn’t just about gamers; it’s about memory.

If you’ve noticed that the price of a new laptop, a high-end smartphone, or even a replacement RAM stick for your PC has suddenly “taken off” like a rocket, you aren’t imagining things. We are currently in the grip of a Global Memory Supply Shortage, and the culprit is exactly who you’d expect: Artificial Intelligence.

What Exactly is “Memory” (And Why Does It Matter?)

Before we dive into the chaos, let’s clear up the jargon. When we talk about “Memory” in this context, we are primarily talking about DRAM (Dynamic Random Access Memory).

Think of your computer or phone like a kitchen: The hard drive or SSD (Solid State Drives) is the pantry where food (data) is stored long-term. RAM is the counter space where the cooking (processing) happens.

If you have a tiny counter, preparing a complex meal becomes slow and cumbersome. You’re constantly moving things back and forth to the pantry, which slows everything down.

RAM is the high-speed workspace that allows your device to multitask. Without enough RAM—and without it being fast enough—even the most powerful processor becomes a bottleneck.

The AI Hunger Games: Data Centres vs. Consumer Electronics

So, why has the price of this “counter space” spiked by 200% to 400% in some sectors over the last year? It comes down to a massive shift in where the world’s silicon is going.

1. The HBM Gold Rush

AI models like ChatGPT, Claude, and Gemini are incredibly “hungry.” To train these models, companies like OpenAI and Google need thousands of specialized chips (GPUs). These chips require a very specific, high-end type of memory called HBM (High-Bandwidth Memory).

Major manufacturers—the ‘Big Three’—Samsung, SK Hynix, and Micron—have seen the writing on the wall. The profit margins on AI-grade HBM are massive compared to the standard RAM used in a budget Dell laptop or a Samsung Galaxy phone. Consequently, they are reallocating their factory lines. Every wafer of silicon used for an AI data centre is one less wafer available for consumer electronics.

2. The OpenAI Factor

To put the scale of this into perspective, some industry statistics suggest that OpenAI alone—just one company—is consuming a staggering portion of global specialized memory resources ¹. When you realize that every tech giant is currently building “AI Superclusters,” you can see why the supply for everyone else is drying up. Data centres are effectively “outbidding” the average consumer for the world’s memory supply.

3. Geopolitical Challenges

It’s not just about demand; it’s about the friction of moving product. Recent tariffs and trade restrictions have introduced additional direct costs and administrative ‘red tape.’ Customs inspections and compliance paperwork have increased handling times, and those extra logistics costs—higher freight and insurance—are being passed directly to you, the consumer.

The Collateral Damage: From Consoles to Cars

This isn’t just a problem for hardcore tech enthusiasts. The shortage is trickling down into every corner of the consumer market:

• Smartphones: Brands like Apple and Samsung are facing higher “Bill of Materials” (BOM) costs. While they try to absorb some of it, expect the “Pro” versions of phones to get even pricier.
• Gaming: Sony and Microsoft are in a tough spot. To keep consoles affordable, they have to find ways to monetize their existing users (think higher subscription fees or digital store cuts) because the physical cost of building the boxes is rising.
• Small Tech: Companies like Framework (known for their modular laptops) are heroically trying to keep prices stable, but smaller manufacturers are being forced to choose between raising prices or delaying launches.

The reality is simple: We are in an era where processing power, especially for AI, is the new gold.

And right now, the biggest drills are all pointed toward AI data centres.

By the Numbers: The Price of Progress

The impact on the market has been nothing short of breathtaking. Here is a look at how the landscape has shifted in just twelve months:

How to Navigate the Shortage

If you’re planning to build a new PC or upgrade your laptops, the ‘wait and see’ approach could backfire this time. Most analysts expect this shortage to persist through the end of the 2026, if not longer.

Here are three tips for the “Memory Crunch”:

1. Audit Your Current Gear: Before buying new, check if a simple software cleanup or a storage upgrade can extend the life of your current device.
2. Buy “Last-Gen” Wisely: If you need a RAM upgrade for a PC, DDR4 memory (last-gen) is currently more stable in price than the cutting-edge DDR5 being heavily impacted by AI demand.
3. Watch the Refurbished Market: Large enterprises often cycle out hardware regardless of market conditions. Certified refurbished units from 2023 are currently the “sweet spot” for value.

What’s Your Tech Temperature?

The shift toward AI is permanent, but the supply chain will eventually catch up—it always does. The question is: Are you willing to pay the “AI Tax” for the latest and greatest, or are you holding onto your current tech until the dust settles?

Have rising memory prices impacted your tech upgrades? Are you planning to wait out the shortage or upgrade your gear now?

Share your experiences in the comments or reply to this newsletter.

References

1. Yang, Heekyong, and Hyunjoo Jin. “Samsung, SK Hynix Set to Supply Chips to OpenAI’s Stargate Project.” Reuters, 1 Oct. 2025, www.reuters.com/business/media-telecom/samsung-sk-hynix-supply-memory-chips-openais-stargate-project-2025-10-01/. ↩︎