Azure’s Chip Shift: What It Means for Nordic Companies Operationalizing AI

If your company is starting to turn AI from strategy into action, there’s a subtle shift happening that’s worth paying attention to.

Microsoft Azure — the leader in cloud AI infrastructure — is now officially opening the door to non-NVIDIA GPUs, starting with AMD’s powerful new MI300X chips.

It may sound like a technical tweak, but this is actually a strategic pivot — and one that Nordic businesses, especially mid-sized ones planning pilots or early rollouts, should be factoring into their AI roadmaps.

Why This Matters Now

Until recently, running modern AI models-especially large language models-pretty much meant using NVIDIA GPUs. They’re powerful, but expensive and in seriously high demand. Even major cloud providers like Microsoft have had to juggle access and availability.

That’s why Azure is now supporting AMD’s MI300X chips. Why is this a big deal?

• 192 GB of memory-great for GenAI and LLMs
• Comparable performance to NVIDIA’s flagship GPUs (at potentially lower cost)
• Seamless integration across Azure’s AI services

In short: You now have real choice when planning your infrastructure. You’re no longer tied to a single vendor’s availability or price curve.

So, Why Is Microsoft Doing This?

Here’s what’s driving the shift:

There simply aren’t enough high-end NVIDIA chips to go around. Azure adding AMD is like an airline bringing in a second jet supplier — it’s not about replacing what works, it’s about meeting growing demand.

AMD-powered virtual machines offer another price-performance option, giving companies like yours a chance to right-size infrastructure to your budget and workload.

Different workloads benefit from different architectures. For tasks that need massive memory bandwidth, AMD’s architecture might actually be a better fit.

Using more than one chip supplier protects Microsoft — and you — from future supply chain hiccups, trade restrictions, or vendor-specific vulnerabilities.

This Isn’t Just About Chips. It’s About Sustainability, Too.

The Nordic region has an edge here. While major European data center hubs face energy constraints, the Nordics benefit from:

• Abundant renewable power
• Natural cooling advantages
• Mature digital infrastructure

As AI infrastructure grows more power-hungry, location matters more than ever. Microsoft’s chip diversification makes it easier to run sustainable, efficient AI workloads-especially when paired with Nordic data center strengths.

How Innofactor Helps You Make the Most of This

Here’s where Innofactor comes in. As a leading Nordic Microsoft partner, we’re working closely with companies like yours to operationalize AI in a way that’s smarter, greener, and more adaptable.

Flexible Architecture by Design
We help design Azure-native environments that support both NVIDIA and AMD, so you’re never locked in.

Infrastructure Readiness Assessments
We’ll assess where you are now, and where AI fits into your strategy-then map out the infrastructure you’ll need to get there.

Green AI Strategies
We help you explore hybrid models or fully local compute options that run on Nordic renewables-good for business, better for sustainability.

Proof-of-Concept Accelerators
Want to see what AMD chips can do for your workload? We can spin up a quick, focused pilot to show you.

Your AI Infrastructure Planning Matrix

Final Thought

For many Nordic companies, AI is a real part of the business roadmap now. And as you shift from strategy to implementation, the infrastructure choices you make now will shape your ability to scale, innovate, and stay cost-effective.

Microsoft’s chip shift is your signal to think broader.
And Innofactor is your guide to making it real.

Let’s build AI infrastructure that’s ready for tomorrow-flexible, sustainable, and made for the Nordic way of doing business.