Three Layers. One Compounding Advantage. The IREN Thesis.

There's been a lot happening at IREN recently.

Expansion across North America, Europe and Asia-Pacific.

The NVIDIA partnership.

The Mirantis acquisition.

New GPU deployments.

New customer discussions.

A growing global footprint.

Underneath all of it is a fairly simple view of where the world is heading, and a deliberate strategy for how we position IREN within it.

That strategy is built on three layers. Together, they compound into a structural advantage that gets harder to replicate every quarter we execute.

Layer 1: Physical infrastructure. Power, land, substations, data centers, cooling. The foundation that everything else sits on.

Layer 2: Compute infrastructure. The GPUs, servers and networking that go inside those buildings. Deployed at scale. Generating revenue. Building execution track record.

Layer 3: Software and operational capability. The orchestration, deployment tooling and enterprise expertise that makes the first two layers work harder for customers, and opens the door to a broader, higher-value market over time.

Layers 1 and 2 are where the overwhelming majority of IREN's value is being created today. Layer 3 is where that advantage compounds further over time, but only because Layers 1 and 2 are built, owned and controlled at scale by IREN, not subscale nor contracted from a third party.

Think of Amazon. They didn't win e-commerce by building a great website. They won it by controlling the fulfilment infrastructure at a scale nobody else could replicate. The foundation you don't control becomes the ceiling on your business.

That is exactly how we think about IREN. The physical infrastructure - the land, the power, the substations, the data centers - is owned and controlled by us. The compute deployed into it generates the revenue and execution track record. And the software, orchestration and enterprise capability we are more methodically building on top is what turns the total product into a vertically integrated AI Cloud platform that compounds over time and deepens into a competitive moat.

AI is still early. The bottleneck is increasingly physical. And we have spent eight years building the foundations.

The Digital World is Scaling Faster Than the Physical World

AI demand grows exponentially.

Infrastructure doesn't.

Every few months we see another meaningful step change in model capability. Better reasoning. Better coding. Better multimodal understanding. Better agents. The latest generation of frontier models feels like one of those moments.

Every time that happens, usage increases, new products emerge, enterprises deploy more workloads, and entirely new use cases become viable. All of those things compound on top of each other.

Think back to the dial-up era. The internet existed. Email worked. Websites loaded - eventually. But the sluggishness of the experience fundamentally constrained what people imagined doing with it. The technology's own limitations shaped the universe of perceived use cases.

AI is in exactly that phase right now. The friction of slow inference and expensive compute subtly depresses AI ambition and imagination. That friction is a function of compute scarcity. It will not persist.

And here is where it gets interesting. It's not just that more compute serves existing demand. It's that more compute creates demand that didn't previously exist. It's like building more highway lanes to reduce traffic. The lanes don't just move the cars that were already queuing, entirely new trips get made that nobody was taking before. People move further from the city. New suburbs get built. New businesses open along the route. New industries restructure themselves around the road.

AI compute behaves the same way. A manufacturer discovers they can run real-time process optimization across every plant simultaneously. A hospital system can model patient outcomes at a level of granularity that was previously unthinkable. A logistics company rebuilds its entire routing infrastructure around live AI inference. None of those use cases were in anyone's demand forecast, because they only became conceivable once the compute existed to make them viable.

The capacity doesn't just serve the demand that exists. It creates the demand that comes next. And then that demand creates the use case after that. It spirals up.

Jensen said it plainly on NVIDIA's most recent earnings call: "Today's data centers are revenue generating AI factories constrained by power." And then this: "Demand has gone parabolic. The reason is simple. Agentic AI has arrived. AI can now do productive and valuable work. Tokens are now profitable, so model makers are in a race to produce more."

Meanwhile the physical world moves slowly.

Permitting. Grid connections. Construction. Cooling systems. Power generation.

You can't just add another 5GW of capacity every time models improve and compute demand accelerates. The physics don't care that every industrial company on earth is about to integrate AI into its operations.

The real constraint in AI is increasingly time-to-compute.

Two Kinds of Infrastructure. One Compounds the Other.

When people talk about AI infrastructure, they tend to conflate two very different things. The distinction matters.

Layer 1 is the physical foundation - power, land, substations, data centers. Layer 2 is the compute that sits inside them - GPUs, servers, networking.

IREN has spent eight years securing that Layer 1 advantage. The land. The power. The substations. Data center sites being developed and built across multiple continents. We were building before it was obvious.

And we are deploying Layer 2 into it right now, at scale. GPUs online, generating revenue, building the operational track record that makes every subsequent deployment faster and more credible.

The barriers to entry are already forming. Developing a large-scale new data center today won’t be online until the end of the decade. And the asset-light neocloud trying to compete by renting capacity is discovering that sites were locked up years ago, and the operators utilizing them aren’t subletting. By the time new entrants solve for land, power and permitting, IREN will have gigawatts online, execution track record, and customer relationships that took years to build. That gap doesn’t close. It compounds.

Vertical integration across both layers is not just a structural advantage, it is a commercial one. Controlling the physical infrastructure and the compute sitting on top of it means faster deployment, greater certainty for customers, tighter operational optimization, and lower dependency risk at every stage. There are no landlords to negotiate with, no capacity constraints imposed by third parties, no contractual barriers between an operator and their own infrastructure, no misaligned incentives between the infrastructure owner and the operator. The economics improve as the platform scales, and the customer experience improves with it. That is what owning the full stack actually means in practice.

Building a Global Platform

A big part of our strategy has been building at scale, and building globally.

Today IREN has secured power and active development pipelines across Texas, British Columbia, Oklahoma, Spain, Australia, and more. Five gigawatts of secured, grid-connected capacity in total. Multiples of that coming behind it.

AI demand is global. The infrastructure layer supporting it needs to be global too. Capital and customer diversification across regions reduces concentration risk and strengthens the overall platform. Some customers care about latency. Some care about data sovereignty. Some care about renewable power. Some simply need capacity wherever it exists.

Europe is one of the largest and fastest-growing AI demand markets in the world, and it is significantly underserved from an infrastructure perspective. Asia-Pacific is home to roughly 60% of the world's population and some of the fastest-growing AI adoption on earth, with an infrastructure deficit to match. The supply simply does not exist at the scale the demand requires, and building it takes years – if you can find it.

Our Spain footprint positions us directly in the European market. Our Australia pipeline with direct submarine fibre connectivity to Japan and Singapore anchors our Asia-Pacific presence. And for customers where data residency matters, consider a company operating under GDPR: routing sensitive workloads through a Texas data centre may simply not be an option, regardless of how competitive the pricing is. Regional infrastructure may not just be a preference for those customers. It may simply be a requirement.

The NVIDIA Partnership

NVIDIA re-segmented its data center business on its most recent earnings call into two categories: Hyperscale, and ACIE - AI Cloud, Industrial, and Enterprise. ACIE revenue reached $37 billion in Q1 FY2027, growing 31% sequentially, with AI cloud revenue more than tripling year over year. That is the segment IREN operates in, and it is the fastest-growing part of NVIDIA's business.

Jensen's framing of what is happening is worth sitting with. 'Today's data centers are revenue generating AI factories constrained by power.' It's not hardware. Not chips. Infrastructure, constrained by the physical layer that IREN has spent years building.

NVIDIA's DSX architecture is a repeatable, scalable blueprint for deploying those AI factories in the real world. In our recently announced partnership, NVIDIA has chosen IREN to operationalise that architecture at gigawatt scale - precisely because of our unique position to own and control all three layers: the physical infrastructure, the compute deployed into it, and the operational capability to run it reliably for the world's most demanding customers.

We signed a five-year, $3.4 billion AI Cloud contract with NVIDIA, for Blackwell GPU deployments across 60MW of gross data center capacity at our Childress campus. Our Sweetwater campus in Texas is positioned as the flagship DSX deployment. And as part of the broader partnership, NVIDIA received a five-year right to purchase up to 30 million IREN ordinary shares at $70 per share - a potential investment of up to $2.1 billion. Those investment rights vest in tranches tied to the delivery of up to 600,000 NVIDIA GPUs across our platform.

The structure reflects that alignment directly. They win as we scale. And we scale faster with NVIDIA deeply aligned alongside us.

Every frontier model and every open source model runs on NVIDIA. They control chip allocation, the customer ecosystem and engineering support across the entire AI stack. And they have a direct and growing economic interest in our operational success.

Why We Acquired Mirantis - And What Layer 3 Really Means

Layers 1 and 2 are where the majority of value is created today. Layer 3 is how we further compound both into something harder and harder to displace over time.

Mirantis has spent more than a decade helping enterprises deploy and manage cloud infrastructure. They are a founding Independent Software Vendor partner of the NVIDIA AI Cloud Ready Initiative. Their k0rdent AI platform is designed to manage AI infrastructure across bare metal, virtual machines and Kubernetes environments - exactly the operational complexity that emerges as GPU deployments scale into production. The acquisition strengthens IREN's platform across four areas: deployment capability, operational visibility, customer support, and market access.

The software commoditization process is well underway. Operationalizing AI infrastructure reliably at scale - provisioning, monitoring, performance visibility, enterprise support across demanding production environments - is much harder. That is what Mirantis strengthens.

There is a longer-term opportunity here too. As the enterprise AI market matures, we see real potential to serve a much broader and more diverse customer base. Large anchor contracts - hyperscalers, frontier labs, model trainers, demand aggregators - are foundational. But the rest of the world increasingly wants compute too. Enterprises standing up AI environments. Mid-market companies that need managed infrastructure without the complexity of building it themselves. Customers who want orchestration, support and reliability wrapped around the compute - not just bare metal. Mirantis's decade of enterprise relationships with over 1,500 enterprise customers globally is directly relevant to that opportunity.

The downstream expansion will still be available in two years. In three years. In five years. Serving a more diverse customer base. Offering higher-value managed environments. Capturing more of the value chain. First movers will have an advantage there too, but nothing like the advantage of owning the infrastructure layer that everything else runs on. The window to get hundreds of thousands or even millions of GPUs online and establish IREN as a globally significant compute platform? That closes. So we know what we are doing first.

Building the Brand to Match the Ambition

Infrastructure at this scale creates a platform. Becoming the partner customers think of first requires more than operational excellence. It requires presence.

That requires showing up consistently in the markets where decisions get made. We are investing in the media, content and partnerships that build that presence - including sporting partnerships that build brand presence, open doors with enterprise decision-makers, and help us compete for the best talent in the world. We are also conscious of our roots. IREN was built in communities - in the regions where we source our power, build our facilities, and employ the people who operate them. That does not change as we scale.

The ambition is simple: when the world needs AI at scale, they think IREN.

The Road Ahead

IREN has been building Layer 1 for eight years. We are deploying Layer 2 at scale right now. And with Mirantis, we have taken the first deliberate step toward Layer 3.

The physical infrastructure advantage comes first.

Everything else compounds from there.

It has become almost cliché to say we are still in the early innings. But it genuinely feels that way. The opportunity feels generational. We have assembled the key inputs to be in a position to compound a genuine time-locked competitive advantage – and right now, the biggest opportunity in the world is simply bringing more of that compute online. These journeys are rarely linear. There will be speed bumps and challenges along the way that we cannot yet see. But we have never been more excited about what lies ahead, and intend to make the most of it.