Building AI-ready infrastructure in ANZ with hybrid cloud

Australia is investing heavily in AI and data centre capability, supported by government initiatives like the National AI Plan. The goal is clear: accelerate AI adoption, strengthen digital capabilities and build secure infrastructure to support it.

But there is a growing disconnect between AI ambition and infrastructure readiness. Across Australia and New Zealand, organisations are accelerating AI adoption, from machine learning to generative and emerging agentic models, but the underlying infrastructure is not evolving at the same pace. The result is a widening gap between what businesses aim to achieve and what their environments can realistically support.

In this context, hybrid cloud architecture offers clear benefits for businesses. However, an IDC InfoBrief, Unlocking Business Agility Through Private Cloud Modernisation in Asia/Pacific, commissioned by Dell Technologies shows that adoption in Australia and New Zealand (ANZ) remains limited, with fewer than half of organisations using this approach.

This creates challenges for organisations, affecting both their ability to innovate with AI and operate effectively. Ageing IT leads to higher costs, greater cybersecurity risks and problems working with modern digital and cloud tools. Businesses that don’t modernise may find it difficult to keep pace in a changing economic environment.

The importance of a hybrid cloud strategy

One key part of modernisation is hybrid cloud, which is not new. It combines on-premises infrastructure with public cloud environments, allowing organisations to balance control, performance and flexibility. In theory, it gives businesses the ability to place workloads where they make the most sense: close to data, close to users or in the cloud when scale is needed. This flexibility is why hybrid cloud remains a preferred model as organisations scale and adapt to new demands.

Key challenges to hybrid cloud success

The reality, however, is that hybrid cloud is not always easy to execute. Operating across multiple environments introduces complexity, reduces visibility and makes governance harder to maintain. For many organisations, rapid cloud adoption can expose gaps in legacy on-premises infrastructure. It requires IT teams to re-platform applications into cloud-native environments, such as virtualised and containerised architectures to ensure efficiency and compatibility.

Also, skills shortages can make it harder for organisations to set up and manage systems, so they rely more on automation to fill the gaps. Security and compliance also become more difficult when working across multiple environments. At the same time, high upfront costs can be a challenge, leading many organisations to prefer pay-as-you-go pricing models.

To address these operational challenges, leading organisations are turning to AI-driven operations, commonly known as AIOps, to automate routine management tasks, predict potential failures, and optimise performance across hybrid environments. This shift from reactive to predictive infrastructure management is critical to scaling hybrid cloud effectively.

Winning the AI race with hybrid cloud

As AI adoption and hybrid cloud environments become key parts of enterprise IT, infrastructure needs to evolve to handle more complex demands. A strong hybrid cloud strategy is essential for scaling AI in a cost-effective and efficient way. It provides a practical foundation for AI by combining the scalability of public cloud with the control and performance of on-premises systems. To support both current and future AI workloads, on-premises systems need to handle high-performance computing and large amounts of data, and work smoothly with cloud services.

However, legacy systems create significant technical debt, leading to issues like scattered data, silos, duplication and poor data quality. These problems make data harder to use and trust, which in turn weakens analytics, slows AI adoption and increases operational and governance risks.

Modern AI infrastructure must support large, distributed data workloads, microservices-based applications, and hybrid environments. Crucially, all components, from edge to core to cloud should operate within a unified, open framework that ensures interoperability, simplifies management and delivers consistent performance regardless of where workloads run.

Australian organisations are already proving the value of this approach. Macquarie Cloud Services, a leading local enterprise cloud provider, recently modernised its infrastructure with AI-driven operations and advanced storage architectures to meet growing client demand for AI-capable environments. The results were significant: a 360% increase in throughput and an 83% reduction in latency, enabling them to support clients’ most demanding AI and data workloads while maintaining strict data sovereignty within Australia. Their experience underscores a critical point: when the underlying infrastructure is purpose-built for modern demands, both AI performance and business outcomes improve dramatically.

Strategic imperatives for modernisation in ANZ

Agentic AI systems can autonomously plan, reason and execute multi-step tasks, and therefore place even greater demand on infrastructure. These workloads require low-latency, high-throughput environments with robust data governance, making the case for modernised on-premises and hybrid architectures even more compelling. As agentic AI grows, both on-premises and hybrid cloud are becoming strong options for deployment. More than half of ANZ organisations (57%) are planning to invest in on-premises infrastructure to support their AI initiatives, reflecting a growing need for performance, control and data sovereignty.

At the same time, organisations using cloud-based AI need to make sure their applications and data can move easily between systems so they can shift back into hybrid environments when needed. This helps to avoid being locked into one vendor and reduces the risk of building new technical debt.

This trend is also driven by growing cybersecurity concerns, with 89% of organisations in ANZ planning cloud repatriation. It shows a clear move toward more balanced and resilient hybrid approaches.

Modernising on-premises infrastructure and adopting a hybrid cloud strategy is no longer optional for organisations in ANZ: it is fundamental to remaining competitive in an AI-driven economy. CIOs need to rethink infrastructure as a strategic business issue, not an operational one. That means moving beyond fragmented environments, avoiding vendor lock-in, and designing for portability from the outset. It also means adding automation and monitoring across all parts of the system to help manage growing complexity. Ultimately, the organisations that succeed in the next phase of AI will not be the ones that invest the most in AI tools, but the ones that fix the infrastructure beneath them.

Top image credit: iStock.com/jullasart somdok