When Gartner predicted that global spending on sovereign cloud infrastructure would reach $80 billion in 2026, it was the clearest sign yet that the cloud market is entering a more complex phase for technology leaders.
Growth is driven not only by demand, but also by concerns around control, resilience and risk. For CIOs and CTOs, this moves cloud planning beyond optimization to more complex decisions around cost, capacity and placement.
Senior Cloud Economist at Nutanix.
Recent results from vendors like AWS show that demand for the public cloud remains strong, and this will only increase with interest in AI tools. Capacity is increasing, services are increasing and investments remain intense. But for enterprise technology leaders, this growth doesn’t eliminate the need to make compromises.
Article continues below
As AI-driven workloads require more and more memory and compute, assumptions about elastic capacity and predictable economics are increasingly difficult to maintain, particularly outside of hyperscale platforms, where exposure to cost volatility and provisioning delays is more immediate.
AI application
This pressure is driven by the speed at which demand for AI is moving from experimentation to production. According to an Omdia analysis, global cloud infrastructure spending reached $102.6 billion in the third quarter of 2025, up 25% year-over-year, as companies expanded their AI workloads across their core systems.
At the same time, Deloitte research highlights that AI is no longer limited to individual applications, but is becoming a fundamental layer of the enterprise technology stack.
This shift significantly increases demand for memory- and compute-intensive workloads, changing the assumptions CIOs can make about cost, scalability and availability.
As prices become more volatile and provisioning is less predictable, CIOs face friction in programs designed to make things easier.
Projects are now delayed, budgets are revised and, in some cases, existing infrastructure is kept in place longer than planned because alternatives are unavailable or no longer economically viable.
The problem lies not simply in increasing spending, but in the growing gap between technological ambitions and what the underlying infrastructure can reasonably support.
Cloud strategy
For much of the last decade, cloud strategy has often meant a steady migration to public platforms. As AI-driven workloads place constant demands on memory and compute resources, this assumption becomes increasingly difficult to maintain.
CIOs must increasingly distinguish between workloads that truly benefit from hyperscale elasticity, those that require tighter control over costs or data locality, and those that need the flexibility to scale as conditions change.
In practice, this leads to a more selective approach to cloud adoption, one that balances public cloud, private infrastructure, and hybrid models to manage cost, performance, and risk.
In practical terms, this means that CIOs can no longer treat workload placement as a one-off architectural decision. They need to have a clear view of which systems are truly elastic, which are cost-sensitive, and which are mission-critical.
This requires a closer look at memory and compute requirements, more realistic assumptions about price volatility, and contingency planning in case of delays or shortages. It also means avoiding rigid designs that lock workloads into a single environment.
Organizations that fare best are those that build in optionality, the ability to rebalance workloads, defer non-essential demands, and protect critical systems when capacity tightens or costs rise.
Hybrid models
For many organizations, hybrid architectures appear to be the most pragmatic way to manage this complexity. Public cloud continues to make sense for workloads that benefit from rapid scaling, peak capacity, or access to managed AI services.
Private infrastructure, on the other hand, provides greater predictability in cost, performance, and availability for memory-intensive or business-critical systems. Hybrid models allow CIOs to combine these strengths, placing workloads where they make the most sense and retaining the flexibility to adapt to changing conditions.
Done right, it’s about creating a cohesive operating model that aligns infrastructure choices with business priorities rather than consolidating everything onto a single platform.
Of course, hybrid alone is not a panacea. Private cloud projects are themselves exposed to many of the same pressures that shape the broader market, including memory availability, delivery times and cost. Hardware constraints don’t disappear just because workloads leave hyperscale platforms.
The difference is that hybrid models give organizations more control over how these constraints are managed. By distributing demand, sequencing deployments, and maintaining the flexibility to modify workloads as conditions change, CIOs gain flexibility that a single-platform strategy rarely allows.
The goal is to prevent this constraint from becoming a single point of failure.
Risk management
For CIOs, this makes cloud strategy inseparable from risk management. Decisions about where to run workloads increasingly affect financial exposure, operational resilience, and regulatory compliance, not just performance metrics.
As a result, cloud planning is evolving, or expected to move closer to the center of enterprise governance, requiring closer alignment between technology leaders, finance teams and boards of directors.
Looking ahead, cloud adoption is entering a new phase. AI will continue to drive demand, while constraints in memory, compute, energy and supply chains will likely persist. In this environment, cloud strategy requires regular reassessment rather than periodic overhaul.
The most effective CIOs and CTOs will be those who plan with uncertainty in mind, test assumptions early, and maintain the ability to adapt as conditions change. The cloud has become a permanent part of modern organizations.
What is changing is the level of focus and governance it now requires to support growth without exposing the business to unnecessary risk.
We have presented the best cloud storage.
This article was produced as part of TechRadarPro’s Expert Insights channel, where we feature the best and brightest minds in today’s technology industry. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you would like to contribute, find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro