AI is clearly accelerating demand for cloud computing, but not in the way many expected. The biggest story right now is not software innovation but the extraordinary capital flowing into physical infrastructure: chips, networking gear, power systems, and massive data centers. Providers are racing to support model training and inference workloads, and this shift is reshaping the cloud market.

Key Facts at a Glance

• US tech companies including Alphabet, Amazon, Meta, and Microsoft are expected to spend roughly $650 billion on AI-related infrastructure in 2026, up from $410 billion in 2025.
• Nvidia plans to invest $2 billion each in photonics companies Lumentum and Coherent, highlighting the importance of data movement and energy efficiency in AI systems.
• Enterprises often start AI projects in the public cloud for speed and flexibility, but many are reconsidering as workloads become persistent and costs rise.
• Repatriation and neocloud providers are gaining traction as cost-efficient alternatives for steady-state AI workloads.
• The AI-driven cloud market is likely to become more segmented, with public clouds, on-premises environments, and specialized providers each playing distinct roles.

The numbers behind this infrastructure push are difficult to ignore. According to analysis cited by Reuters, US technology companies are expected to invest about $650 billion in AI-related infrastructure in 2026, compared to roughly $410 billion in 2025. That kind of growth signals a fundamental change. AI is not just another software wave that sits neatly on top of the existing cloud stack. It is forcing a redesign of the stack itself, from the silicon up to the orchestration layer.

This redesign reaches deep into networking and data movement. Nvidia recently announced plans to invest $2 billion each in photonics companies Lumentum and Coherent. The move underscores where the pressure points are emerging. The issue is no longer only raw compute. It is also how quickly data can move between processors, racks, and clusters without creating unacceptable bottlenecks or power inefficiencies. As AI systems scale, latency, throughput, and energy usage become first-order economic concerns. The physical layer of the cloud is becoming the strategic center of gravity.

All of this suggests AI will absolutely drive more demand for public cloud computing, but the demand will be uneven. Public cloud providers remain the fastest way to access advanced infrastructure, global scale, and managed AI services. At the same time, the cost profile of large, persistent AI workloads is prompting many enterprises to reconsider whether the traditional hyperscaler model should remain the default destination for every stage of the AI life cycle.

Most AI starts in the public cloud

When companies are experimenting, speed matters more than optimization. Public clouds give teams immediate access to GPUs, foundation model APIs, vector databases, orchestration tools, security controls, and integration services. They also allow businesses to quickly start pilots without waiting for procurement cycles, data center expansions, or specialized infrastructure teams. The cloud dramatically lowers the barrier to entry, providing compute as well as a full operating environment for AI experimentation.

Given the high level of uncertainty, the public cloud is often the right choice for first-generation AI. Enterprises do not yet know which use cases will deliver value, how much inference traffic they will see, or which architecture model will ultimately survive. At this stage, the ability to quickly try many things is more important than squeezing every dollar from the underlying infrastructure. Managed services reduce friction, and friction is the enemy of early adoption.

We are seeing strong initial demand for AI land in public cloud environments. Enterprises are building chatbots, copilots, knowledge assistants, document automation systems, and code generation tools there because the cloud makes experimentation fast and accessible. This pattern is consistent across industries, from financial services to healthcare to manufacturing. The hyperscalers have spent years building out their AI offerings, and that investment is paying off as enterprises look for proven platforms to launch their first AI pilots.

Next-gen AI systems present choices

The second generation of enterprise AI systems looks different. Once a use case proves its value and usage becomes persistent, the financial model changes. A workload that looked inexpensive during a proof of concept can become shockingly expensive when it runs at production scale, especially if it depends on premium GPU instances, high-performance storage, constant network traffic, and managed services layered on top of one another.

That is where repatriation enters the conversation. We are starting to see a pattern in which enterprises build first-generation AI systems on public clouds, learn what works, and then move some of those workloads back on-premises or onto so-called neocloud providers that offer AI-optimized infrastructure at a lower cost. Neocloud providers

Source:InfoWorld News