For many companies, the hardest part of artificial intelligence has little to do with the technology itself. It is working out where the systems will run, who will control the data, and how the monthly bill will be justified when the pilot phase gives way to daily use.
At Nutanix’s annual NEXT conference in Chicago last week, the hybrid multicloud software company made clear how quickly the AI conversation is moving towards hosting, governance and accountability, alongside computing power.
Nutanix, which says it serves more than 30,000 customers worldwide, used the event to declare a new category taking shape in the AI market: specialist cloud providers built around artificial intelligence services and infrastructure for enterprises that want tighter control over data, costs and operational responsibility. Nutanix calls them neoclouds.
The company says: “The AI era has fuelled the emergence of neocloud providers offering rapid access to GPUs through flexible, on-demand services. While demand has largely been driven by AI training workloads from a small number of large customers, the next phase of AI will centre on scaling inference and running agentic AI applications in production for a large number of enterprise customers.”
The largest cloud platforms still dominate on scale, reach and capital. Many enterprises, however, are looking for something narrower and easier to answer for. They want AI infrastructure with clearer lines of responsibility, closer jurisdictional fit and fewer billing shocks. They also want stronger control over where workloads run and how data is handled once systems move from experimentation into everyday use.
Thomas Cornely, executive vice president of product management at Nutanix, says: “Demand for sovereign and specialised AI clouds is accelerating as organisations look for ways to access AI while maintaining control over their data. The Nutanix Agentic AI solution, with its secure multitenant and AI management portal, is designed to enable neocloud providers to rapidly deliver advanced high value AI services to enterprises and public sector organisations looking for powerful AI capabilities from trusted regional providers.”
That takes the argument away from model benchmarks and chip counts and places it squarely inside the discipline of running a business. Finance teams want visibility into recurring spend. Compliance officers want jurisdictional clarity. Security teams want tighter control over access, tenancy and resilience. Boards want fewer leaps of faith. Once AI starts moving into production, every one of those questions lands on the same table.
Nutanix says its Agentic AI platform is designed to address those concerns with “lower and more predictable token costs”. It says providers will be able to host multiple enterprises on shared GPU infrastructure with “strong tenant isolation and granular resource management”, while maintaining “predictable performance, security, and data isolation.” It also says Nutanix Cloud Manager will provide “usage-based metering, enabling providers to track and bill customers based on GPU usage, API calls, or model consumption.”
These are operational details, but the commercial consequences are plain enough. AI becomes difficult to justify inside a company when the cost model is opaque, the governance structure is loose and the data trail stretches across jurisdictions that legal teams would rather avoid. Regional and specialist cloud providers are trying to turn that discomfort into a market opening.
Nutanix tied that opening to a broader push across its platform. The company says organisations are modernising infrastructure while trying to balance “the flexibility of hybrid multicloud infrastructure and the need to maintain sovereignty of their data and applications.” It says Nutanix Cloud Manager 2.0 is generally available now, with multisite and multidomain management across multiple Prism Central instances. It says Nutanix Data Lens 2.0 is generally available now and can run “fully on-premises, including in air-gapped environments.” It also says Nutanix Cloud Clusters is expanding into AWS GovCloud now, with AWS European Sovereign Cloud planned for later in 2026.
Those details point to a change in the way cloud value is being judged. For years, competition in cloud infrastructure revolved around expansion: more regions, more services, more processing power, more spending. Enterprise AI introduces another test. Providers now have to show that they can operate inside legal, financial and operational boundaries without turning every deployment into a consulting exercise.
That creates room in the market. At one end sit the hyperscalers, with vast infrastructure estates and deep balance sheets. At the other sit companies and public sector institutions trying to make AI useful without taking on a fresh layer of uncertainty. Between them, space is opening for providers that can offer narrower service portfolios, closer customer relationships and a firmer grounding in local regulatory and operational realities.
Nutanix wants to supply that middle layer. Its own outline of the neocloud service catalogue reads like a condensed menu of managed AI infrastructure: “GPU-aaS, K8S-aaS, VM-aaS, Notebooks-aaS, VectorDB-aaS, and Models-aaS.” Business customers may well prefer to buy AI as a controlled service package with visible governance and measurable usage, instead of assembling a collection of components and discovering the economics later.
There is another force pushing in the same direction. Nutanix says hardware supply constraints are driving the need for more flexible infrastructure platforms. Its response is to widen the options around hardware, cloud and deployment models. The company says NKP Metal, which extends its Kubernetes platform to bare-metal infrastructure, is in early access now and due for general availability in the second half of 2026. It says Service Provider Central is also in early access and planned for general availability in the second half of 2026, with multitenancy aimed at service providers delivering hosted infrastructure and AI services.
The largest cloud operators will retain immense advantages for years. Their scale, reach and capital base remain formidable. Regional and specialist providers face their own hurdles, from margin pressure to the hard work of turning local trust into durable growth. Yet a gap has opened between technical possibility and institutional comfort, and that gap has commercial value.
Scott Sinclair, practice director at Omdia, says: “The deployment of autonomous agents is rapidly becoming the next frontier in enterprise AI, but this rise is introducing significant new risks related to data security, governance, and unpredictable performance. Organisations cannot manage this transformation on legacy infrastructure.
“Given these demands, Nutanix’s focus on strong governance, security, performance, tenant isolation, and predictable resource management in its purpose-built Agentic AI solution provides a welcome option for CIOs as they seek to deploy an enterprise-grade foundation for their AI agent strategy.”
* Arthur Goldstuck is CEO of World Wide Worx, editor-in-chief of Gadget.co.za, and author of “The Hitchhiker’s Guide to AI – The African Edge”.