Nutanix expands agentic AI

Nutanix expands agentic AI Nutanix is trying to move one layer up the artificial intelligence stack. Instead of only helping customers run servers, storage, and virtual machines, the company said on April 7, 2026 that it plans to add new capabilities to its Nutanix Agentic AI offering so “neocloud” providers can sell secure, governed artificial intelligence services to developers and enterprises in the second half of 2026. (nutanix.com) That shift matters because the market around artificial intelligence infrastructure is changing fast. A year ago, many cloud and hosting providers were mainly selling access to graphics processing units, the specialized chips used to train and run large artificial intelligence models. Nutanix is betting that raw chip rental is becoming a thinner-margin business, and that providers will want to package those chips with management software, security controls, and self-service tools that are easier for developers to consume. (siliconangle.com) (computerweekly.com) The key word in Nutanix’s announcement is “neoclouds.” The term refers to a newer class of cloud providers built around artificial intelligence workloads, often with heavy investments in graphics processing unit clusters and regional data centers. These companies are trying to serve customers that want more computing power than a traditional enterprise data center can provide, but also more flexibility or geographic control than the biggest public clouds always offer. (nutanix.com) (finance.yahoo.com) Nutanix’s pitch is that these providers need more than a pile of graphics processing units. They need a way to divide infrastructure among multiple customers, expose services through a portal, enforce policies, and let developers request resources without opening support tickets every time they want a model endpoint or a Kubernetes cluster. The company said its planned additions include a multitenant, multiservice portal and artificial intelligence management features designed for secure self-service use. (nutanix.com) In plain terms, Nutanix wants to help infrastructure providers sell an artificial intelligence platform, not just a machine in a rack. The company said neoclouds using the software will be able to offer graphics-processing-unit-as-a-service, Kubernetes-as-a-service, and an enterprise-ready artificial intelligence platform service built on Nutanix Agentic AI. That is a more ambitious product catalog than simply renting out compute by the hour. (nutanix.com) This announcement also builds on a product Nutanix had only just introduced a few weeks earlier. On March 16, 2026, at NVIDIA’s GPU Technology Conference, Nutanix unveiled Nutanix Agentic AI as a full software stack aimed at helping enterprises build, scale, and operate what it called “AI factories,” with integrations tied to NVIDIA AI Enterprise. In that earlier launch, Nutanix emphasized performance, compliance, security, and lower aggregate token costs. (nutanix.com) The April 7 update broadens that story from enterprise internal use to service-provider resale. Nutanix is no longer just telling companies, “Use our stack to run your own artificial intelligence workloads.” It is also telling cloud providers, “Use our stack to package and deliver artificial intelligence services to your customers.” That is a meaningful expansion because it turns Nutanix software into a potential revenue engine for providers, not only an internal operations layer. (nutanix.com) (siliconangle.com) The timing is not accidental. Nutanix made the announcement at its.NEXT conference in Chicago on April 7, 2026, where it rolled out a broader set of cloud platform updates around hybrid multicloud operations, storage, Kubernetes, and artificial intelligence. Multiple reports from the event framed the company’s strategy as an effort to present Nutanix as a complete platform for the “agentic AI era,” rather than a point product for virtualization or hyperconverged infrastructure. (nutanix.com) (virtualizationreview.com) (diginomica.com) There is also a competitive angle. Nutanix has spent years positioning itself as an alternative to older virtualization-centric infrastructure stacks, and recent coverage suggests it sees multi-tenancy and managed cloud capabilities as a way to win more service-provider and enterprise migration business. In that context, agentic artificial intelligence is not a side project. It is part of a larger attempt to make Nutanix the control plane for hybrid infrastructure, containers, and now artificial intelligence services. (crn.com) (siliconangle.com) The phrase “agentic AI” itself is doing a lot of work here. In vendor language, it usually refers to artificial intelligence systems that do more than answer a prompt once; they can plan, call tools, chain tasks together, and act with some degree of autonomy under policy controls. If those systems are going to be used inside enterprises or public-sector organizations, providers need ways to govern access, isolate customers, monitor usage, and keep data boundaries intact. Nutanix is packaging those governance needs as a platform feature. (nutanix.com) (financialcontent.com) That may be the most important part of the story. The industry spent much of the first wave of generative artificial intelligence selling speed, scale, and access to chips. Nutanix’s new message is that the next sale will be governed delivery: secure portals, policy controls, multitenancy, and repeatable services that let developers build quickly without giving up enterprise oversight. (computerweekly.com) (siliconangle.com) Nutanix has not yet shipped these new neocloud-focused features; the company said they are planned for the second half of 2026. So for now, this is a product roadmap announcement, not a launch with broad customer deployment data attached. But the signal is still clear: infrastructure vendors increasingly think the money in artificial intelligence will come from selling managed, policy-aware platforms on top of compute, not from compute alone. (nutanix.com)