At the enterprise level, agentic AI has become less of a prospective tool for businesses, evolving into a valuable asset for many organizations.

As more companies shift from the experimentation phase and begin to scale agentic AI use, their infrastructure requirements are changing. Enterprises are now looking to integrate AI into existing infrastructure, creating hybrid environments that raise new questions. Cost, reliability, operational demands and governance are now the priorities, and companies are having to determine how to balance it all.

“Most enterprises don’t have an AI ambition problem,” said Sam Grocott, senior vice president of product marketing at Dell Technologies Inc.. “They have an AI execution problem.”

Dell is one company investing in a strategy that helps enterprises manage the move toward hybrid infrastructures built to support agentic AI. Following Dell Technologies World earlier this year, that strategy is becoming increasingly relevant as companies move past pilot programs.

This feature is part of SiliconANGLE Media’s ongoing coverage of enterprise AI infrastructure and the shift from AI experimentation to production. (* Disclosure below.) 

Enterprise data remains at the forefront of priorities

Companies are already sitting on critical data that fuels their existing operations. However, the shift toward a hybrid model that integrates AI is forcing them to rethink how they use and store that data.

“The vast majority of data still resides on-premises, and being able to bring the compute and the data management and the orchestration and the automation to where the data lives is not only cheaper, it’s more efficient,” said Travis Vigil during theCUBE’s coverage of Dell Technologies World. “Customers, more than ever, need to turn the data that they have [into results]. That’s the oil — everything else is just piping to get value from the oil, and the ability to leverage that in place is critical to our customers.”

To address this need, Dell has begun focusing on the development of solutions, such as its AI Data Platform. This is helping businesses bring together their storage, security and data management capabilities to support AI workloads across enterprise environments.

“From an application development perspective, the competitive advantage is enabling developers to access trusted, governed data consistently across the AI lifecycle,” said Paul Nashawaty, principal analyst for theCUBE Research. “As enterprises move beyond pilots, the organizations that connect their data infrastructure to production-grade application delivery will be better positioned to turn AI investment into measurable business outcomes.”

That focus has continued since Dell Technologies World. Earlier this month, Dell announced additional AI Data Platform capabilities designed to give AI applications and agents more consistent access to enterprise data, reinforcing the importance of data readiness as organizations move toward production.

Agentic AI creates new operational demands

As hybrid infrastructures evolve at the enterprise level, agentic AI is becoming increasingly important to businesses. Because it can help automate long-running, complex processes, agentic AI can offload tasks from employees, freeing up time for other priorities. Agentic systems, however, are creating new priorities for companies that are moving from the experimental phase to production. Continuous observability and security are now critical to enterprises that need to closely manage their data. Governance and orchestration also require oversight as engineers manage increasingly autonomous systems.

For these reasons, one approach involves developing an AI environment that prioritizes building and operating autonomous agents from the start. Rather than attempting to add AI capabilities to existing infrastructure, more enterprises are considering infrastructure designed to support AI at the foundational level. Dell’s AI Data Platform is one example of how solutions are now being designed to connect enterprise data with AI infrastructure.

“The biggest impediment to navigating the legacy environment into the AI era is this mistake of thinking you’re doing it by having the legacy environment be the anchor,” said John Roese, global chief technology officer and chief AI officer at Dell. “The only way it works is [if] you actually have an AI environment. You have to have a place to build agents and run them autonomously. You have to have a security architecture that works with things like agentic. The brownfield becomes a feeder system to the new AI environment, but if you don’t have the AI environment at all, you’re just putting some AI features around a brownfield.”

The cost of moving from AI experimentation to production

With companies moving from experimentation to production and scaling AI deployments, they are also shifting their focus toward proving the return on investment of agentic AI. At the same time, it’s becoming critical to demonstrate that their hybrid infrastructures can be supported economically. For this reason, cost remains a pain point for enterprises scaling AI workloads. Token consumption, inference costs and data movement are all factors that businesses must now take into consideration when deploying AI. There are also more financial decisions surrounding the use of the public cloud, private cloud, edge and endpoint environments and data centers.

Dell has been investing in financial services capabilities to get ahead of this curve. Through its financial services arm, the company has positioned itself to help customers finance infrastructure investments needed to support AI deployments.

“We’ve booked $64 billion of AI demand last year, but it’s accelerating. It was actually $34 billion in Q4 alone,” said David Kennedy, executive VP and chief financial officer at Dell. “When you look at that, you can see the appetite that’s there. We have a financial services arm — we’ll do our due diligence and the durability of that growth is the opportunity. There’s many ways to kind of source and provide that capital to the tremendous buildout that’s out there.”

Dell has also invested in a strategy around a disaggregated model that separates compute and storage, allowing each to scale independently.

Dell’s cost-saving infrastructure strategy

As Dell sees more companies look for AI infrastructure that moves “from the desk side to the data center, to the edge, and all the way out to the cloud,” as Grocott puts it, it’s making moves to meet these needs. In collaboration with Nvidia Corp., the company is helping enterprises that do not want to run their AI agents exclusively in the cloud, but rather locally. Deskside Agentic AI is helping businesses achieve this goal, offering several potential benefits. To start, companies can address some data security concerns by working with their information in-house, reducing reliance on cloud AI services. The approach can also help control costs that would otherwise be spent exclusively on external data centers or third-party cloud services.

With more enterprises scaling their agentic AI use, they are also beginning to focus on workload placement. Data locality and latency are two factors influencing these decisions as companies look to ensure they are spending wisely. Security, governance and sovereignty requirements – specifically in industries such as healthcare – are also dictating workload placement.

“The PC is a great place to start,” said Jon Siegel, SVP of Dell portfolio marketing. “When you have a free token generator, you give your employees a free place to experiment, to innovate, and then when they want to run into production, they could then run it in another spot like in the data center or in the cloud.”

By making a broader shift toward placing workloads on the appropriate platforms, companies are also positioning themselves for long-term cost savings. Azure Local is one example of a solution that is helping businesses adopt a hybrid approach while keeping selected workloads on-prem.

“What has worked in the past won’t necessarily work moving forward,” said Kenny Lowe, technical staff member and cloud platforms evangelism and enablement lead at Dell. “We need to be able to right size workloads in the right way on the right platform so you’re not overpaying for the infrastructure that you’re running on. The core foundational model of the data center is shifting — shifting away a bit from hyperconverged to a more disaggregated model.”

Dell Private Cloud is another example of a solution that is helping companies choose the appropriate platforms for their workloads. With PowerStore Elite, for example, Dell’s clients can take advantage of storage performance and data reduction capabilities. Combined with Dell Private Cloud’s disaggregated architecture, these capabilities give customers greater flexibility in managing compute and storage resources, according to Dell. By putting more infrastructure control into the hands of its customers, for instance, the Dell is providing companies with more autonomy over their hybrid infrastructures.

“The reality is that Dell Private Cloud, with disaggregated infrastructure, with PowerStore now, has a six to one data reduction — you have up to 65% acquisition cost savings,” said Caitlin Gordon, SVP of product management for private cloud and AI solutions at Dell. “On top of that, the private cloud software gives you the operational savings. It’s really a cost benefit analysis that is undisputed for every one of our customers. We’re really trying to help them get out of the handcuffs of HCI and get [them] into the freedom that they need with Dell Private Cloud.”

Dell’s broader AI infrastructure strategy comes into focus

Since Dell Technologies World, the company has continued to introduce capabilities addressing enterprise pain points linked to scaling agentic AI. Businesses are now faced with not only determining where their data should reside, but how to continuously govern these environments. Through solutions such as the AI Data Platform and Deskside Agentic AI, Dell is showing how its infrastructure strategy is developing in response to customers’ needs.

Infrastructure economics, workload placement and the management of agentic AI operations will likely remain at the forefront of enterprise concerns as organizations move into the production phase. Dell’s strategy reflects the growing importance of connecting these priorities as enterprises build the infrastructure needed to support AI at scale.

(* Disclosure: TheCUBE is a paid media partner for Dell Technologies World 2026. Sponsors of theCUBE’s event coverage do not have editorial control over content on theCUBE or SiliconANGLE.)

Image: SiliconANGLE/ChatGPT