From AI Ambition to Production Reality: Dell Technologies and NVIDIA on What It Takes to Operationalize Enterprise AI at Scale

Enterprises have already decided to invest in AI. The bigger challenge is turning promising pilots into production systems that deliver measurable business value at scale. Success depends on far more than the model itself. It requires the right infrastructure, data strategy, governance, and operational discipline to support AI in real-world enterprise environments.

At Dell Technologies World 2026 in Las Vegas, Patrick Moorhead and Daniel Newman sit down with Varun Chhabra, SVP of ISG Marketing at Dell Technologies, and Jason Schroedl, Director of Product Marketing for Enterprise Platforms at NVIDIA, to discuss what the next phase of enterprise AI demands. Drawing on announcements from DTW26, including advancements to the Dell AI Factory, new approaches to agentic AI deployment, and investments in data readiness, the conversation explores the building blocks organizations need to move from experimentation to production.

Varun and Jason also examine how enterprises can expand AI capabilities while maintaining control over security, costs, and operations. As agentic AI adoption accelerates, they share practical guidance on local execution, infrastructure strategy, and the priorities leaders should focus on as they move from AI ambition to enterprise-wide deployment.

Key Takeaways:

• The Dell AI Factory has spent more than two years moving beyond co-marketing into co-engineering, and that depth of collaboration is what enables enterprises to deploy AI with confidence rather than troubleshoot integration failures in production environments.
• Agentic AI demands secure deployment models, predictable cost structures, and local execution capability. Enterprises that cannot govern where agents run, what data they access, and what actions they can take autonomously will not be able to scale agentic systems responsibly.
• Data readiness is the most consistently underestimated barrier to enterprise AI success. Making data truly AI-ready, structured, accessible, and governed at the speed real-time agentic workloads require, is a prerequisite that most organizations have not fully addressed before deploying models.
• Enterprises want AI capability without surrendering operational control. The Dell and NVIDIA platform is designed to deliver performance and scalability across the full AI lifecycle without forcing trade-offs between capability and governance.
• The organizations that move forward with a clear prioritization framework will compound their advantage. The ones still debating where to start will find the gap harder to close as agentic AI moves from a strategic option to an operational requirement.
• The pilot phase of enterprise AI is ending. What separates the organizations that scale from those that stall is the infrastructure foundation, data readiness, and governance architecture they built while everyone else was still debating the models.
  

Watch the full video at sixfivemedia.com, and be sure to subscribe to our YouTube channel so you never miss an episode.

Disclaimer: Six Five Media is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded, and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors, and we ask that you do not treat us as such.

JASON SCHRODEL:
The future of AI is hybrid. We are seeing more and more demand for AI and also the opportunity to bring the AI to the data.

PATRICK MOORHEAD: 

The Six Five is On The Road here in Las Vegas at Dell Technologies World 2026. It's been all about agentic enterprise AI and that really folds into really the theme of 2026, which is making AI real for enterprises.

DANIEL NEWMAN: 

Yeah absolutely. Pat it's been a big focus the last couple of days. You're bringing not just new technologies out but bringing customers through hearing from them. How exactly are they taking these investments these technology spend all the ambition and all the proof of concept that they've been spending their time and building around for the last few years. How are they turning this into enterprise value that can be measured.

PATRICK MOORHEAD: 

That's right. And I think the most impressive thing that I've seen all week is some of these model makers going on prem. Yeah. You know Google had been there for a while with GDC. But you know we've got OpenAI Palantir and who knows. Oh XAI. Yeah. And who knows. Maybe we'll see Anthrop again there too.

DANIEL NEWMAN: 

Yeah I mean over time I think what eventually happens is the landing will find itself somewhere in between the cloud and the prime. That's right. We saw that in the last era. You know I think we've done a few victory laps over time when everyone said everything's going to the cloud. And you know we were like. I don't think that's how it's going to end up working out. And in the end it's worked out very hybrid. And I think we're seeing that now with models. And it's really a it's a state and it's an era of abundance. All types of heterogeneous compute all types of memory, all types of architectures all types of software. They're winning because the demand is just so significant.

PATRICK MOORHEAD: 

Yeah, it is. By the way, it's also been the Dell and NVIDIA show here. It sure has. If you haven't noticed, I was with Jensen on stage. Just a lot of co-engineering that went on. And with that, I'd love to introduce Jason from NVIDIA, Varun from Dell. Great to see you guys. Thanks for having me here. Great to have you on the show. Thank you for having me again. It's always a pleasure. Absolutely.

DANIEL NEWMAN: 

Varun, it's been like your 400th time with us? Who's counting? It's always fun. Yeah, it is good to have you. So look, it's been a few years since you launched the Dell AI Factory, and it's been a massive success. Just listen to those backlog numbers. Just listen to the growth of NVIDIA. But what makes this partnership fundamentally different? As an analyst, we attend a lot of events. We sit in a lot of rooms. We listen to a lot of executives. Many have partnerships. with Nvidia someone like what is really at the root of this alliance that makes it so different and that it's become so important for your customers.

VARUN CHHABRA:
Absolutely. I think there's a few things that make this a really really defining partnership for the A.I. in the enterprise era. from my perspective or our perspective, there is a strong foundation of trust that the two companies have that comes from working together for a long, long time. Like we're talking 20 plus years, right? You can see it in the engagement that Jensen and Michael have on stage. And it goes to every level of the organization, including the engineering teams. So I think that's one. Second one is, I think there is a shared understanding that amongst the companies that for AI to be successful in the enterprise, it truly needs to be end-to-end. It needs to be simple. It's not just about compute. It's about compute, data, networking, the solutions on top of it. It's the hardware and the software working together. So I think there's a joint shared philosophy there. And it's no accident that Dell AI Factory with NVIDIA was announced actually not at Dell Tech World, at GTC by Jensen two years ago. We have 5,000 enterprise customers now. It's such a big number that sometimes you have to just take a step back and be like, it's crazy, 5,000 customers. Wasn't it like 52 years ago or something? No, but when we just announced.

DANIEL NEWMAN: 

Well, I mean, when you announced it.

VARUN CHHABRA: 

Yeah, I mean, and it was 4,000 a quarter ago. That kind of gives you a sense of the acceleration we're seeing. And it only happens because the two companies are working together, product, go to market, joint storytelling, all of those things.

JASON SCHRODEL: 

Yeah, no, that's absolutely right. We're thrilled with the partnership and the momentum we've seen. I've seen some of the customers that were showcased on stage, like Lilly, Honeywell, Samsung, Hudson River Trading, just fantastic customer success momentum. And it's because of that deep partnership at every layer of the organization, from our product engineering teams out through the go-to-market, and at every layer of the stack. So NVIDIA bringing the accelerated compute, the networking AI software, and then Dell bringing the infrastructure, the integration, the services, the deployment know-how to make it work in the enterprise to deliver business outcomes. It's that combination of the technology and the deep expertise and enterprise relationships that make this partnership successful.

VARUN CHHABRA: 

Jason did an amazing pitch for Dell. That tells you how strong our partnership is.

PATRICK MOORHEAD: 

Yeah, I mean, you made a lot of tangible announcements, right? This isn't just ambition. And a lot of this is doing AI in areas, you know, a lot of people think it's cloud only, right? But what we laid out is basically a hybrid model, including even client computing. Can you talk about some of the specific announcements that the two of you made together? Jason, I'll start off with you.

JASON SCHRODEL: 

Yeah. I mean across the board we touched on some of the announcements earlier with some of the agents and models coming on prem. And I think that's a phenomenal change that we're seeing now that really just recognizes that the future of A.I. is hybrid. We are seeing more and more demand for A.I. and also the opportunity to bring the A.I. to the data.

PATRICK MOORHEAD:

Yeah.

JASON SCHRODEL: 

And the reality is 70% of the enterprise data is on-prem. So by bringing the AI to the data, you're going to get much better business outcomes. But from an announcement standpoint, it was really across the entire Dell AI factory stack. I mean, from the desk side all the way with GB10, GB300 systems, and the new desk-side agentic AI solutions, all the way up to the VeraRubin MVL72 massive scale infrastructure. And solutions that we've built with the AI ecosystem, including some of these foundation model builders, but also ecosystem software partners out there. We've talked about Palantir as an example, but there are a wide range of different business outcome-led solutions that Dell is now bringing to market with these AI leaders and pioneers in the industry. So it's a fantastic refresh of the overall Dell AI factor with NVIDIA that was announced this week.

VARUN CHHABRA: 

And I think the complement to everything Jason's talking about is the data piece, right? You can't really, you're seeing this, we're seeing this. Enterprise AI going from POCs, success rates of going from POCs to production comes down in large part to whether you get your data strategy right, are you getting your enterprise-wide data fed into your models that are now coming on-prem? So what we've also announced on the Dell AI Data Platform side with NVIDIA, working closely on accelerated data ingest, you know, QVS, QDF, KVCache, Dynamo, you have a deep integration at every level with the Dell AI Data Platform stack, whether it's the storage engines, whether it's our orchestration engines, et cetera. That's a huge part of the story as well. It's not just about compute. As Jason said, it's compute, storage, networking, software, and the services all jointly published together.

DANIEL NEWMAN: 

Lots of announcements. I think that storage inflection is still coming. I think we're just at the very beginning where as these agents spin up and all the data gets created I don't think people are fully appreciating how much more storage we are going to need. Jason I want to double click a little bit on the data comment. You know. You guys talked a lot about data and its challenge in terms of enterprise AI success. And then you talked a little bit about what you announced. But talk a little bit about why what you announced is really going to make enterprises more ready because everything in terms of scale is getting bigger. Everything in terms of time to value is getting faster. And you know unlocking the data has got to be the key.

JASON SCHRODEL: 

Yeah. No you're absolutely right. I think that's where we feel like like data is the fuel for an AI factory. And that's why that you know getting that foundational data platform right is key to having success in AI going forward. So some of the innovations that we introduced around the AI data platform with the media are include things like the evolution of the data orchestration that was built on the data loop acquisition, but also some of the new data engines with, you know, QVS libraries for unstructured data, QDF for structured data, and some of the ecosystem partners, for example, like Starburst and Elastic and others, we've built on top of that. A lot of that innovation is around making sure that that data is easily accessible, but also that it's secure and available for enterprises as they scale out their enterprise AI and now agentic AI initiatives.

PATRICK MOORHEAD: 

So how holistic is that data, meaning does it incorporate SaaS as an example?

VARUN CHHABRA: 

Yeah, I think it's a work in progress. The data estate is so big. But our goal is to really, with what we're doing with the orchestration capabilities in the AI data platform, and the ingestion and processing capabilities, is to bring data across the corporation, across the enterprise, whether it's sitting on a Dell storage device or not. The goal is to really bring all of it together. As Dave Morin said today in the conversation with Jeff, Agents need context. And context is just another word for data, right? And as we look for agentic AI to become more autonomous and for that autonomous operations to be more accurate, more successful, the difference is going to be the data piece. So bringing data across the enterprise, that's really the goal of the AI data platform. Of course, we hope most of that data sits on Dell storage platforms. But even if it doesn't, we want to bring all of that together for customers.

PATRICK MOORHEAD: 

Yeah, that is a key. I mean, everybody wants to go. I mean, I see, you know, some of the cloud players kind of doing it. Hey, it's kind of all or nothing. Yeah, we're building this. And yeah, we might have a, you know, something out here that you can connect. But I think in the end, enterprises have made their choice. And listen, I'm equal opportunity, I love the public cloud, I love on-prem, colo, sovereign, whatever. But we're 15 years into the public cloud and 75% of the data, enterprise data is still on-prem. It's different for consumer, but very much so. And we predicted a long time ago that that's how this would end up. The good news is is is that again we've got the core model makers. Yes. Your key. Yes. ISE fees coming in and participating and participating in that.

JASON SCHRODEL: 

And that's I was just I think the other thing that's that's key here is is agentic. I mean it really has changed everything. So I think we are seeing just a phenomenal step change in terms of the enterprise adoption and the token consumptions with agentic. And that that really has changed the game. I think we're now like a lot of the innovations that we announced this week are based on preparing for the era of agents. Yeah. And from again from the desktop that side to the data you know all the way up to the rack scale architecture in the factory is preparing for this age of the agents where again you know part of this is is responding to the demand you know the consumption. Yes user level. But also part of it is brain for the economics of it. You know as you think about the scale that the classical Jevons paradox is we're seeing the you know the token cost per token decrease. We're going to see this consumption just go parabolic. Yes. And so that man is just going to skyrocket. So you would have to look at the economics. Absolutely. Of your of a particularly as as these agents take hold within the enterprise.

VARUN CHHABRA: 

And it's no accident that all these large frontier models are coming on-prem because they're seeing the same thing, right? It's like one feeds into the other. So, yeah, absolutely. The world will be hybrid, but we think that a large part of it will be on-prem because customers want that. They want to bring AI to their data, not the other way around.

PATRICK MOORHEAD: 

That's right. So guys, we have time for one more question. And you know, this AI thing is so complex. And even in the last two days, just the amount of new information coming to your customers is amazing. And Varun, I'll start with you. What's the one thing that you want enterprise to take away throughout the future of AI?

VARUN CHHABRA: 

It's a great setup. I think so many announcements, but the one thing I would say is what Dell and NVIDIA have been working on the last two years, and the reason why the AI factory with NVIDIA is so successful is that we are focusing on simplifying things for the enterprise. So one thing we didn't talk about, We announced this ecosystem program jointly where we're bringing in ISVs, software providers, et cetera, model providers from across the ecosystem. And we're using the Dell Automation Platform, NVIDIA Blueprints, all of these things together to make it easy. Because customers are ultimately looking for use cases. They're not just getting up in the morning. They are getting up in the morning and saying, hey, I want to do AI. But the successful Pilots going into productions are places where there are specific use cases identified. There's a stack that they're working with. Bringing that software and hardware stack together is not easy. So the automation that we've built into this to actually make this as turnkey as possible is a huge part. And it's all parts of the journey, data, processing, whether it's tuning, et cetera, all of those things. That's what I would love for enterprises to really look at. I think once you look under the hood, the choice is clear. It's the Dell AI factory 

with NVIDIA. Yeah.

PATRICK MOORHEAD: 

So Jason, how about you? Anything to add to that? What's the one thing? I would say it's OK. You can agree with Varun.

JASON SCHRODEL: 

You can agree with Varun. So we think alike. The one thing I will say is the time is now for the enterprise. And enterprises need to prepare for the era of AI agents. And frankly, this tsunami of token consumption that's going to happen within their organization. So they need to prepare. And we're frankly that you know I think Jeff Clark did a great job outlining it this morning and the day to keynote where you know the companies that are successful in the future are those they're going to be a native native. It's not a birthright. Yes it's an operating model. So companies need to embrace that now and prepare for that change because it is it's coming in the enterprise and they need to be ready.

DANIEL NEWMAN: 

Yeah. Yeah. Yeah. I think that's a great point because in the end we're so incredibly early still. And while there is immense pressure into all the enterprises out there that are that are looking I mean I saw a stat recently about 2 percent of of the consumers are paying for AI right now. Any of it. And about 20 percent of businesses. You know we live in this kind of world. It's a vacuum. Every day we wake up and it's like oh my gosh AI is taking over the world.

VARUN CHHABRA: 

And it is. Yeah.

DANIEL NEWMAN: 

But in terms of if you're an enterprise is a big spectrum and you're not down the road there's still it. Yeah. The time is still there and you can turn on this operating model. And because of the power and the capabilities of these models you can actually go really fast. Yeah.

VARUN CHHABRA: 

Once you decide the choice you. It's a choice. Exactly. It is a choice.

DANIEL NEWMAN: 

I think that part is really important because I would argue yes in a positive way. The technology that has been built by companies like Nvidia and Dell is world class. Yeah. It is capable of delivering agentic experiences that work across very targeted models and across very broad use cases. But having said that, you do have to choose your culture to be an AI-led, AI-native culture. Gentlemen, I could talk to you all day. Varun, Jason, so much fun. Happy Dell Technologies World. Let's have a conversation again soon. Thank you so much for having us here. And thank you everybody for being part of this 6 5 on the road. We are here in Las Vegas Nevada at Dell Technologies World. We got to step out for a little bit. Stick with us. We'll be back.