Latest / Technology Now / The great VM reset, hybrid AI, and agentic enterprise: rethinking infrastructure in 2026 | Patrick Osborne
Transcript
- MICHAEL BIRD Well, good morning, Sam. Good morning. in 10 seconds or less, what are the biggest trends or changes we are seeing in the world of technology? I guess specifically enterprise technology today. Go. SAM JARRELL the great VM reset. People wanna change how and when, how much and when they consume, virtual machines. the AI gold rush, the rush to extract value out of AI rather than just plugging AI into everything. and then the security and resiliency imperative. Security can no longer be bolted on after the fact. It has to be built in from the ground up and assumed from the inception of a product. And then resiliency, backup. it's not if you will get attacked, it is when MICHAEL BIRD Sam, I knew I could rely on you. you're very succinct with it. Well, today we're gonna be doing a bit of a roundup of various topics because we talk about all of these on quite a lot on the podcast, but none of them exist in isolation, and I think you'll agree it is all connected SAM JARRELL I do agree MICHAEL BIRD I’m Michael Bird SAM JARRELL I'm Sam Jarrell MICHAEL BIRD And welcome to Technology Now from HPE. MICHAEL BIRD So Sam, we're chatting about some of the big shifts happening in 2026 and beyond, I guess, in the world of enterprise technology SAM JARRELL Yeah, and to be honest, if we had made this episode at the start of the year, it would probably look completely different too MICHAEL BIRD Yeah, absolutely. We probably need to get this episode out quite quickly because things are changing so quickly, aren't they? there's like the huge rise in agentic AI and the increasing focus on sovereignty. Both organizations and even governments are rethinking how they run their infrastructure SAM JARRELL Yeah, and all of this is happening simultaneously too. We're talking about all these shocks kind of hitting at once, AI, sovereignty, security, the works MICHAEL BIRD Yeah, they are. And as a result, we're seeing a couple of mega trends popping up across all the different parts of the industry. so to find out more, I spoke with Patrick Osborne, SVP in Hybrid Cloud and Office of the CTO. And, to kick us off, I asked him what the mega trends are in 2026. So Sam, it's time for us to find out if you were right. SAM JARRELL Let's do it PATRICK OSBORNE the first one is the great VM reset. So everything changing in the virtualization landscape. the second, is, people's pathway to AI, both deploying it, making it scale and work for them in the enterprise, and then we talked considerably about the path to an agentic, AI enterprise. And then lastly, which is like always top of mind, is around the security imperative. So security, resiliency, et cetera MICHAEL BIRD So the great VM reset that you mentioned, why are we talking about it now? I thought, virtual machines was, something we talked about 15 years ago, and that problem was solved. We knew how they worked. We knew how it was licensed. Yeah. Why are we talking about it now in 2026? PATRICK OSBORNE Yeah, so for the customers, there's been quite a bit of consternation in the market, Specifically around economics and some relationships with their vendors. And so they're being forced to rethink not only the finance and the economics around it, but the technology set, right? And it kind of bleeds a little bit into the AI portion as well too, which is anytime you have a forcing function and they have the ability to rethink what they're gonna do, whether it's for virtualization, now they have to start thinking, "Am I ready for these AI workloads to be running in the same data center, like co-resident with all my new, opportunities around virtualization?" So it's a discrete pain point that's going on right now, but it's making people rethink what they're doing for infrastructure, uh, in the virtualization space, and then once you open that up, am I ready for AI as well? MICHAEL BIRD Yeah. what are the big challenges with, the great VM reset? why is it a challenge today? PATRICK OSBORNE Yeah. So, on the technology front, right? the challenge for customers is the virtualization space and those workloads are growing, But then I'm starting to introduce a whole, number of new workloads in, right? Containerized services for running AI applications. I'm gonna be deploying agents, right? To help those workloads run more efficiently at scale. So there's the technology implications of it, and then the second is, you go into a room and you ask people, "Raise your hand if you're getting an increase in your budget," right? Very few raise their hand. So not only are you trying to mitigate what's going on in virtualization, costs are going up, commodity costs are going up. You also wanna create headroom to be able to invest in all these new fantastic AI and agentic enterprise technologies. So you gotta do more with the same or more with less, right? MICHAEL BIRD and the way that we did virtualization 10 or 15 years ago, I mean, the technology's largely the same, but what we're asking of that technology is different PATRICK OSBORNE Absolutely. You're asking for more efficiency in the infrastructure layer. You're asking for the ability to run multiple run times, right? So this isn't a point in time rip and replace. Customers have these journeys they go on for six months, a year, two years to get to a new architecture. So it's not just a point in time. It's definitely, a complicated journey for everyone. And so, it's very difficult for them to mitigate, and understand how they make decisions today, how are they gonna play out in the future MICHAEL BIRD Can we talk hybrid cloud? Because, I want to understand how hybrid cloud is changing in the era of AI PATRICK OSBORNE Yes. Yeah. So there's a couple implications on hybrid cloud. AI is definitely a hybrid workload, right? So, as we all know, there's, AI services that people use at the commercial consumer level, right? There are models and services people use in the enterprise to help with engineering, coding, content creation. but one of the things that's difficult for customers to understand is the data, Data has a certain gravity to it. Data needs to be protected. It needs to be sovereign in some cases. it's like the digital, capital of, what most customers are doing. And so being able to use AI, whether it's in the public cloud, or some of the neo clouds, or being able to run those services on-prem, which is co-located with, customers' probably most mission critical and most important data it's a hybrid workload at the end of the day. And there's a lot of implications on the technology. Do you have a network that can support that kind of, performance, security, multiple places this data's gonna live, in the cloud, on-prem at the edge? and so, there's a lot of choices customers need to make MICHAEL BIRD isn't the big thing and perhaps one of the challenges with AI is that, it requires, serious amounts of data to be useful. And the challenge with cloud environments is that often the data is far away from where it needs to be used and maybe in a different location to where the computer is. So, how does that work in the world of hybrid cloud? PATRICK OSBORNE Yeah, so we, have a number of, capabilities that we allow customers to do these AI workloads in a hybrid manner, So starts with the data layer, right? So we have some scalable systems and architectures that help people, specifically for data for AI workloads, right? It's a lot of unstructured data or semi-structured data. There's new protocols involved. a lot of the AI actual services are being embedded into the storage layer now, so data indexing, data queries, data tagging, right? and then not all data is new, right? So we have the capability to, build, a data fabric for customers so that you can have newly created data. You can import, brownfield datasets, whether they be databases or semi-structured, and have access to that, secure access. And then we have a number of technologies we deploy at the networking layer, right, to provide secure, high-performant connections into where that GPU-accelerated compute is gonna sit, And so, it is a hybrid workload, and we have a number of the components of the stack that help customers do it in a efficient and secure manner MICHAEL BIRD 'Cause you don't want those GPUs idling. You want them being used- PATRICK OSBORNE All the time. MICHAEL BIRD chat all the time PATRICK OSBORNE chat. and I think a lot of the implication for customers too is that when they're gonna run these services in their traditional data center, some of the data centers don't have the capability of doing this. Once you move from air-cooled into liquid cool, do you have the power, capable of running these, right? So oftentimes customers are asking: Can I, use a co-location facility for this, but I need secure access into my data, et cetera, et cetera. So we provide that for our customers. MICHAEL BIRD Which leads quite nicely onto, sovereign, sovereign AI, sovereignty. what's your perspective on it in this world of AI that we live in? PATRICK OSBORNE Yeah. So sovereign is definitely a loaded term, right? It means a lot to a lot of different people, depending on their, the customer and their use case. But sort of simply put, we have customers, whether it's through, regulation or compliance, they need to understand physically where their data is, where the infrastructure lies. they also want providence over the people that are actually managing and supporting, that infrastructure. They have to be oftentimes like, you take the EU, right? That infrastructure needs to sit in a specific country. They need to have, folks that are citizens of that country working on the data, also, audited and secured. MICHAEL BIRD And disconnected is disconnected from the internet, disconnected from- PATRICK OSBORNE everything. Yeah. Yeah. I was talking to some partners this week. They're, you know, talking about, how do we get, AI services and models up into the International Space Station, MICHAEL BIRD Wow PATRICK OSBORNE right? So it's like the ultimate lazy connection, up there. So there's a lot of challenges in how I can do that in different manners MICHAEL BIRD so we've talked through the mega trends. in fact, the last thing I guess was agentic enterprise. Yeah. Can you just talk through that a little bit? PATRICK OSBORNE Yeah. So we started this journey, here in HPE with our GreenLake cloud platform about six years ago, and when we created the platform, it was to give a better product and operational experience, for our customers. So it was often, how can you do orchestration, get telemetry on the products, do predictive analytics, et cetera. And then with the rise of agentic, we feel that as a vendor, you need a platform to be able to really take advantage of agentic AI. So all of the discrete portions of your infrastructure stack will have agents, right? Agents will be able to register. You can create an agentic mesh and then actually start to build agentic services, right? So a good example is, you know, we're in the process right now of migrating, tens of thousands of virtual machines that support our development environment, and you can't do that with a pick and shovel, right? It has to be automated. So we created, an agentic service that does virtual machine migration for us. So we're providing those type of, user experiences, and they can only happen if you have a platform, right? A platform that has all the levels of the infrastructure registered, managed, in our cloud platform, and then the ability to, have agents, put guardrails around that agents, create a, service mesh, MICHAEL BIRD Okay, fast-forward to 2030. What do you think are gonna be the biggest, changes or biggest trends that we're gonna see in 2030 in the tech industry? PATRICK OSBORNE 2030 MICHAEL BIRD It's not that far away. PATRICK OSBORNE I know, I know it's just, it's wild that we're already talking about 2030. But, I think there's gonna be a couple things. there's a lot of talk right now around, you know, the AI market is pretty frothy, and some people put the term bubble out. But I think we are just getting started in terms of physical infrastructure build-out, data centers, access to power, green power, and so I think there's a lot to be learned there. I think, for customers, I think the agentic AI piece is gonna be... It's gonna take off, right? Because the one thing that is very difficult is to get time and people's time back, right? So the more automation we can introduce, those folks can do less, grunt work, repeatable, operational work, and focus those resources on what's differentiating you as a company. So the agentic piece I think is just getting started, and it's gonna take off. and so I think there's that. And then, I think it's not a technology problem, but it's gonna be, more of a regulatory, stance, how these things are gonna be, regulated by individual sovereign governments, right? how they, secure their data, who has access to which models and which services. and so like many other paradigm shifts in the past, the technology is outpacing the ability to control policy, right? And so I think that's gonna be, pretty prevalent over the next few years. MICHAEL BIRD Amazing. Well, Patrick, thank you so much for your time on Technology Now PATRICK OSBORNE Yeah. Thanks for having me. SAM JARRELL Well, I think based off of that conversation that, uh, I was right, right? MICHAEL BIRD You are spot on. You are spot on, Sam. SAM JARRELL Thank you. Thank you. But I did appreciate, you guys having the double-click conversation into my, rapid 10-second delivery of these very high-level topics. What really stands out to me is that none of these trends seem to happen in isolation, and, organizations aren't just dealing with, the great VM reset. They're not just trying to capitalize on AI gold rush, and they're not, also just trying to navigate the security resiliency imperative. They have to do all of these things at once, whether or not they, they want to. They have to either try to cope or capitalize, across all three of these things, and it seems to me like this is just gonna continue. Even in the closing when he was talking about 2030 stuff, it just seems like that's the acceleration of what's already happening right now. To me, it feels like technology planning is more complex today than it was even, like, two or three years ago. MICHAEL BIRD there are so many things that maybe 10 years ago we thought, "We've solved it. We don't need to worry about that anymore." And then they're resetting themselves. The great VM reset. Like, organizations were like, "Great, I understand virtualization. It's just a license I now have to pay. Easy peasy." But now it's changed. And AI has also changed the way that our organizations are having to think about the basics. we did the episode recently on security, where we talked about how AI is just impacting the way that we think about security, and it's just completely shifted how organizations, how software developers, how hardware developers think about security. we're in this phase where, everything is interconnected, but also the stuff that we could rely on maybe 10 years ago, we now can't necessarily rely on that, and so I wonder if organizations are really having to think more holistically about their infrastructure. SAM JARRELL I think so. I think, discussions that were once, cost discussions on, an annual basis have become full-on strategy discussions. MICHAEL BIRD Yeah. Hmm SAM JARRELL it's kind of funny to me that many of these organizations were viewing a lot of this stuff as this is dead set, this is done, we don't have to worry about this anymore. When with the rapid pace of technology, nothing is a, guarantee, right? It seems like you have to constantly be keeping up with what the latest and greatest is to ensure that you're prepared for what's next. MICHAEL BIRD infrastructure outside of technology. if you think about, bridges or roads or railway lines, there are railway lines in the UK that were built in the Victorian era. So, you do think of infrastructure as something you put in, and you're not necessarily replacing it all the time. it's something that's, it should be part of the fabric. And so, with the rise of AI, I think the way that we think about our infrastructure, both hardware and software infrastructure, I think is, has changed. and then it's had this, knock-on impact across multiple, all these multiple different things that we maybe wouldn't have even thought would, would be an impact back in 2022 when AI really sort of hit the public consciousness, hit the boardroom, so to speak. and often it's like big financial institutions or, supermarkets or whatever because all their infrastructure runs 24/7. if you imagine banking infrastructure, you can't bring it down. and so it's almost cost prohibitive to change it. Obviously, we'll have to change it at some point. MICHAEL BIRD anyway, we digress. let's go back to the mega trends because we talked about what the mega trends are, but I wanted to end on something directly practical which our listeners could do. So the final thing I asked Patrick was this: What advice would he give to people who are thinking about adopting AI into their work or into their wider infrastructure? PATRICK OSBORNE What advice would I give to people adopting AI? So there's the technology piece of it, right? which you gotta dive in, right? Everyone's gotta be comfortable with the tools. It's a very fast-moving space, right? So now's the time to jump in. Not tomorrow, not next week. I'd say the thing that people don't realize too is that there's a huge cost, in the management of change, in the way people actually do their jobs on a daily basis. So, process redesign, process change, leading from the tops down on, getting folks to adopt AI and augmented, AI services in the way that they do their workflow on a daily basis. So I think one of the things is people shouldn't discount the amount of cultural and, process friction this can create. And, definitely lean in there and, it becomes a culture of a company at the end of the day and an enterprise. SAM JARRELL Okay that brings us to the end of Technology Now for this week. Thank you to our guest, Patrick Osborne, And of course, to our listeners. Thank you so much for joining us. MICHAEL BIRD Yes, and if you've enjoyed this episode, please do let us know. Rate and review us wherever you listen to episodes. And if you wanna get in contact with us, send us an email to technologynow@hpe.com, subject line, "Megatrends." Oh, oh, cowboy. "Megatrends." I'm hoping producer Harry can add a bit of reverb to that just to sort of make it sound more epic. Technology Now is hosted by Sam Jarrell and myself, Michael Bird This episode was produced by Harry Lampert and Eva Higginbotham with production support from Alysha Kempson-Taylor, Nik Damarell, Arlyne Chin, Beckie Bird, Alissa Mitry, and Jenessa Ayache. Our theme music was composed by Greg Hooper. SAM JARRELL Our social editorial team is Rebecca Wissinger, Judy-Anne Goldman and Jacqueline Green and our social media designers are Alejandra Garcia, and Ambar Maldonado. MICHAEL BIRD Technology Now is a Fresh Air Production for Hewlett Packard Enterprise. (and) we’ll see you next week. Cheers! SAM JARRELL Bye y’all