Jason Kelley, IBM

More from MWC Barcelona 2026

Jason Kelley

General Manager, Blockchain Services

IBM

Info
Transcript

Jason Kelley, IBM

Jason Kelley

General Manager, Blockchain Services IBM

In this interview from MWC Barcelona 2026, Jason Kelley, global head and managing partner of core business applications at IBM, joins theCUBE’s John Furrier and Dave Vellante to discuss the evolution of enterprise modernization in the era of agentic AI. Kelley unpacks how IBM is helping organizations move beyond isolated software rollouts to orchestrating complex, end-to-end workflows across multiple cloud environments. He highlights IBM’s unique position as a trusted system orchestrator for major business applications – such as SAP, Salesforce and Workday – ... Read more

Jason Kelley, IBM

>> Welcome back here to theCUBE's live coverage here at MWC26. I'm John Furrier hosting you with Dave Vellante, my co-host. Our next guest, Jason Kelley from IBM, Global Head, Managing Partner of Core Business Applications. Again at IBM. Jason, great to see you. Thanks for coming on theCUBE.

>> Great to be here. Thanks for having me.

>> So I love the IBM titles, Core Business Applications, basically the partners, the big partners for IBM. Explain quickly who the partners are that you manage and you oversee.

>> Core Business Applications, key word there is core. So think of those apps that are core to any of your given enterprises, there's always an ERP in the backbone. So you're going to have an SAP or an Oracle. There's going to be a front-end to that, an Adobe, a Salesforce. You're going to have capabilities of a Workday as well as a ServiceNow. And so those are the core business applications that my teams work with globally and continuing to see the business grow exponentially.

>> You guys have had great ecosystem relationships. The stock price is at all-time high. You guys are performing really well. The Client Zero stuff, all that productivity of AI, we heard that this morning in your booth. Talk about the ecosystem relationship to the company and why that's so important to the new IBM, I'll call it the new IBM, again, thundering away on the business performance. So, yeah-

>> We'll take the new IBM any day you want to say that,

>> as long as the new means the continually growing and-

>> All-time high. - ...

>> value to our clients.

>> And I tell you that, that's where the thought of an ecosystem has really evolved. Our clients aren't ignoring the fact that they have multiple applications. Many of your large enterprise clients have grown over the years through acquisition. So they may have made a landscape decision at one point, but then as they acquire more, that landscape changes. And we know that with the landscape changing, technology has changed also. So those are moving targets that someone has to be able to at least get into some type of coordination and structure, and that's where IBM comes into play. We work from bare metal to software to consulting to research. And so we're the only player on the planet that has that consulting business at scale inside of a technology company. And being that entity, working with those ecosystem partners is really, I think second nature, if not first nature for us.

>> Yeah, I love attending the analyst meetings at the IBM's new headquarters in New York. One of the brief things that we saw on the consulting side was the ecosystem, but also the business model of the service with leverage from technology. That was a big part of the pitch. I want to ask, because now that we're seeing the agents come onto the scene, as Generative AI becomes the standard interface for how people can interact with applications, the rise of MCP servers where people and the ecosystem's connected now. So it's not just, "Hey, we're going to do a Barney partnership. Hey, we love each other. " There's real integration involved now with agents. Talk about how that's changed your business model, your approach, and ultimately the products for the end customer. Because again, it's multi-vendor environments, heterogeneous, okay, great, but now you have this new dynamic where there's a lot more Generative integration going on with agents.

>> So I think with the Agentic approach, people have also ... When I say people, I mean our enterprise clients and clients as a whole, start to question do they really need everything they currently have? That's the first thing. Do I have to have all of the things that we've acquired over the years? And so before you even start about integration, there may be an optimization play there, and that's the first thing. Does it mean you're cutting something or does it mean you're just cutting back? Well, let's decide. Let's understand. And that has to happen by industry and by desired outcome, not just some generic, oh, cut this, cut that. So that's the first thing. And then when you say the agents are all working together, it sounds really nice, but is it real? Did someone walk into that company with a big bucket of agents and say, "Go do your thing. Fly, fly. Go, go, go"? And so I think as we joke, the reality is that there are agents that will do some things better than other agents. There are data sources that they should be working with, whether it's a small or large language model. Is it really MCP or is it agent-to-agent, A2A, and are they really working together? Ultimately, what has to happen is understanding, what's the end goal and looking at the landscape end to end, and then orchestrating that to get to that outcome. And you may start with one part of it, and when I say that, I do not mean a pilot. It's understanding the end-to-end objective and then orchestrating part of that end-to- end objective in phases to get to that end state. So it's not let's run 10 pilots and see ... That too has to have a view and a vision that gets to the outcome.

>> So last May, IBM Think, we had an AMA with Arvin. You were there. I was there. Love Arvin's AMAs. He's just very candid. And we were talking about the new IBM and we're talking about partnerships. He said, "You know, there maybe was a day where we looked at another company and said, "Well, we want to do that, so we're not going to necessarily partner with them. The mindset today at IBM is much different. We look at that company, say, 'How can we, with what we do well, partner with that company and then rise us both up? '" And that seems to be the new sort of model today.

>> I'd say yes. As they say, high tides raise all ships, and I would say as well as a seaworthy ship, it will rise. But you're right, I do think what we really made a pivot to was not just a one plus one equals three. It's a one plus one plus one equals five, because more often than not you're going to have multiple clouds, not one. And you're going to have a new sovereign capability that everyone ... The buzzword of the day. You're going to have that. And now you're going to have not just hybrid clouds, but think of hybrid Agentic capability, because some is going to be private, some is going to be public. And those data sources haven't changed. So I go back to this challenge that people have of being able to navigate that rising tide. And so I think that that's where we play well.

>> When I think about your partners, you got CRM, you got HR, you got service management, you got ERP, you got design. Thinking about those core apps, where is IBM adding its value, whether it's services, whether it's Watson, data, governance, help us understand how you partner and add value to these organizations, customers out there?

>> We'll stick with the AI discussion as we're talking about. Each one of those areas may have multiple players to begin with. In the ERP space, you could have both SAP and Oracle, and when you start to think of how you interact with your clients, your end customers, you could have a Salesforce and an Adobe. And so now, what company doesn't have an HR capability. All those domains you called out, every company has that, and most companies will have more than one tool if it's not something that they've also developed themselves. So when you think of the AI component, all of a sudden you have data sources that need to be integrated or at least orchestrated to get to that outcome, again. So when you begin to look at how that's orchestrated, there's one company that can understand the data. There's one company that has made the investment in AI. We had our challenges with AI a decade ago and people say, "Oh, look, you're not doing that well. " Well, guess what? We were the first ones. And it's easy to sit in front of people and say, "Well, you can actually fall back or fail if you want to call it that, and then move fast-forward. " And that's where we sit now very confidently, knowing that we did that more than a decade ago. So I think the value that we have that we bring is around our expertise in data, our experience and expertise with AI, and of course, the fact that we're doing this across multiple clouds and multiple environments in an open way.

>> When we went from on-prem to SaaS, a lot changed. The pricing model, the business model changed, the operational model went to cloud. It's interesting, this list of companies, because Salesforce actually was the OG of SaaS before cloud in 1999, right? And then Workday-

>> I think Marc would smile at that, he's the SaaS OG, right?

>> It's true though, right? - He was cloud before cloud.

>> Right, exactly. Okay, but Workday rode that wave.

>> ServiceNow. It was interesting,

>> actually ServiceNow had their own cloud. They probably still do. Oracle made the transition, as did SAP from on-prem to SaaS pretty successfully, obviously two very successful companies. Adobe as well. Two questions. What changes in this new AI world? And make the case that these core enterprise apps don't just vaporize because I can say, "I can talk to an LLM and vibe code them".

>> Right, right, because the LLM is going

>> to take care of everything. Well, we've said it. There's still data. There's still way too much data for people to be able to manage. It still has to be integrated. And these applications still have workflows and valuable outcomes that they're driving within the given enterprise and industry. And I keep saying the industry part because people are forgetting that component, that in fact, it's not one workflow for everybody, it's a workflow by industry. And so those components make me believe that there's not one magic agent that's going to come in there and take over all of those applications and make it work. You're still going to need capable humans to do that. And you're going to still, as an enterprise, rely on those applications. Sure, pricing models will change. I think and welcome the outcome-based models that are now becoming commonplace. Those are things that we should all welcome, because that's going to help us deliver higher quality and faster business outcomes.

>> I want to stay on this if we can.

>> All right. - So I would also say, to your point,

>> there's the data, but there's also the business logic, the process logic. That all exists inside today's apps. They do have to take advantage of that. We always talk about the 4D map, the system of intelligence, people, places, things, and processes or activities, and that lives inside of those apps. It's changing because agents are now going to have access to those and agents are going to interact. All these companies with agents are going to interact with each other. I would say this, and we wrote about this recently, that the single user productivity apps versus the enterprise apps may be somewhat more threatened, because those file formats are available and I can-

>> Not that complicated. That's right.

>> And I could go use those

>> and then maybe use something like a PowerPoint to actually produce the output.

>> Right, right. - Maybe don't need that as a

>> work surface anymore. But that to me is more threatening than your partners because they have that process logic, that data, that business logic. Does that make sense to you? How do you see adding value to that equation from, say, Watson or-

>> I think that you're on a great point. First, something that we didn't talk about was governance and security, which comes into play, and that's someplace where we obviously play very, very well and I think with great value to our clients. I think also your example of consumer versus enterprise is a great example, because that's where I think the difference lies in the complexity. You start to think about a standard model. And we've run these tests, exercises with our clients. You could query, you start using the same ... You can ask the questions from many different angles and your prompts can come from different people, thought process, but you'll start to, in a consumer world, come up with the same answers.

>> Well, and ironically, there was a little war of words between Marc Benioff and Satya Nadella. I don't know if you remember this, but Marc called Copilot "Clippy" and then Satya responded with that SaaS is just going to be agents talking to CRUD databases, CRUD, create, read, update, delete. Ironically, what Nadella said is kind of true, but for his own personal productivity apps, because they were trying to sell a Copilot for $30 per user per month. I think the industry is rejecting that pricing model, to your point. As consumers, we want outcomes, and that's a whole different ... It's not going to be that old seat- based pricing model jamming us. Now, that's going to not transition overnight, but I think that, your point about governance is really important.

>> Well, data gravity and

>> governance are two of the big things.

>> These are becoming platforms where governance

>> and security are built in. They're probably invisible. Ideally, they're invisible, but they're there and-

>> There's more to the end user.

>> That's right. And I think also you hit on this, this thought of the complex versus simple, and I'll go back to the client and where they're operating. That customer, the enterprise client is in a certain region, a certain environment with a certain landscape, in a certain industry. Therefore, they're unique, and we're not going to get rid of that uniqueness and that complexity. You'll hear the term the "sea of sameness" or the "cauldron of commoditization. " All those things are, we have to remember that there are some levels of complexity and specificity that still add value.

>> Jason, I want to get your thoughts on the customer. You mentioned the outcomes. You guys are very outcome-oriented with a customer focus. The enterprise modernization trend feels like it's been two decades. Is there fatigue? I mean, customers have gone through cloud migration, digital transformation, ERP transformation. Now they're looking at AI transformation, which everyone pretty much agrees is pretty big and important. Are enterprises overinvested or are they underprepared? How would you rate this next chapter in the modernization sets of things that have gone on? You mentioned before, some things have to be unwound or changed as optimization. Is there fatigue? What's your view on this? Because again, they've been through these things. I mean, not as massive. AI, I mean Andy Jassy, Arvin, all the CEOs are saying, "This is the biggest shift we've seen and we've already crossed the threshold. It's happening."

>> Well, look, we just had this

>> conversation here about all of the core business applications, and it used to be the thought that we're going to modernize because we're going to focus on one part, and we're going to modernize our CRM, we're going to modernize HR. Now you really can't look at it in a vacuum. And so, whilst we may have fatigue in the old way of looking at things, the exciting part and where we don't have fatigue, but I think the market has a lot of momentum and excitement, is when you're saying, "Let me look at this end to end. " I can't tell you how many of our clients or even partners have talked about a Client Zero story. We're Client Zero for transformation. Well, we know that too, and we have our story and that's literally how we looked at it, was to say, how do we look end to end and begin to truly modernize? Those two words, we'd eye roll when we heard modernize, transform, reinvent. Those things would cause people to go, "Oh, not another one. Root canal."

>> I want to talk more about that though.

>> Let me finish my thought because I think this is important, because earlier this morning in your booth, we heard Luca talk about the Client Zero. And there was a lot of details for analysts, so it was a little bit more detailed. But high level zooming out, it was basically like, "Hey, we got AI in hybrid cloud with Red Hat. " That was a big part of his subtext, was his conversation, like, "We have the tools. " So talk about the importance of the AI piece that IBM has and Red Hat, the hybrid cloud, and what that means for customers.

>> So you have a hybrid cloud,

>> which allows an open environment with Red Hat to work across your multiple clouds, on-prem, off-prem. We called ourselves an AI hybrid cloud company way back when people were like, "What does that mean? " Now we're seeing it all the time. When we have done it, we have done it from an open and AI-first point of view. And so when you're able to have that tool set to say, "I'm not just going to walk in and say, we need to orchestrate this differently. We can actually walk in and in your current landscape, start where you are, client, and not just orchestrate, but actually conduct and help work with those bets that you've already made, because we have our open granite models, you can work with multiple clouds, and as I said, across multiple data sets. " That's where we can continue to move forward.

>> Just one more point, Dave, and I'll get this is. Because I think one of the things that people see, obviously the happy customers, you guys see that, but the stock price, that's a scoreboard for success. What's IBM's unfair advantage right now if you had to say it? What would be your version of your unique differentiation? What's the unfair advantage? Because GSIs are evolving. We heard that today, clearly. Workflows, end to end, a lot of this domain. What's the IBM's unfair advantage from your standpoint?

>> I'm loving the fact that you put stock price and IBM's unfair advantage in the same sentence. So we could just-

>> Well, the stock price is the scoreboard. We all know the results are the results, and they're good.

>> I just like the two. If we could print that and say that we're talking, that's wonderful because that could be it. But I think our unfair advantage is our unique place in the market. And so unfair advantage is that we can go from bare metal to software to consulting and research. We're not just looking at today, but we're looking at further out, tomorrow. Quantum, for example, when you say quantum safe, it means something. The only way we'll be able to manage these levels of data, this amount of AI with the safety and security that we need to at the speed is using quantum in the future. So we're already there. So we're future proofing today. And so I think our unfair advantage is just that. The depth and breadth that we have across the technology stack and the bets we're making in the future, looking around the corner to say-

>> - ...

>> where we're going to be. I think it's a safe bet to continue

>> to bet on IBM as the player in the game.

>> And Jason, when you were talking about modernizing each of these individual stovepipes, if you will, and we kind of rolled our eyes, and you called it orchestrate before, but you just alluded to it's more than that. What I'm envisioning is, if you take IBM, you take Watson for data and governance, you can orchestrate, let's say across the six partners that you have and others, there's more than that, I realize, but in addition to the data and the process logic, there's context. That AI needs context, agents need context in order to act in a governed, safe way. And they're going to get that context from the analytics system, from the transaction system. There's structured data, there's unstructured data, multimodal data. There's a system of intelligence that's emerging that kind of sits in the middle, so it's more than orchestration that feeds all those agents.

>> That's it. You got it. - Okay.

>> Yes. - So that to me, is your unique opportunity, is

>> because you're Switzerland to all these apps, right?

>> Switzerland with an opinion, but yes, Switzerland.

>> Okay. No, that's good. You got an opinion-

>> I am a neutral orchestrator.

>> Yeah. I want to make sure

>> that we're not calling it just neutral because people value an educated and very pointed opinion.

>> And you have an opinionated stack,

>> but you're not taking on head-on these companies. Maybe there's some overlap. Of course, everybody competes with everybody these days.

>> Correct. - But that system of intelligence has

>> to be somebody who can work with the entire ecosystem. It can't-

>> Correct.

>> I mean, maybe. It's going to just somebody emerges as the dominant player, but I don't think so this time around-

>> No. - ...

>> because there's just too much going on in the

>> ecosystem. Your thoughts on that.

>> I think it's trust has come to the forefront, and I say that in a business sense and that used to get an eye roll. But it really is to be the trusted player, and I think that's the safe version of you saying Switzerland. Because while we are in the middle of those and being the orchestrator, we're not just, to play on the analogy, orchestrating the woodwinds or the brass or the percussion. We're orchestrating all of that. And that's why I said, yes, we're actually conducting in the orchestration, because some of those partners will orchestrate in their specific areas very well, and we need to make sure that we're conducting that and bringing that to the forefront. Because that system of intelligence is going to be a different system of intelligence for each one of the enterprise clients, and it's going to require someone, some company. Not a GSI, because a GSI is smaller than what we're talking about here. We're looking at not a systems integrator, but we're looking at a systems orchestrator, a system of intelligence orchestrator that's able to do that.

>> There's also an element of composition though, because you're creating new intelligence and insight-

>> Dave, you're going to make me go down that hole front-end?

>> I am going down a rat hole here.

>> It's, you composition.

>> Sometimes, as we see now, it's not as orchestrated as one would think. It's not Beethoven's 5th. We're talking about a jazz ensemble at times where you're saying, "Okay, we're going to have to come up with some new creative ways to do X, Y, and Z inside of a client. " So I like your composition.

>> And I'm tapping into intelligence in each of those partner platforms and in their agents, and then I'm putting them together in a digital representation of my enterprise, digital twin essentially, of my enterprise, and no individual system is going to do that themselves. Somebody or something has to bring that all together to create that digital twin.

>> That's exactly it. And that's where I think, to circle back to what we were talking about with regards to governance and security, and then my prior thought around designing end to end. That's where when we sit there, you almost want to look the client right in the eye and say, "Are you ready to make that big decision? " It's not a waterfall, because when I say big decisions, make a commitment for that end-to-end capability and not a bunch of pilots.

>> Yeah, and that's the key, the agents in the futures nail the compliance, governance. Jason, thanks for coming on. Really appreciate it. Final question, just 30 seconds we've got left. What are you working on this year? What's your goals? What's your focus? What's your to-do?

>> My to-do is to make sure that our clients understand the need for the orchestration and that our teams are the best trained teams on the planet to deliver that, from bare metal to cloud to consulting, and to make sure that we ourselves are empowering our AI platform that we have in consulting as quickly and as dominantly as we can.

>> Yeah. There's a lot of action on the field, as they say, with AI transformations. It is real. It's the biggest one we've seen of all the waves. Everyone's kind of seeing it. Now it's execution time. So congratulations. Thanks for coming on.

>> Look forward to it. Thank you.

>> Jason Kelley, Global Head

>> and Managing Partner of Core Business Applications at IBM. I'm John Furrier with Dave Vellante. More live coverage after this short break.