What Makes a Company Exponential in the Age of AI? — With Anthony Pappas

0:00 In the age of AI, what makes a company exponential? Let's talk about it with Anthony Papus, the chief marketing officer of DXE Technology in a conversation brought to you by DXE Technology. Anthony, great to see you. Welcome to the show. >> Thank you, Alex. I appreciate you having me here right now. >> Yeah. No, it's great to be here with you. And u for the audience, this is going to kick off a series of videos that we're doing here on the channel that are going to focus on big technologies guide to AI. So, we're going to do this in partnership with DXE. We're going to talk about lots of interesting things, including enterprise agents and plenty more. Cyber security is another one of our segments. And so, these will continue to drop on the feed through the year. and Anthony, would just love to hear your perspective. You know, you're you, at DXC, you talk about the exponential enterprise, right? And it is interesting because the technology seems to be moving in an exponential, right?

0:55 There's no doubt about that. the technology is moving fast and the innovations are happening faster than many of us could pay attention to. but doesn't mean companies necessarily will track with the pace of change. So your company thinks a lot about what makes an enterprise exponential. so what does that look like in practice? >> It's it's a really interesting time. it's an exciting time for us in technology, but it's also if you're a fan and and you've watched the evolutions, it's it's unprecedented to be honest. And I think you hit the nail on the head. The technology can move at pace, but really enterprises need to evolve and and and move at pace.

1:34 That means teams need to that means decisions need to forget systems by themselves, but the processes need to change and evolve. So when we talk about exponential enterprise, what we're really saying is it's an enterprise that can do more than what it's been capable of achieving recently or the last 10 years or 20 years. It's it's an unprecedented component, right? So we say speed, quality, and scale. That's really what makes an enterprise truly exponential.

2:08 And what's I I think important here is that when we also talk about the enterprise and we'll maybe we'll get into this when we start chatting it it's not one aspect or another. I think the big misnomer right now is that you bolt on AI and you do a few pilots and and everything will be great and all this savings and efficiency and productivity will occur and the street loves it and it's a great story. But the reality is is that there are far many more things to think about and do in order to actually achieve what we are calling more enterprise productivity or real value for the enterprise. So it's an interesting time and a wild ride right now. So >> and then what is it about AI that will make companies in particular be able to move in the ways that you're discussing? I mean these ideas of speed and quality and scale that's always been around with enterprise and I think for a lot of our viewers like you know the question is is are are we in this moment that's linear progression from where the enterprise technology has been for a long time or is this as you know we sort of talked about in the title exponential. So when you talk about these goals, is it the same goals but just the ability to accomplish them faster with artificial intelligence or is there something different about AI that sort of changed the equation itself?

3:27 >> Yeah, I think this is the moment that we have to actually rethink what we've been doing the last 20 30 years in developing technology. This is the precipice to start to say how we've been building systems and how we've kind of trained the world to think about computing and working by logging into systems and having people work you know through the system to actually collaborate. This is the moment where all of that changes.

3:58 AI is redefining because obviously our our president labeled it super intelligence but if we look at an intelligence layer right now we really start to think about now that we've got unprecedented ability to ask complicated questions that we as business leaders or any team members need to actually get how do we get that information which historically was in many systems and variety of teams had to go and get that data and then resurface it and taking days, weeks, and months to be able to do it. Now, we can ask those complicated questions and complex questions so that we can achieve what we're actually trying to achieve. I've I've I've used this a little differently with my team and thinking through this a little bit more when we really talk about what's the real benefit to AI. You know, if we looked at what we've been doing, we spend 80% of our time seeking and finding information and 20 maybe 20 or even less 10% of our time actually doing something with it.

4:58 right now where we are, the capability that AI is giving us is all that 80% we no longer have to do. We no longer have to see confined. You can have all that 80 90% of that time back. So the question really for us is what do we do now that we have all that time and we have the ability to pretty much get almost any answer that we're looking for however many systems it needs to jump into. what do we actually do and how do we actually progress going forward and I think that's why I'm saying this is a precipice of a new of a new moment.

5:30 Everything has to be rethought now that I have that ability to just get there. Processes have to change. I mean we have six principles to the exponential enterprise that we can kind of go through but it's a fundamental change to rebuilding the enterprise than what we've been doing today. So at some point we can kind of get into that if if you want to we can do that now. >> Definitely. I mean I'll just just make one note. You know Anthony it's interesting you know we talk about the knowledge economy or we have historically talked about the knowledge economy which we move to from the industrial economy and the industrial economy was basically like somebody has an idea like let's make widgets and then all everybody would just spend the entire day making widgets right which was not really an economy that prized new ideas. with the knowledge economy, it was about what you knew. but it's amazing how much of people's time was actually spent on like the quote unquote widget making. And that's like the 80% of time that you're talking about, right? It's like, all right, we're in this knowledge economy, but how often do you get a chance to use that knowledge as opposed to how often do you basically have to do all these execution tasks before you can actually get to the knowledge part? Yeah, it was a it was a time of doing, not an actual time of using is what I would say. And right now we're really in that stage where we now can create use. you know, one of the one of the real key things in my organization, you know, we were we were presented with a real fundamental challenge. You know, our CEO came in and said, "Look, we need fluency. We need AI fluency at such a degree. Everybody needs to move now." and obviously being a services organization to help organizations move through technology but also being you know a a core function in the business. I'm both the customer as well as somebody who would be servicing a client. And you know the real unique challenge is as a business leader and a business owner in order to get my team to operationalize and stop the building right and the time to build I needed them to get to well what is the use? How do we how do we make yourselves and the team actually produce more for the benefit of DXE, right? And that's kind of the real chasm that organizations are starting to look at. And you know, one way I really just classify that is my outcome dramatically changed and I'll use your words of this knowledge economy. It it my my outcome changed dramatically. I needed to now prove and I'm sure there's lots of conversations about core functions in the in the boardroom, you know, about how are you actually contributing? How are we measuring a contribution to the business that's meaningful? I mean, numbers and stats and data is great, but what is it actually doing to lift the needle for the organization? And so, when you think about marketing, they're two key things. We have to make sure people know who we are. thanks to thanks to yourself, people will today.

8:27 but also I think the other aspect is how are you driving business? How are you driving actual revenue to the business so that we can actually grow and and continue that growth. Now when you put that in the lens of marketing that's always a a one side you know one one one CMO might think about it one way one might CMO might just think about it the other way like in data right now we are forced to think about that in both contexts and I have to move at such a ridiculous pace in order to show that enterprise value I'm not able to and and organizations were not able to do that in the past so you know it's a really complicated space right now and when we think about using all of what my team needs to do today is basically enable themselves to do more, not build something for me. And and that's kind of the the real I think change also to the operating model. I think when you start to hear and see people say, "Oh, this is going to reduce our jobs or move our jobs," it's actually not going to do any of that. I think it's it's it's one of these places and again this precipice that you're going to change how you think about the roles that you're playing and what we actually need to do which I actually since I've been doing this for a little while having been through the internet economy and the very very early stages in the mobile economy where we've given everybody those devices you know that evolution that opportunistic view of it is is a much desired view right now I I think a lot of people have more of a fear-based reality to what's actually going on and and everybody's a little confused.

10:09 So, you know, again, it's it's not until you really get in and start to think about this at an enterprise level can you really start to see where AI is going to be super effective and actually just change how we're operating and how we're doing business. >> Yeah. And but the interesting thing is that DXE itself has actually started to implement this technology. and we're going to talk about the principles you've used to do it. And I love the way that you talk about it. It's not, you know, you just go ahead and throw AI at something. You have to actually be thoughtful about it, which makes a lot of sense given how many pilots have not worked out. but I I I'd like to just get a little bit of an explanation from you about DXC, the company, like what is DXE? And then we'll go into how you've used AI to change your operations.

10:54 >> Yeah. So DXE technology is an enterprise technology organization. We're we're a partner here who is really focused on driving innovation throughout the entire enterprise. And when we say the entire enterprise, we mean both in operating infrastructures and security and you know dealing with things like workplace and and employee you know operations. But we also have a build side or what we would call an innovation side where we're really thinking through new strategies, new developments, enterprise application implementations.

11:33 So we're one of the only few companies in the world that can operate across the entire enterprise on either side helping organizations and a lot of our customers are very missionritical organizations. So in regulated businesses, insurance and finance and banking amongst others and you know they rely upon us to deliver without fault. You know we have a saying if if something happens with one of our systems the world is is going to know about it. We're we're really operating some of these these mission critical systems that are happening in the world today. So, you know, we come from a heritage of having clients with us for 60 years, 50, 30, 40, 50, 60 years. And and we're one of those organizations that we really like the hard stuff. And when I say the hard stuff, AI is a great example of the hard stuff. It's it's a complex environment. It's across the enterprise.

12:35 it needs to have deep industry knowledge within a specific industry or vertical market. and we have all of those and our heritage really comes from some of the large organizations that you might have heard of like EDS and CSC and HPE. so we have people we're about 120,000 people globally in 70 countries. So our diversity, our engineering you know our engineering strength, we have full stack engineers you know over 50,000 of them. You know we have a lot of what I would call technology chops that companies are relying upon and industries are relying upon for for what we're doing today. So, you know, a as a systems integrator, which is figuring out all the pieces of how things need to work. AI right now is where we're really shining because it is that unique challenge that is how do we get this to work? How do we plug in the pieces? What has been the historical context for all this technology? And then how do we actually move it going forward? So, we're kind of really excited about this. We're a bunch of geeks in that space. But it is interesting because for a company who as you said if something goes wrong you hear about it and if you're working with companies that liter like doesn't seem like they can tolerate many mistakes.

13:53 You know AI is a technology that it would almost seem the conventional wisdom would be that you'd want to stay the furthest away from because of the unpredictability and the probabilistic nature of the technology. So, have you wrapped your head around the idea that, you know, it's it's still important to use or worthwhile to use or is there a a tolerance of error that wasn't there beforehand? Talk a little bit about that. >> So, yeah. So, I think look, I think the reality is is that we're we're a technology organization and and the only constant in technology is change and we embrace that notion. You have to learn it. You have to get involved with it. have to be able to understand where we're going. And right now AI isn't going away. It's going to evolve and change. We don't believe AI is you know a single technology or a single moment in time like like sticking an enterprise application into your system.

14:48 AI is about an era and that era is going to evolve and the technology is going to evolve and we're seeing the models change with time and because we've got such deep experience in you know some of our heritage is building satellites the first ones some of our heritage is building the internet for the first time right so we've got decades of experience in dealing with this frontier if you will so one of the great examples of that is this idea of security and governance that you're talking about. You're going to meet some of our our folks. You're going to meet our CIO and and CISO. and they're really at the frontier of this. But because we have that legacy, because we understand that we can't fail, we've not only used a customer serial approach, which is really means that we've implemented it across our entire enterprise at what we call production grade AI at scale governed and secure and actually we're really leading in that security space which they'll come up and and do much more justice to it. But because of our heritage, we have to embrace this at such a level. So it's not that you have to tread lightly. It's that you have to understand what is actually out there and how to actually operate with it and then build what's necessary to continue that going forward. And so that's actually what we do really well. And and our CEO is a is a very entrepreneurial, innovative CEO. And you know from his concept he's been building leadership that has that frame of reference as well. It this we all have to move with caution. We all also have to understand the real challenges that are going on in the world. But we also can't shy away and say we're not going to actually touch that and and not approach it. Now the systems that we are building we are testing. They are tested. They are approved and some of the data will actually show that. So when we use these with our customers we're not just putting a pilot into practice. We're putting realworld, real tested, production grade products that have been built with a lot of expertise and you know we we some of our results will kind of speak for itself in that category.

17:01 >> Yeah. And I think now we should take it from theoretical to practice, right? So we've talked about like a willingness and and the way a willingness to implement AI the way that you operate u but I actually want to hear about how these projects have gone because ultimately that's where the rubber meets the throat, right? It's like you could talk about AI, but a company like yours has actually worked to implement it and has a lot of things that you've learned that we can hear learn a little bit about what's happened from things that have gone right and things that you'd like to improve. So, I think we just start with how you've implemented it internally. can you talk a little bit about the AI program that's going on within DXC and and then either during or afterwards we can get to the principles through which you've built it? Yeah. And I can what I could do is I can I can talk about the principle and then I'll just give you an example with with that customer zero approach, right? So the the general gist of of being an exponential enterprise you know the the key here is that you need and we want to talk to our customers and saying when you go down this road of trying to implement AI into the enterprise it is not about one little aspect that changes or the bolt-on as we talked about. So using marketing and communications which is my group and you'll you'll hear some examples on IT and and and some products that we've built within our organization but I have an organization that's global. We operate in a global context and you know when we were faced with a couple of different things at DXE one was a rebrand. I came on as a C CMO as a new CMO and and we had to rebrand the organization reposition for this AI era but we also had to start to generate demand and when we when we think about that you go okay well you know just start some demand genen program but when you don't have the infrastructure in place when you don't have any systems that can tie any of the data for what is being thrown out there in a marketing organization let alone that we had very very few what I would call outreach or direct channels or campaigns being run at all. but we have no tie to really understand what happens when that kind of information is out there and then how does it tie actually into what you know what we call total contract value or TCV or actually just end up being revenue that we can close. You know there was zero linkage to that. So if you can imagine 9 months ago we're dealing with those three those situations.

19:32 My outcome as a marketing organization wasn't make more campaigns. My outcome was I need to add revenue to the core business so that I can show enterprise value by growing the demand in in the business. Right? So just fundamentally the first principle is define the outcome and the outcome and the focus on that outcome is dramatically different than what we've been thinking about. Most people their outcomes are make my people do more or the teams do more. In my case, marketing across all of marketing globally had to provide data and real world value to the organization. Very different place. The second thing in order for me to do that I had to say okay well how can my team operate or start to you know utilize the AI AI quote unquote AI pilots? how can we use this to actually do more with it? So the the first inclination is you take the different groups within your organization and start to say hey let me let me develop a pilot or a little prototype of something that maybe an individual might do and they might be using a cloud or GPT and they might be using a thread base but the reality is is that you need to actually put AI at an enterprise level across M andC. So that's what we did. But the second principle is really getting a scaled production grade AI. It's secure. It's in a in an environment that you can actually work in. So it's not being shared externally. It's it's dealing with what we have and our data, but it is also governed in a way that I understand my team understands the guardrails. You know, when we think about this, I had to make a lot of new friends in a marketing organization. I had to make friends with the CDIO and I had to make friends with the CISO and say we need to enable this team so that we understand how we can utilize these tools. Right. So once we did that production grade AI and scaled it and they actually implemented across the organization.

21:33 >> By the way, can I just ask what technology do you use? Like what technology when you said you want to implement AI at you know across the organization, what what was your what did the technology stack look like? >> Yeah, so that's a fantastic question. and again, I would I would tell you if you tell if you tell Mr. Russell Jukes, who will be on your show. I said this, I'll deny it, but he was brilliant in his >> this is on camera. I can't help with that.

21:57 >> Oh, no, now it's too late. yeah. So, so I think that, you know, one of the things that he did that was, really brilliant is we looked at a variety of stacks and this is actually one of our principles a little bit later on. We look we have claude running as an enterprise application set. We have AW quick AWS quick running across the entire organization. So we implement that to 120,000 of our employees. We've got co-pilot in there. We've got chat tpt and then we have layers on top of that as well. So we also built our own AI playground. So when we start talking about governed production grade, you look at all of those frontier models that all the employees have access to, the one thing you might say is, "Oh my god, this is a free-for-all and it's so much." But when you start to look at AWS quick to build your agents and claude to kick off and start building some skills and all of that data is actually inside your enterprise that each of those frontier models are collectively actually building upon that turns this into a very different capability set.

23:13 It's no longer individuals with their individual threads. We actually have and then of course we built something on top that allows us to tie all of those together for ourselves as an organization so that our employees can start to ask complicated questions and work with all of those models simultaneously. Now in that kind of security and governance model several layers were built on top and when we start to look at like what is the new enterprise that's where this level of implementation this level of scaling AI starts to reveal new services new solutions new products that need to get built upon this complicated more open ecosystem and you know Russell and and and the team doesn't like the word open but really it's a flexible environment that allows any technology to kind of come in and out of but in a very secure and governed way. So it starts to then get complicated. But right now because things and the way models go down and up and they're a little bit unstable, you have to almost separate yourselves in your environment from all of that activity and not be reliant upon just one. So some of it is a fail safe, but some of it is flexibility. like my team actually utilizes all those models and then what what I would what I would tell you is we also use a whole other set when we generate our own content in a very different way of nodebased u models that are out there that allow us to get some gen AI kind of capabilities. So the seed dances and the veos and the clings and the 26 other models that are out there because when we really start to think about production grade AI, you also have to think about tokconomics. And I'm sure that you've had this conversation about the tokconomic challenges that organizations are dealing with. And so when we built our own playground, we get to try things and we get to spend less money. And only when it was actually something that we believed can go across our entire team did it actually elevate into something that our our IT organization would release to us. And what that did is it really put a dependency on my team to make sure that they were very focused and fixated on building the tools and not just little pilots but a tool that the entire marketing organization can can utilize.

25:40 And so what that did in that in that step number two on that on that production grade outcome is it I realized that my organization needed to change the the roles, the skills, the way that we were building our campaigns, the how we were we were getting as much of that content out to the outside world and making all these con that had to dramatically change. So our third principle is you have to change the way that you're actually operating because you've built an organization building on the old tech. Now we're in a completely new enterprise space and what the demands of of the of your team are dramatically different and their roles are going to change. What they're doing is actually changing. How they think about it. They're still experts in the field but actually their day-to-day is going to be changed from seeking and finding to now doing stuff with it.

26:30 Right? So when we changed our operating model, I developed a structure so that marketing can actually work in a linear fashion. It sounds kind of non-efficient, but meaning that every group could actually work together for the greater whole of basically driving demand. And then what we then did and and again back to the demand gen once I've got AI production grade AI, once I actually change the way that the team is working with that production grade AI and I'm really focused on driving demand, I needed to have them build an agentic workforce. And I'm sure you've heard those terms as well. But when we really talk about the agentic workforce, what we're really saying is that we need to automate the things that the team historically had been doing in a manual process. And that automation would not have happened without the expertise of the team because they know inherently what needs to get done. And so what we did is we offloaded some of those, you know, seeking and finding type missions to an agentic force that actually went out and got all the data. So we we we dipped into Salesforce, we dipped into our, you know, Marquetto environment, we dipped into all the leads, we hooked it into all the channels that we distribute all of our content out there in our events. And essentially the team built a dashboard on our own with just a little bit of help from it. But we didn't implement a new system which again is this fundamental change to aggregate all the data so that we know right now that when we put something out in the world if it's actually been responded to when by whom and then who on our sales organization has actually followed up on it. So we deploy it to the sales team say this is a real activity and a real opportunity you need to follow up where they are and it can translate and we can track that all the way down to what we call like I said before total contract value. So the reality is within a 9-month period of time, my team reorganized, redeployed, and built on their own an ability to devel a demand genen system that now we're tracking upwards of over a billion dollars in net new leads on a quarterly basis from not knowing that data historically. And the other flip side to this is we in order to be able to feed that monster, you you really have to build more content. You have to build more campaigns. You have to be more active in the outside world. And from a marketing perspective, that's actually, you know, from a CMO, that's actually what I want to be doing because that helps that first point of building brand awareness. So the only last two points to our principles here is that once you be able to automate some of those tasks, you really need to be able to plug into that, you know, open or a flexible ecosystem of lots of tools, lots of technologies because not everybody's outcome is going to be the same and there's not the same way to get to all of those outcomes. So, like I had said, in order for us to produce a lot of content, we're using, you know, I would say probably about 10 different tool sets, engines that help us get that data. And that's the real reality of it.

29:56 They're not existing systems. They're new, but I know that those systems are going to evolve and change, and they're changing on a weekly basis. So having our IT department, having our security group, having our supply chain be able to help me move within that ecosystem is a super important effort because of the speed and innovation. And then you know those numbers that I threw out are really our our point of number six. If my outcome was really to help, you know, drive demand, what can I point to to show that I've driven demand? And that's where we start to get into this net new opportunities all the way translated into, you know, total contract value of of dollars and cents.

30:37 That would not have been able to happen historically in the what I would call the systems economy. And that is an my group marketing communication. That's one example of what it means to be an exponential core function within an a new enterprise and how AI is evolving. But unless you do those steps, you are not going to be able to achieve your outcome because they're compounding. They're actually on one another. And that's where we get the exponential.

31:06 Once you begin, you've got to go down that path and then it takes off. And now we're in a in a state where we're every day, every minute checking on that data and being able to have that level of activity. And anybody who's done that before, you know that there are systems out there that you have to kind of work within their bounds and their interface and it has to adopt to your model. Well, we actually built it ourselves around our process. So, you know, when you put that into real context of what kind of power is AI doing for an organization, it is fundamentally changing as long as you can kind of get through these gates and really embrace the change that's really in front of you. And we know that some of our clients are are more riskaverse in that and they want to take each step which you can absolutely do but again it's one of those components where you'll you'll find incremental value in each one but you don't get that real enterprise value and that real exponential kind of growth until you embrace them all. You know, Anthony, an AI critic, just to give voice to what AI critic might say, you know, would say potentially, this notion of having a unified view of a customer or a prospect, intelligent routing to sales, better measurement has sort of been the thing that, marketing and sales have been working through on software, initiatives for years. And so I think that, you know, when when it comes to the exponential side of it, you know, the critic might say this isn't exponential. Maybe this is papering over some things that the software you know existing software has come up short on. So when you think about the exponential side of this where do you see it as exponential versus AI being let's say a normal technology?

32:48 >> Yeah. So, I'm going to put my business hat on a second when I'm pressed with a with a a budget and the budget says here's the number that you have and I now need to operate and deliver everything I need to operate with a significantly different budget exponentially is what I didn't tell you is there was a huge number for demand, but I also saved $15 million right out of the gate in my marketing organization and budget and reduced my dependencies on outside organizations and made my team in complete control of what we produce, how we produce it, and how we launch it to the world. So all the data is ours, all the information is ours. We're creating it from scratch.

33:36 It's as authentic as it can be. And we don't and we did this with a time frame that I' I' I'd ask any organization out there because I've been doing this for a while. How do you move within 6 to 8 months from zero to going 200 m an hour and not have the dependencies of any of those and it's actually operationalized? It's not really possible. And so when we really talk about exponential that that defines everything moving at speed the quality of the data and I've scaled it in a global way and it it it wasn't my organization. It wasn't me personally but it was my organization who's worked with the CIOS and the CISOs. It's such a different relationship than they have in the past.

34:26 It's not a client kind of vendor. We were handinand being able to actually execute at pace at speed that I haven't ever in my 30 years in in this in being in marketing have never seen something move as quickly as that. So it hit on the dollars hit on the growth and it and it hit definitely on the speed factor. >> Not possible without generative AI. >> It's not possible without AI at all. >> You can so so here here's the thing.

34:57 I would have had to have implement 10 systems. I probably would have had to have had worked with five or six organizations on the outside to be able to implement all the different technologies to achieve this at a level that we've achieved today. And I think AI is really the one that that that put us over the edge. It gave us the ability to do to do that, >> you know. And you talk about the savings that you were talking about. But I would imagine that any token cost is well not any but you even at the going rate of tokens you could spend a lot and still come in with savings but you did mention the tokconomic side of things. So yes when you think about doing this in a way that has real ROI for organization you figured out a way to handle the token spending and make sure that you're not one of those companies. I think there was a headline that one company spent like 500 million in a day or month by accident with Claude. so the token side of it is working for you.

35:54 >> Yeah. So what we So the interesting part back to the Russell who you'll you'll meet our CIO. you know, we we we weren't handed an open what I would call the open checkbook of of tokens. We actually had to make sure that we had, you know, true products in the playground. And once we did, we were able to deal with tokens a slightly different fashion. What I had my team do is we built our own token calculator based on the models that we use. And actually before we go ahead and run either Gen AI type tools and using different models or we start to run agents in a certain way or a series of skills that might you know take on a lot of compute power. we actually do an estimate of what it will cost and we'll either turn it to free models for a draft versus a final. And so that kind of token calculator that spans all the different models that I mentioned actually gives us an indicator where they would come back to me and basically say look here's the cost here's what we're going to get for it and it allowed the teams to think in a much more structured way. So it's not that the teams are just going off and spending frivolously. They're actually measured against it, but we're using the calculator to kind of keep us in check.

37:22 And then yeah, there are times where we are spending you know significant dollars on it, but we know that that money spent is actually producing a real output, a product, a content generation. It's being utilized for something. And so that little piece of upfront would not have been able to happen if we didn't actually get what I would say have a disciplined approach in being allowed to get what we were getting in the core functions and and and again that's where my credit goes to Russell Jukes. I think that philosophy in the enterprise is one that you know should be commended because I I have also spoken to other customers and organizations and other CMOs and other organizations where you know they've given everybody sort of their ability to use tokens or I should say just their own credits but they also have independent accounts and it's not even aggregated. So now they're in the mode of of of collecting everybody and saying stop the spend. We built the calculator in and some of our other products like you'll hear of DXE Oasis which is a is an operational product for Knox and Socks that has that functionality in there. So when you're starting to generate agent compute power which is 100% of the computing and coding within Oasis and a lot of running those operations you understand those dollars and cents before the operation. So you know we're being mindful of it. It's going to be it's going to be ever evolving. but we don't feel like we are at the mercy of tokens which I think a lot of organizations feel that they are.

39:03 >> Anthony, last question for you if I can. >> sure. >> You are a software user. You are your company works to integrate systems. you're a software builder as well. do you have any sense about what the future of software is going to look like? because it's a big question that we deal with all the time here. is it going to be like the traditional SAS system? Is it going to be everything in an agent? Some hybrid of the two. Where do you see this going?

39:32 >> Yeah. Now, I'm not the full technology person, but I'll look at it as a business owner perspective and and being a business leader having to look at systems for my organization. you know we do services, we do the solutions and we we build the products and what I think is is going to happen is that the systems that we already have and are utilizing today will remain. They're not going to go anywhere. are we going to use them the same way? I think that's where the evolution is going to really happen. So for instance, you know, we have our CRM, we have our our analytics tools, but we're using them dramatically different than what we've done before in the past.

40:16 And what we really are seeing is that these layers are being built above these systems. You know, things like the mainframe or things like, you know, stuff that's out in the cloud or or even just large enterprise systems that are out there like the SAPs and and the Salesforce. But we need the the data repositories. We need to understand where all that is. And you've been you you spent all that money. We've spent all that money on that investment. Do we throw them away? Absolutely not. but what we now are able to do is keep those systems in place, but I believe that the layers that are going to be above that are actually going to be individualized for the organizations using it.

41:00 My example in marketing was we built our own dashboard. But we've also built our own content engine. And that's not a third party node system that's out there. That's not clawed. That's not GPT. That is a content engine that utilizes all of those texts, brings it in into a process that is built for our organization. And so it's going to get a lot more personal because you have the ability to do it. And I think that the product game is going to evolve. you know, when I look at organizations out there coming in and saying, "Well, you implement our new software system and you train the model on your brand and you train the model on all of these things and that's going to take a year's time." I as a business leader say I don't have that time within my organization to be able to do that. So unless that product or that software is flexible enough to get ramped up now then it's a little bit easier if my team and my expertise and my knowhow within my organization and where everything resides and lives that we have an ability to build that layer into it and then work with the systems that are more what I would call data stores in a better way. So again, that's my philosophy because that's where my team has seen the most success. and as I look to the future, I would love it if there are products out there that, you know, can replace some of those services. But I think one of the challenges that we have is training your teams because we're all moving and changing so dramatically.

42:38 Having to do yet another element to your organization also stresses out your organization. So the the path of least resistance for somebody like myself to for my team to be most effective is to say what's going to be the most comfortable way that my team can be fully in you know involved in moving the needle on us working towards our outcomes rather than waiting for outside products or tools or systems to come in. because right now speed is of the essence as you know and that's kind of what everybody's being measured on.

43:14 >> Fascinating stuff, Anthony. Thank you so much. >> Absolutely. This was a pleasure. Thank you very much, Alex. >> If people want to learn more about DXE technology, where do they go? >> You can go to dxe.com and you'll get everything that you need to know there or obviously on socials as well. but the easiest thing is to go to dxc.com. >> All right. Amazing. Thank you, Anthony. Thank you everybody for watching. we will be doing a series of videos here on the show going deep into areas like enterprise AI, AI agents, cyber security and AI ROI doing that in partnership with DXC. So very thrilled to be kicking this series off and to have introduced DXE technology to you here on the feed and we look forward to being back soon. Thanks again for watching and we'll see you next time on Big Technology.