From Workflow To Blueprint: How Pega Reimagines Processes

What if AI could design your business in minutes—and still pass audit? We sit down with Pega CEO Alan Trefler to unpack how Blueprint uses generative AI to reimagine complex processes while preserving the predictability enterprises demand. Instead of asking models to improvise outcomes at run time, Blueprint captures objectives, proposes stages and data flows, and feeds a best-in-class workflow and decisioning engine that’s built for change.

We get into the core challenge every leader faces: turning AI hype into production reality. Alan breaks down three kinds of AI that matter in the enterprise—statistical next best action, generative features that assist workers, and generative process design that rethinks how the business should run. The twist is timing. Let the model be expansive at design time, then lock the logic, policies, and decision frameworks before go-live. That keeps outcomes consistent, explainable, and ready for regulators and auditors.

You’ll hear how this approach reshapes sales and consulting. With just a few inputs and a client’s website, a first meeting can produce a tailored blueprint that feels shockingly complete. Consulting partners can now embed their proprietary IP via partner-branded blueprints, marrying their best practices with Pega’s runtime to deliver four to five times faster. We also cover trust guardrails—data isolation, no training on client data, transparency into decisions—and why predictable AI will separate pilots that stall from systems that scale.

If you’re evaluating AI for process transformation, customer engagement, or operational excellence, this conversation will give you a practical framework for choosing the right AI for the right job—and a playbook for moving from idea to production with confidence. Subscribe, share with a colleague who owns transformation, and leave a review with the one process you’d redesign first.

Transcript

SPEAKER_00 0:01

She said, you know, I'm tired of all these banking demos. You have something more interesting. And I said to her, Well, have you ever thought of starting a business? And she said, Yeah, actually, I had at one point wanted to create a flavored water business. A company to sell flavored water. And I said, you know, great, you know, flat as far as book. And I took it out, I put three sentences into my iPad. And what came out was amazing. First of all, it goes to our intellectual property, which knows nothing about flavored water, but knows a tremendous amount of how workflows work and how customers should be engaged. So it knows, for example, that if you're going to be doing something, you need a set of objectives you're trying to achieve, uh, you're going to execute to a series of stages and steps, those are the workflows. You're going to have to connect different backend systems, you're going to have to do certain record keeping. You might want to have certain user experiences that come out at or generated. And literally in under 10 minutes, asking a couple of questions along the way, it basically designed her business.

SPEAKER_01 1:23

Today on Customer Land, I have the honor privilege of speaking with Alan Treffler, who is CEO at Pega. And for me, this is kind of a milestone because I've spoken with uh a lot of people at Pega over the years. Of course, everybody points to Alan uh for one reason or another as the leader and kind of giant brain behind it. I know you would dispute that, but um but for all those reasons and more, I'm honored that we finally get to talk. So thanks, Alan.

SPEAKER_00 1:52

Well, thank you for having me. I would say they mostly point to me to blame me for something.

SPEAKER_01 1:56

There was that. Oh well, you know, one of the things that uh that I that I love about these kinds of conversations, especially with people like you who sit at the top of their organizations, and even more so sit at the top of a space that that I think gives you perspective that's really unique. You get to see a lot of stuff that a lot of other people, like a lot of other people who are pushing the buttons and pulling the levers and you know, actually doing the the the work, don't get to see. We're very task-oriented down at this level. Um, on the other hand, you get to see trends and paths and potholes and uh potential risks and hazards that probably a lot of people just don't get to. Um so I'm very interested in your perspectives from from that level. But maybe just to lead off the conversation and for people who may not know, could you tell us just a little bit about what Pega is doing in the marketplace right now?

What Pega Builds And For Whom

SPEAKER_00 3:01

Sure. So um Pega is uh a 5,500-person company. We're very technical, we're very much involved in building technology aimed at sophisticated clients. And the technology falls really into two main buckets. On one bucket is that we support workflow automation. That is, how do you capture the workflows that make up a business or that describe the relationship between a customer and a company, perhaps? And then how do you put in reliability and predictability and automation into that as much as possible? And the second part, which is related, is how can you use real-time decisioning powered by a variety of statistical and machine learning type concepts and measures to make great decisions and then to reevaluate and do the next best action in those workflows. So you get both great decisions and great execution. And we do this for uh many, about 750 of the world's most sophisticated companies. Our customer base tends to be very uh large firms, often global, uh, or large government agencies. And, you know, I think what that gives is uh me the perspective of trying to look at this set of problems in this world from our clients' perspectives, which is a whole nother vantage point, looking at them and their different uh business choices, technological choices, the different values they're trying to promote from agility to reliability, and how can you use this decision-making technology and this operational workflow technology to do a great job? And how does AI now empower you to think differently about that?

Why Pega Is Winning Against Giants

SPEAKER_01 4:59

I'm I'm really interested in your perspectives on those things, but but let's put it into a little bit of context. Um you're you're competing against all kinds of giant companies, and I'm thinking of Salesforce and a few others um that are offering very sophisticated solutions, and yet you seem to be winning. I mean, you're you're recently well, yes. Yeah, and I don't I don't say that as if I can't believe it's not no, you're you're winning because you are doing some things very, very right. So can you elaborate on that? You know, what what's behind the wins, if I could ask it that way?

SPEAKER_00 5:42

Well, you know, we're public, so our financial results are are public. And we just saw our growth rate tick up a little bit uh in the last in the last quarter. And what I would attribute it to is the really creative application of generative AI through something we call blueprint to how we handle both elements of the business. How to be able to allow a customer to have a business problem, a business challenge, a business objective, and literally engage with our blueprint agent to be able to redesign their processes. And then also how to have the agent figure out when it's a customer-related issue or customer-related engagement, how they should engage with clients, what sort of what sort of types of offers they should make, how they should present them, how they should think about them. This blueprint capability we have really uh walks on both sides of that decision making and execution line. And it's really jazzed up our customers and prospects a lot.

SPEAKER_01 6:52

So um tell us what it does. What what is it?

Blueprint: The Generative AI Approach

Live Demo Story And Secret Sauce

SPEAKER_00 6:56

How does it work? Well, actually, you know, it's it's interesting. It's it's uh free and available to anybody at Pega.com slash blueprint. Um, and what you do is you put in a description of a business. Um, I I'll share you a story. I was with uh the COO of a big Nordic bank, and I was gonna tell her what we're talking about, what we did. And uh she said, uh, can you show me? And I said, Yes, let me show you this, because took out my iPad and what to do with it. She said, you know, I'm tired of all these banking demos. You have something more interesting. And I said to her, Well, have you ever thought of starting a business? And she said, Yeah, actually, I I I had at one point wanted to create a flavored water business, a company to a company to sell flavored water. And I said, you know, great, you know, flat or sparkling. And she said, both. And I took it out, I put three sentences into my iPad, and what came out was amazing. It it first of all, it goes to our intellectual property, which knows nothing about flavored water, but knows a tremendous amount of how workflows work and how customers should be engaged. So it knows, for example, that if you're gonna be doing something, you need a set of objectives you're trying to achieve. Uh, you're gonna execute through a series of stages and steps. Those are the workflows. Um, you're gonna have to connect to different back-end systems, you're gonna have to do certain record keeping, you might want to have certain user experiences that that come out it are generated. And literally in under 10 minutes, asking a couple of questions along the way, it basically designed her business. And she was blown away. And we get that reaction um a lot. So this took something, you know. Let me tell you what the secret sauce is here. Pega has been automating workflows like that for our whole history. And we've built a system that is really comprehensive and very powerful at being able to take those artifacts, automate, automate them, make them changeable, our tagline is built for change, make it so that a business could run in uh an effective way. But historically, it could be a little tricky to learn and use. What we were able to do with the generative AI was say, hey, we're gonna take our core workflow engine, which is broadly considered absolutely best in class, and we're gonna teach a whole new way to explain to it what what you want done. You may have heard this vibe coding stuff people talk about these days. This, I would say we we didn't model it after that because we've been doing this now for two years. But this is sort of a vibe way in English to describe a business objective, a business problem, or a business challenge, and and have the system actually create the automation steps.

SPEAKER_01 11:08

Wow. So uh two questions. One, did that COO actually run with the business as it had been laid out?

SPEAKER_00 11:16

Are we gonna no, but we we succeeded, we succeeded with the system we were really trying to sell.

SPEAKER_01 11:21

Well, there you go. There you go. Um, again, I think this is uh uniquely viable coming from someone like you, but you know, what has it been? Three years of uh we're in in the AI hype cycle right now, three years-ish, maybe four. And PEG has been operating and building AI-based tools for whatever, I think you said 10 years-ish, more or less. Oh, 15.

SPEAKER_00 11:51

I would go back in 2010.

Three Types Of AI Explained

SPEAKER_01 11:52

Okay, all right. It's that that's quite a while. So there's a looking at the hype, looking at the reality. Um, I talked to a fair number of high-level executives who I can tell are um still very much caught up in the hype, which creates a heady mixture of excitement and anxiety and fear and hope. Um, maybe you can dispel some of that. What what do you what is the reality right now from your perspective?

Features vs Process Redesign

Design-Time Reasoning And Auditability

SPEAKER_00 12:27

So the the reality depends a lot on how you do it. So, first of all, um, we think of there as being three types of AI that a business and a uh uh a user needs to be able to differentiate with. The first is statistical AI, which is or machine learning, which is hey, how do I take this pattern that I see my customers following? How do I evaluate it? How do I understand what the next best purchase or the next best offer should be for a particular customer? And that's really using using math. And it's um it's what we started with in 2010, and it's very, very popular and very powerful to be able to do that. But there's also now, to our mind, two different types of generative AI. One is what I would call generative AI features. So that's when you have a business application, like um an application that that creates sales quotes, or an application that uh takes complaints from a customer. And you use the AI for certain features, for maybe being able to craft a note, or maybe being able to summarize a set of interactions so they're easier to understand, or maybe to be able to do translation to a foreign language, which has now become incredibly, incredibly powerful and reliable. That's what I would describe feature sets of capabilities, are what some of the hype is about. And what I tell my customers is they're gonna build some features themselves, because they're all huge companies. They're gonna buy features from other companies, and we have a couple of dozen as well that we want people to acquire. So, for example, we have a feature that listens on the phone line when a call center representative talks and figures out from what's being said what it should fill in on the screen for the person to then just be able to approve it. You know, it's a case where it's an assistant sort of sort of discussion. So the second type was features. Um, but the third type, which I don't see anybody else doing, is actually using the AI to reimagine and redesign the process itself. So to be able to get the customer to describe their business problem, and from that description of the business problem, be able to say, hey, this is the way you want your flavored water company to work. You're gonna need chemists, you're gonna need, you're gonna need delivery trucks. If it's direct to consumer, you're gonna have to have the setup to be able to do the mailing and the personal and all those other sorts of things. And being able to use AI to really design and plan um is very, very powerful. However, that's where those big companies like Microsoft have gone wrong. Because what they say is you should create agents that are gonna figure things out and do them. You want to create an agent based on prompts, based on a textual description of what you want. And then based on that agent, it's gonna go do stuff for you. So, you know, maybe it will decide whether it wants to pay a claim in an insurance company, or maybe it will decide um, you know, what the whether whether it should authorize a certain procedure based on certain rules and regs. They'll tell you to do that with agents. And what we say is no, no, no, no. You might have some agents as type two, that that central center type to do maybe a summarization because that's safe, that's controlled, that's limited. But in terms of the actual process, the actual complex thing you're actually doing for your business, you cannot do that over and over again when you're interacting with it. You should do that in advance. You should create the library, as it were, of process definitions and workflows, as we call them, that allow you to describe how you want to work with your customers. And by creating those in advance, by doing your reasoning in advance, what you're able to do is look at it and make sure it does what you want. And you can show it to an auditor or you can show it to a regulator. And you can have confidence that if two people come in, that they'll get treated in a fair and same way. But you can get all the inspiration and power that you get from having the language model there. We're the only people who have a workflow engine that we sit in front of, as opposed to what the others are doing, is they're just turning things over to the language model. And they're saying, hey, solve this for me. I mean, you never do this. If you were if you were running a business and you were select paying insurance claims, you know, which is one of our use cases that we do for a bunch of our customers, and you hire a really, really smart person, smart as a language model, you wouldn't just sit that person down and say, hey, here's a stack of your claims to pay them. You'd say, hey, we've got we've got policies and procedures and other types of things. And by the way, we've interpreted them for you. You've got a workflow that you could follow, and you've got decision frameworks around when you make certain decisions and when you don't. The danger with just turning it over to the model is uh is candidly, the models are very, very sensitive to small variations of the data. And and just because a company, you know, customer had a particular variation in their experience, it was different, those might not be the reasons that you want a different outcome. So by doing that reasoning at design time, you eliminate the foibles. And the reason we could do it is we had already this super powerful set of engines, this runtime engine for um for workflow and this runtime engine for decisioning, that we could learn to feed and make it easier to build and design, as opposed to saying, hey, I'm gonna start generating code and other sorts of things. Trouble with all this code generation, a lot of these guys do, it's very hard to go back and change the code. Yeah, it might be easy to spit something out, but you want to go back a couple months later, you want to change something. How how do you know you're changing on anything that's big? Small stuff or stuff that doesn't have to be predictable and reliable. I I love them. We we use we use the engines all the time when we're doing creative exercises or you know, we're trying to figure out the thing. Right. Or names for product that could even be very, very creative. Or, you know, we will say we can't we don't know how to we don't we don't we we have a product that's as this we don't know what to call it. Give me 10 names that are are are uh copyrightable and bang, you have it. It's it's a it's uh amazing. You have it in seconds, but it's being interpreted by a knowledge work, it's not actually leading to outcomes direct for the customer or you know that might affect the customer.

Creativity Without Code Generation Debt

SPEAKER_01 19:38

Does that make sense? It does. I'm just I'm I'm trying to interpret that. Um, and tell me if I'm if I'm anywhere close to right on this. If if your systems to you know, second system generative AI, call it features and applications work. Yep, um, I understand it. That's probably 99% of what I see out there. Um, but the more interesting conversations I think have to do with what you're calling the third scenario, which again, this is my interpretation, what I think I heard you just say, is that the more interesting and maybe even more appropriate use of these engines is it their ability to rethink, redesign, and reimagine the systems rather than just being uh uh applications for task work. Do I have that anywhere in your clothes?

SPEAKER_00 20:32

Yeah, you do. And and tying them together and and making it all work, making it visual, collaborative, etc.

SPEAKER_01 20:41

So let me ask you something. Um, I can I can see you see an awful lot of people being really intrigued by this tool. And, you know, props to you and your team for for coming up with the idea. But it's it's it might be one of the the very best lead gen tools I've ever seen. I don't know if it's really effective for you that way, but it seems like it would be.

Sales Transformation With Blueprint

SPEAKER_00 21:04

Um well, you know, our customers tend to be um, our customers tend to be uh pretty good sized companies. And so a lot of what we do is we go and we call on them. But what this has done, this has completely changed the way our salespeople engage with customers. A salesperson used to engage with a customer around like slides, and they'd go, you know, four or five meetings often to be able to get a uh a right to have a more technical person come back with a custom demonstration, etc. Now, every salesperson can do a real demo of something that's almost shockingly good in the first meeting, and it's specific to that customer. Yeah. It's exactly and and by the way, if we had put the your website in, it would go out and scratch your website and get all sorts of information about you that it would use for construction of. So this idea of using AI to rethink a process at its core, as opposed to uh either using it only for features. The features are good, but you don't just want to use it for features. You want to re you know, sort of rethink and reenvision and reimagine what you're trying to do, you know. And I I I I think that's revolutionary, and it's something that at this point is unique to us.

SPEAKER_01 22:26

So uh one of the things that occurs to me is like, you know, first of all, there were very few sentences you use to kick off this this whole process. I you know, create a business podcast, I think, and a few other parameters. Um, which is so it's remarkable that it could spit this out just in terms of the complexity and completeness of it. On the other hand, the thing that occurs to me is you know, so many people I've done this, and other entrepreneurs that I know kind of approach businesses, business ideas, concepts this way is you layer in everything you can think of um into your plan. But what ends up happening almost by default, and fairly unexpectedly a lot of times, is um you're not, I you're not really allowing yourself to think beyond the bounds of what you already know. And here, you just put in a couple of prompts, and this machine went and looked at the entire universe of possible outcomes and came back with something that you couldn't you couldn't as a human team conceive of.

Jumping To The “To-Be” State

Partner-Branded Blueprints And IP

SPEAKER_00 23:33

And and what it does, partially because of the way that engine, our engine goes out repeatedly and and looks looks to sources, is it it really tries to get it to be uh state of the art. You know, what are other people doing? What could we be doing? Remember, we were, if you upload information about what you're currently doing, it will try to take what you're currently doing and enhance it. It's not gonna be bound by what you're currently doing. The whole idea, you know, I think this whole idea of first I'm gonna sort of map my as-is, you know, whether it's uh a customer journey or something that my business is doing, and then I'm going to try to envision a 2B. This just blows that up. You just put your objectives in. It really does. It actually takes you right to the 2B. People, people are pretty stunned by this. It's really, I think, game changing. And uh we just introduced something really I'm excited about called partner-branded blueprints. So we went out to some of our big partners, companies, companies like um Amazon Web Services, Cognizant, Ernst ⁇ Young, Accenture, and a set of them, 11 of them, have agreed to put their intellectual property into a private database that only they can see. And when when when one of those people, when a salesperson or a person from that consultant of the company goes to one of their customers, they and only then have access to their blueprint ID. So if they know something about how a particular type of business is supposed to work, or they have best practices that they've developed, they can put that in the database. And when Blueprint grinds these together to re-envision the process, it will take what we do, which you know has a lot of workflow knowledge. Um, it will take what the internet thinks, because it will go out and ask general questions of the internet, and then it will take that particular partner's IP and grind them all together. And the partner and only the partner can use that in conjunction with that customer of theirs. And I think this is going to be, yeah, next year, as this really starts to get rolled out in anger. I think that's gonna be really interesting in terms of letting us not just empower our own selves to add value to our customers, but to enable our partners to add value to their customers using our engine as the technology.

SPEAKER_01 26:08

Extend that out a little bit. And um, and there's a lot of fear we've just created in the ranks of the uh of the consultancy world because, you know, yeah, yeah, facetious, yes. But you know, they just became what, 100x more powerful in what they can do.

SPEAKER_00 26:24

Well, they are. And you know, it's interesting because the whole consulting business is under enormous pressure, um, especially with some of this AI. And you know, you see, you see businesses that say we want to try to reassert more control. We want to try to be, you know, independent of some of the traditional consultancies there. Um, and you can also see them trying to write code on themselves by themselves with some of these very powerful co-pilots and coding tools that exist to help people write computer programs. This, I think, gives the uh consultants something of a weapon to fight back it with, because they can go into one of their customers and say, look, I can bring you my intellectual property, which I've developed over a long period of time. I can marry it with your definition of your problem, and I can get something done four or five times faster than we ever could have done before. Now, the fact that it just happens to run on PEGA, and to actually execute it, they'll have to run it on a Pega system. We don't we don't make them execute it on our system. If they wanted to take what this generated and recode it all by hand, they could. But that would be, I think, kind of kind of foolish when this actually becomes a state-of-the-art running system with with screens and back ends and interfaces and agents. You know, everyone's talking about agents. This this generates an agent that I can converse with. And you don't have to do any of that in the manual way. So that's to me what the superpower of AI is.

Consulting Shakeup And Execution

SPEAKER_01 28:04

Yeah, that's a big deal. One of the themes that has, and I I just said that's a big deal and then skipped right over it. No, that's a really big deal. One of the themes that keeps bubbling up, and uh it I don't see a real cadence to this, but it just keeps showing up somehow is the idea of trust in AI and how it's how it gets dealt with from a user standpoint. How much do we really want to trust these machines? But from a development standpoint, how much trust can you and how do you bake trust into how these things think and what their outputs are? And I had a really healthy conversation, beginnings of a healthy conversation with Rob Walker on that. Uh, but I'd love to hear from your perspective as the the leader of the company. How, first of all, do you see it as an issue? Maybe that's just something that, you know, is is a issue.

Trust, Transparency, And Safe AI

SPEAKER_00 29:01

Oh, I I think it's a huge issue. I think there's trust, I think there's making sure that the company and its tools are delivering value in ethical ways, because there's lots of lots of information AI can use that um candidly violates the law. And that's why we think it's so important to have this uh intermediate stage where you use the AI to design what your business should do. But then when you execute, when you're making the lending decision, when you're when you're making any decision, you need to be able to see how it came to that decision and why it did it. And if Rob talks about transparency, which I think is really key. This style of using AI is very transparent. Additionally, we have to complement that with um, we have agreements with our AI providers where no client data is used for training. So they don't have to worry that either we or the AI company will train on their data. And all of the data is encrypted and belongs to them. So those are the insurance assurances our customers uh have required of us all along, as they've trusted us with their information, and we've now extended that to our use of AI. So absolutely trust is critical.

SPEAKER_01 30:37

Does when you're in these conversations and when your business developments team are in these conversations, the idea of trust always has to come up. It just will, I think. But it seems to me that at this phase in the game, and I'd love your perspective on this, saying that we have addressed trust issues by doing A, B, and C is kind of a check the box. And not to diminish anything you're doing at all, because I think it's it's it's spot on. But um do you get pushback in these situations by saying, yeah, but really? Or yeah, but how do we know? Um in other words, how significant is trust really perceived as by some of these big companies? Are they really concerned, or is it more of just a I need to check that box?

Predictable AI And Production Reality

SPEAKER_00 31:29

No, I think they're very concerned. They they understand the level and they're correct, and the level of reputational risk that could exist from doing something untrustworthy. That's why, you know, what what what we do is we use the AI to create a very creative and curated experience for their customers, as opposed to creating the outcome for their customers. I think there's a lot of anxiety, well-placed anxiety, that if you over-delegate to the AI at the wrong time, you're not going to have predictable outcomes. And we actually talk about predictable AI as being the way we want to be perceived and seen. And, you know, it's done through a variety of ways relative to both the technical infrastructure, to the contractual infrastructure that we strike, to the audits we support when customers want to see how certain things work. So I think it's a very big deal for our clients given who they are.

SPEAKER_01 32:29

Yeah, some very big names. I am I'm privy to uh at least a part of your client list, and they're very, very, very big names with a lot of exposure out there.

SPEAKER_00 32:38

And and everyone is terrified, I think appropriately, of being the next headline.

SPEAKER_01 32:44

Right, right. Yeah, nobody needs that. So we're um we're in a very rapidly changing AI environment right now. It would be completely unfair of me to ask you what you think this space will look like in five years. So I'm not gonna do that, but let's let's bring it back home a little bit. Let's say two years, let's say a year, because um, if I know Pega at all, I know you've already built the roadmap for what that's gonna look like in a year or and beyond. So without giving away too much of the secret sauce, what do you think? What do you think we're looking at?

Near-Term Outlook And Hype Shakeout

SPEAKER_00 33:22

So I think you're looking at um ever more complete uh ability to apply this AI technology safely in businesses. I uh you know, we've really focused on the safe, predictable application. I think you're gonna see a a bifurcation between companies who uh uh did things in ways that weren't safe and have to stop at the experimental phase. That's why people talk about so many of these systems never really make it to production. It's it's not because the AI doesn't work, it's by its nature. The AI that they're using and the way that they're using it subjects them to a lack of predictability. So I think you're gonna see uh some shakeouts here as we go, as as we go forward. The the level of hype around you know, the words AI, the words agentic is embarrassing, you know, as a as a technologist, I find it embarrassing. But there's no choice. You have to use those words, and it's gonna do really important things. It's you know, it's it's not gonna, it's yeah, maybe it will cook dinner for you when when Elon gets his robot working. But, you know, it it has some things that does incredibly well, but you've got to use the right AI in the right place. And that's where we've really tried to focus on when do we want to be statistical? Well, you want to be statistical when you need to be able to say exactly how you came to a decision. You can be generative, particularly when you're kind of designing your business, because you don't have to explain that to somebody. Right. The miracle can just happen. And knowing when to use which uh is really important. And the the thing that makes it complicated is there are some pieces of generative that are completely predictable. If if you if you wanted to translate a word or you wanted to be able to uh summarize uh a document, the AI does that incredibly predictably. You can rely on that. If you wanted to, as we were just doing or talking about, redesign the way you want to engage with the customer, use the AI to tell you. But then you better look at it and make sure that it really matches what you want to do.

SPEAKER_01 35:47

So I was waiting for uh for you to say something like, uh, it's all hype, we should be uh avoiding buying AI-based stocks right now. But but we're not gonna say that because we don't need to do that as a stock market. Um but I'm I'm very keen to keep an eye on how Pega, being one of the leaders in this space, and I think one of the best big thinkers in this space continues to think about AI. Because I think we all have something to learn there.

Closing Thoughts And Resource Link

SPEAKER_00 36:17

Well, I'm I'm I'm I'm glad to both keep thinking and keep sharing. I I will I will tell you, uh I never give stock advice. I always think that's a really bad thing to do. But the the thing I think we will see is that uh there inevitably will be shakeouts, and though that's just the way it goes. But our use of AI, we're already seeing the results. And our customers are telling us that we've made it much easier for them to envision how they should improve their processes and decisions, and that's what it is really for us. Can we use the AI to uh impact that creative process that historically needed to be done through training or by hand, and now be able to do that assisted? And there's no question that this blueprint technology uh helps helps do that in a big way.

SPEAKER_01 37:14

I think uh the word you just used, the phrase the idea that um that these technologies can be used to help other companies envision what could be is the big deal. I really do, and I'm I'm excited to keep looking at this. Well, Alan, I have um I've enjoyed this. I'm glad that I pursued your schedule as hard as I did. And um really can't wait until we can do it again, maybe another six months or a year next time you have an opening.

SPEAKER_00 37:43

Oh, excellent. I'd be I'd be happy to, and there'll be more to show and more to talk about, I'm sure.

SPEAKER_01 37:49

I'm sure. I'm sure. Well, Alan, thanks again. Um, I'm gonna post a link to the Pega blueprint uh tool, which is which is fascinating. It's too bad this isn't a visual medium because if you had seen what I had seen. Uh but again, thanks a million, Alan. Really appreciate it. Take care, man.