Alan Trefler keynote at PegaWorld iNspire 2023: Build for Change: Architect Your Business for AI and Agility

Economic volatility. Geopolitical instability. Supply chain disruption. New advances in AI technologies. Change. It’s all redefining how we work, communicate, develop, and live. The real question isn’t how to handle change. It’s how to master it. The opportunities are endless: How can you use AI to streamline customer experiences and improve employee productivity? What is the best way to modernize your enterprise tech stack to save money and stay agile? How is generative AI transforming workflows?

Join Alan Trefler, Pega’s Founder and CEO, for his take on the latest AI innovations and how Pega’s low-code platform for AI-powered decisioning and workflow automation can help you adapt instantly and start building for change.

Transcript:

Welcome Pega! I know I have, but I think we've all missed this. And I'm thrilled to be here, especially after three years. But this year, this isn't all we're celebrating. We're at our 40th anniversary. We've used 40 years developing AI, developing low code, bringing pragmatic value to our clients, and trying to understand how to handle sophisticated problems in whole new ways. I'll tell you from my perspective, it's been an amazing journey. Fresh every day. And it's all driven and all about clients and partners like you. We're so excited by our engagement with our customers. We find that it stimulates, it rewards, it encourages. Occasionally painful, but always worth it. And what I would say is that one of the things that's important is whether you're new to the Pega family or whether you're one of our longstanding clients, that you really embrace working together with us as we work on the journey that I'm here to talk to you about today and that you'll see so much more of in the next two days. I'd say one of the core elements of Pega's value system is that the thing I get the most pride out of and feel the most responsibility for, is that my first two customers, who went live thirty-nine years ago, are still major, major clients today. And you'll be hearing from one of them. Citibank is going to talk to you. I don't think about that whole 39 year transition.

But I think they'll talk about how they've moved to a really modern cloud-native environment to run the latest AI and decisioning architectures. And you know what I'll say is that during these years, technology has changed a lot, like mind-blowingly. Now sometimes that change is evolutionary, it's steady. Other times, the change is radical, and revolutionary, and disruptive, and threatens both technology companies and also all the businesses in a sector. What I would tell you is that it's clear to us that we are at a revolutionary moment. We're at a moment when the emergence of AI in new ways is complementing our traditional views of AI and providing both enormous, enormous opportunities and enormous risks. And we're doubling down on AI. You're going to see how we're doing it in ways that are intended to help the developer, to help the organizational staff, and most importantly, to help the client.

And you'll see generative AI applied in stunning ways, complementing the analytical AI, which has been core to what we do and the process AI, which is a key element, and the AI-powered next best actions, and so much more. Now, the results here are what matter. The results have got to be better client engagement, optimized processes, and ultimately, outcomes that win for the clients and for your staff. We're at a tipping point. If you look at what's happened with generative AI, think about Facebook. They took 10 months to hit a million users, right? Instagram. Instagram took 11 weeks. Generative AI, ChatGPT, took five days. Five days.

So, there's no question that this is here, it's coming and it's sweeping over our businesses. It's going to influence not just bits of technology, but how people think about their businesses in very profound and very fundamental ways. And what this means is there's going to be a tremendous amount of change. Change in technology, but even more significantly change in the way that business is run. And we need to embrace this. We need to learn collectively to make change a fundamental element of how we operate, a fundamental skill.

And this is what I love, that for the last 20 years, Pega's motto, our tagline, our registered trademark is Build for Change. Our brand promise has never felt more vital. Now, this idea of building for change has been core to our thinking right from the beginning. And it's found in our definition and absolutely critical layer cake approach, which I'll talk about a little bit more. And it brings enormous power to workflow automation and AI power decisioning. Now, look, a lot of concepts of even some of these latest technologies have been around for a long time. I mean, neural nets were being studied when I was back in university. And we've been using AI for years. And you use AI in many ways to enhance client experiences. But the thing about AI is, it's evolving. It's creating new ways of thinking, new ways to complement what already exists.

And I started my career actually very much in AI. I had a success at a major international tournament and earned a master's rating. And it was great. I was a college sophomore. It was a miracle actually. And it led to me to be recruited for a couple of things. I was recruited to place 47 Dartmouth College students on the green. And the bookstore paid 10 bucks to anyone who would beat me. And yeah, that's Leon up there that I'm playing with and he beat me. But notwithstanding that, more significantly... And by the way, I don't think I have that shirt anymore. I was asked if I might wear it. I think that's a good thing. But more significantly, I was recruited to work with the computer science department to try to teach computers how masters thought about chess. How to think about what was then called expert systems, to be able to see if we could get a computer to play at a reasonable level.

And it was a different way of thinking about problems. And candidly, the whole way of thinking about this whole topic has changed massively of course over the years. But at that time, it really impacted my thinking. And I think set a path for the journey that Pega has been on. When I went to work in the big city as a software architect, I was faced with many of the common business problems that occur in one of the world's top banks and looking at how you build systems in this sort of way. And it struck me that using this understanding of how you teach a computer to play chess, it could give you opportunities to apply those ways of thinking to different type of problems. Ways of thinking that might entail teaching the computer to do other types of automation. To capture the business processes and the rules in a mode similar to the way that we were trying to capture how a chess master thought when they played chess. And getting it so that business people, NIT people could work together on a common canvas to be able to have them both in control and both able to influence and change the way that the business worked. And it's this thinking that led to the original concepts of the layer cake. The idea of creating a structure to organize the processes, the rules, all the elements, the interfaces, the technology in a way that brought order to it, that allowed things to be shared. But also compartmentalized certain parts so that you could have common rules and processes across the organization. But specialization where that made sense. And what we've come to realize is that the layer cake is fundamental to new ways of thinking in the face of generative AI. Now, the layer cake is the first of two elements that are absolutely critical to the thinking we want to talk to you about today. The second is what we call center-out. Center-out is about creating engines. Engines that have brains, and process, and case management to record the history of what's going on, that can be dropped into your enterprise.

Not in the channels, because the last thing you want is a business logic in the channels because, well, you can't use it across the channels. And not in the backend silos of systems because you don't want to bury business logic in the back ends. Let's face it, lots of us have lots of that. But when you've got it buried in either front end or the back end, you lose the ability to think of what the business is trying to actually achieve, what the outcomes are that you really want. But if you create engines like this, engines that are powered and built on that layer cake concept, you've got the ability to capture the intelligence and execute it.

But it's really important to not think of this as a monolith. We've built it especially to operate in a distributed fashion. And the idea, the vision is this should be part of a connective tissue, a part of a fabric, a micro-services architecture as it were, with a collection of engines, running perhaps in multiple regions to be able to handle local data requirements. But all running off of a common set of layer cake definitions, so that the way you run your business, the way you implement change, has both the locality, the power of being local, but also has the power of supporting global concepts and the ability to change rapidly at a global pace.

Having a distributed architecture provides incredible agility. Having a layer cake to capture the logic of your businesses, is what delivers a coherent experience to your customers wherever they are. And to your staff to make sure that what they do is always on target. Because the way we describe this, is you want to run your business as a process fabric. You want to think of the outcomes of your business and weave together all the core elements. So, Pega, and you'll be able to see it in the innovations hub, has a process fabric hub. Which acts like a Google index for all the work in your organization. By the way, it can include non-Pega work as well. It allows you to see it from a customer point of view. What am I doing for this customer wherever I'm doing it? And what did I recently finish for this customer?

Or it allows you to assign, and operate, and manage it from an operational point of view. How do I get the right piece of work to the right staff member at the right time to meet service levels and do everything we need to be able to do, to ensure that the staff is always at its most productive. Facilitated by AI in the way they do the jobs, but organized by a fabric that ties together the different objectives, helping you get work done most transparently to clients and most effectively. Now, you also want to start in some areas and continue doing this, but do this in a way that's organized and structured. And our solution to this is what we call the low code innovation factory. And the low code innovation factory is what allows you to, well, do this at scale, to bring in new ideas, to leverage what already exists, to be able to support quality and testing, and we use at an enterprise level. And this is how we think about low code. And candidly, it's not the low code of others. This is the layer cake, controlling the low code and accelerating the adoption of these engines through the power of AI. And the best thing about this is you can have an enterprise that becomes self-optimizing, where every piece of it can be evaluating how it's working, looking at standards, minding what's going on with the processes. And we call this continually optimizing enterprise driven by the layer cake, driven by the center out engines, we call that the autonomous enterprise. And it is going to change the game. It makes automation seamless, it brings people in when and where people are needed. And Pega is uniquely positioned to help you achieve this next level.

Now, to make it happen, we are using our AI. The AI we've been using and a massive introduction of capability around generative AI. And we're upping the pace and are tremendously excited. You'll hear from Karim, and Rob, and Don about real specifics and be able to see them in the innovation hub. And what's interesting is I think the emergence of generative AI is going to be absolute destruction for about 80% of the low-code players. Because they're going to discover that they're not adding structural value and people will just generate code. But for Pega, we think we are uniquely positioned to actually turn generative to enormous advantage. So, I'm going to explain to you why we see ourselves as uniquely positioned. And that's because there are certain things that really go well with AI.

And one of the things that goes really well with AI, is cake. You see, even AI likes the cake. Cake is the secret to mastering generative Ai. And our patent layer cake organizes the pieces of your system, the workflows, the decisions into these layers that allow you to reliably add and build, but also to change it, which is one of the most critical elements. Now, this is very different from other approaches. Other approaches are going down the path of saying, "Oh, I'm going to use all these wonderful prompts, isn't it a miracle? Prompts they're going to generate code. And the code, I'm going to stick it together somehow."

And candidly, if you've got small apps or systems that aren't very sophisticated, this may be fine. But the problem is your logic is buried in this code and more code, and more code, which inevitably means more bugs. And the reality is, more disconnected apps. As this happens, organizations, some are already finding that they're accumulating lots of very shoddy code, which in the technical realm we call this technical debt. And the real killer is you do this, and then you want to go back and you want to change a piece of it.

You can't reprompt the whole thing. How do you control, and constrain, and manage what you're changing? How do you have a safety net to let you see what you've done with some sort of, well, vantage point? That's where the Pega approach is radically different. Yes, we're using gen AI. You'll see it used big. But we're using it to augment the layer cake. We're using it not to generate the code, but to generate the rules, the business logic, the AI principles into the cake, into that structure. So, you've got a place where you can look at it, and reflect on what it's doing, and actually show an auditor what you're running without showing them a bunch of prompts. Imagine how that's going to go. And this then becomes the basis for the most important thing, which is being able to change it, and being able to weave it out and control it across your business to make it percolate across the enterprise.

Now, this is candidly a big idea and it was empowered by the architecture that we uniquely have. So, we look at what's potentially a very threatening moment for lots of organizations and see pure empowerment. And we're doubling down on bringing our development environment, the power of generative AI, the power of an autopilot, the power to make AI-driven development natural. To be able to put Pega skills at everyone's fingertips, where the assistance comes to you, seeing what you're doing proactively guiding and automating. We expect the developers with Pega can more than double their productivity to deliver much more capability, much more rapidly. And this is just one of the revolutionary changes you're going to see today and with what Pega is going to be doing over the next year. Now, I think we have the chance to radically improve how people run their businesses. And that this is truly a moment of Genesis. Truly a revolution. Empowered with our history, our knowledge, the awesome customers that we get to deal with.

We are, I believe, terrifically positioned to lead this revolution. But like a lot of revolutions, you have to be alert to some of the dangers. And Rob Walker tomorrow will talk a little bit about that. But just to give you a bit of an example, if you were to ask AI to draw the moment of creation, the Genesis spirit from the Sistinem Chapel, you might get this like we did. Which I'm not sure how it got there, but I would say is more than a little deeply disturbing if this is what AI is going to do with us and to us. So, we need the control, we need that cake in the middle, and we need the right architecture to be able to make this work. Because if you do have that right sort of set of controls and you do have the right power, you can really bring an artistic vision to your enterprise. And you can take this incredible moment of Genesis to the next level.

And that's what Pega wants to do with you. We want to partner with you, we want to understand the needs, we want your feedback on what we are doing, and really take what's going to be an amazing, amazing opportunity, and realize it fully and broadly. We're going to help you, and you're going to help us, and it's terrifically exciting. But it's not all future certainly. With what we are showing you today, which is a tremendous of real technology available this summer, and what we'll be doing over the next year. When I look out to Pega World a year from now, let me tell you what I'm expecting and what we're working for. First, everybody will say that Pega expertise is at their fingertips. This long-standing issue that sometimes you can't figure out the PEGA resource you need to do something, we're going to build that right into the machine. We're going to double developer productivity by making both mundane things, well, go away, and making complex things guided and more readily available. I believe we'll see the emergence of what we call Pega Launchpad, which is a product geared and aimed at our partners for them to be able to create and capture their intellectual property. So, that if they have workflows and decisions based on that, that they're going to be able to bring that to market as sets of products running on Pega Cloud. And by the way, I've been working with it. It's really fast.

And finally, we're going to together see the emergence of the continuously optimizing autonomous enterprise. It's going to be an awesome year. I think it's going to be an absolutely terrific show together. And thank you so much for the 40 years. And thank you so much for coming and being with us today. Have a great PegaWorld.

Tags