What’s the best way to learn AI? Just build something. In this episode of the Pay It Forward Podcast, host Bria Bell, Vice President on the Payments Communications Team, sits down with Zack Anderson, Chief Data & Analytics Officer for Payments and Global Banking, and Alejandra Villagra, Head of Digital & Design, to discuss how experimentation has shaped their approach to AI. Recorded at the March 2026 Global Town Hall, they share personal stories, including building a custom learning app for Zack’s four-year-old son, debugging products through chat, and creating an app that picks Alejandra’s outfit each morning. Together, they argue that personal exploration is the fastest path to professional impact and encourage colleagues to revisit tools they may have dismissed months ago. They also discuss how these lessons are being applied across the business to simplify client experiences and help shape the future of digital payments.
What’s the best way to learn AI? Just build something.
In this episode of the Pay It Forward Podcast, host Bria Bell, Vice President on the Payments Communications Team, sits down with Zack Anderson, Chief Data & Analytics Officer for Payments and Global Banking, and Alejandra Villagra, Head of Digital & Design, to discuss how experimentation has shaped their approach to AI.
Recorded at the March 2026 Global Town Hall, they share personal stories, including building a custom learning app for Zack’s four-year-old son, debugging products through chat, and creating an app that picks Alejandra’s outfit each morning. Together, they argue that personal exploration is the fastest path to professional impact and encourage colleagues to revisit tools they may have dismissed months ago. They also discuss how these lessons are being applied across the business to simplify client experiences and help shape the future of digital payments.
Narrator (00:02):
Welcome to Pay It Forward, a podcast by J.P. Morgan Payments for J.P. Morgan Payments. Get to know our business and gain a better understanding of where we sit at the intersection of technology and finance. Did you know we move nearly $12 trillion dollars globally each day? Through over 170 countries and 120 currencies at more than 5,000 transactions per second? Join us as we take you inside the stories, technology, and people that make it all happen.1
Bria Bell (00:37):
Welcome to another episode of our AI miniseries. I'm your host, Bria Bell, Vice President on the Payments Communications Team. Today's episode features a candid conversation with Zack Anderson, Chief Data and Analytics Officer for Payments and Global Banking, and Alejandra Villagra, Head of the Digital and Design Team. Together, they explore one of the most important themes in AI right now: why the best way to learn is just by doing. You'll hear how both leaders have been experimenting with AI in their own lives, from building their own apps to using generative tools to solve everyday problems. They also discuss why curiosity, continuous learning, and hands-on testing are essential as AI technologies evolve so quickly and how these learnings are being applied inside the business to simplify experiences, improve client interactions, and shape the future of digital payments. With that, I'll hand it over to Alejandra and Zack.
Alejandra Villagra (01:32):
Hello, everyone.
Zack Anderson (01:33):
Hello.
Alejandra Villagra (01:34):
It's my pleasure today to introduce you all to Zack Anderson. Zack, welcome. Zack joined us how recently?
Zack Anderson (01:43):
Nine weeks.
Alejandra Villagra (01:44):
Nine weeks. Okay, so you're still counting in weeks. Zack joined us from NatWest where he was the chief data and analytics officer. He has a super interesting background and I've had the privilege of chatting with him a bit about it over the past nine weeks. Zack, just to kind of kick us off, can you tell us a little bit about how you came to J.P. Morgan?
Zack Anderson (02:05):
I spent the last six years at NatWest and prior to that, I spent almost 15 years at Electronic Arts, the video game company, and a little time in the auto industry before that. I like jumping industries. I like the beginning of things and growing my knowledge base.
Alejandra Villagra (02:17):
Okay. As a parent, I can't help myself. I got to know. What are the video games you're not going to let your kids play?
Zack Anderson (02:23):
Well, right now they're four and two, so I don't let them play any video games unless I make them myself.
Alejandra Villagra (02:28):
Fair.
Zack Anderson (02:28):
I'm going to make them wait a long time, I would say.
Alejandra Villagra (02:31):
While we're on the topic of kids, there's so much news around AI. How do you think about talking to friends, to family, one day eventually your kids about this world that we're in?
Zack Anderson (02:44):
Interestingly, I do, I let that my son talk to Sarah, which is what he's nicknamed ChatGPT and we let it know that. So it always says, "Hi, it's Sarah," and she knows his name. And so I do let him actually talk to her quite a bit. I was pretty amazed. I lived in London and that my oldest son has an English accent, a four-year-old high-pitched English accent, and actually ChatGPT is pretty good at understanding him, sometimes better than me.
Alejandra Villagra (03:09):
I love that. One of the things that we've been talking about a lot just as a community in the payments business, I think within our teams and certainly among one another is the fact that we're at the same place of chaos, lack of understanding, learning curve. This is a brave new world for all of us simultaneously and we're all learning and we're all kind of in many ways starting from scratch. So I want to just spend a couple of minutes talking about how you've learned AI. I mean, you're more of an expert than the average bear, but how are you incorporating AI into your life?
Zack Anderson (03:49):
In a bunch of different ways. I mean, my job is AI, so I definitely put a lot of my effort into staying up with it, but I also just use it in all kinds of ways. And this is an example. My son is four and learning numbers and counting, and trying to understand which number is bigger than the other. He loves dinosaurs and animals and fast things for some reason; everything's got to be fast. I don't know where he gets it.
(04:12):
So I built this little game in an afternoon, one Saturday, where it shows him different animals and he swipes right or left on which animal is faster, has to compare the numbers. And then I made a dynamic difficulty kind of thing in the background, so it gets harder and harder as he gets them right and they get closer together so he has to be able to actually see the numbers. And then he started mastering the one on the right. So I built the one on the left, which is now he's got to rank four of them and switch them around. He loves it. And that's the games that I'm letting him play now. No FIFA Ultimate Team. This is much better. Teaches him math.
Alejandra Villagra (04:45):
And may I ask, what platform did you build this in?
Zack Anderson (04:48):
This is built in Claude Code entirely and actually I have not touched a single line of code in it. It's had some bugs, and I've debugged it just through chat interface and I wanted to see how complicated a product I could build without ever actually writing code. I've written code for games before, but never done it through chat, and it's really amazing.
Alejandra Villagra (05:06):
It is. And I actually built my own app as well. I think the point here is that you don't have to have a PhD to start playing around with these tools. You know, I'm not a developer. I don't code. I started fooling around in ChatGPT about a year ago. It was frustrating, but I tried, and iteratively, iteratively, iteratively until I got to Opus 4.6, it was able to help me build an app. And in this case, listen, I'm a mom. I have lots of different roles and responsibilities. One that I both enjoy but don't like is deciding what to wear. So now AI dresses me, tells me what the weather is going to be. I tell it if I have an important meeting or a casual day or whatever, and it picks out what I'm going to wear.
(05:52):
So I use it every day. It took me a long time of experimenting to get to a point where I had a usable application, but I think what was meaningful and important for me was just the learning curve that I could do it. Now I understand vibe coding and perhaps the point of this conversation is to say, we're all learning, it's okay. Get around, play with these things. They're highly accessible for technical and non-technical people.
(06:16):
And Zack and I just wanted to encourage you guys to find something that is interesting or fun to you, get into the tools and play around because our experience has been that that learning translates so quickly to work. It sparks your imagination, it gives you a sense of what you can be doing, and it starts leading you down this path of how can I take these capabilities and start introducing them into my day-to-day?
Zack Anderson (06:44):
There's two things just to take away I think right now, which is Opus 4.6, essentially since I joined the firm, it's been being used heavily. So in the last nine weeks, the ability of these tools has changed radically. Ask yourself the last time you went and tested your favorite use case. Was it before nine weeks ago? Because if so, the application's changed. And I think you have to constantly be going back and looking at the things you tried before that were too hard, didn't quite work, that's an MLS or in the tools outside or in Claude Code or in Copilot. All those tools are advancing at a really rapid rate. And the thing that didn't quite work six months ago has had four iterations of new tools that have come out since then, and its capability's really evolved.
(07:29):
And so I'd say for everybody, it's not just like they try it once and then decide, "Ah, this is rubbish. I should move on." You got to go back and try it again and again. I couldn't have built that app without coding. In fact, I've built a number of apps using AI coding. I've never been able to build one without any actual in the command line writing code before. This is the first time that that's happened. So the capabilities are really moving quickly. And I just encourage everybody to keep going back and trying things over and over, and try the hardest thing you can imagine, is the other thing I'd suggest. Try things that you don't think it will do and just see if you can get it to work because that's really where the frontier's going and it's quite an interesting thing when you learn what it can and can't do. It's really useful.
Alejandra Villagra (08:14):
I just wanted to pivot to how are we bringing these learnings to work? What are we doing in the digital space as it relates to clients? Zack, you and I have talked about the natural tendency when innovation is ripe to end up inadvertently shipping the org to clients. And that's something that creates confusion, it's poor user experience. And so one of the things that we're focused on in D&D in partnership with Zack is making sure that we have a framework that can take the different agentic capabilities that many, many, many of you are imagining right now and serve it up to our clients in a coherent, easy-to-understand way. And of course, none of this will come to life without deep partnership from Zack and his team. So I'm very, very excited to welcome him to the organization.
Bria Bell (09:01):
One message came through clearly in this conversation. You do not need to be an engineer or an AI expert to start learning. Progress begins with curiosity, experimentation, and a willingness to try. And now I can't help but think about what app I want to build myself. What stood out most in this conversation was how personal exploration can quickly translate into business impact, whether it's building a simple tool at home, streamlining a workflow, or rethinking how clients navigate digital experiences, because small experiments can lead to meaningful innovation.
(09:34):
So, I want to ask you, what is one task in your day that AI could simplify? For me, I sit in a lot of meetings and have to manually write up notes and action items. AI can handle that automatically, so I walk out of every meeting with a clean summary ready to go. You could also ask yourself, when was the last time you revisited a use case that may work better today than it did a few months ago? Or what could happen if you spent just 30 minutes experimenting with a new tool this week? Thank you to Zack and Alejandra for sharing their perspectives and reminding us that learning by doing is often the fastest path forward. Thanks again for listening. See you next time.
Narrator (10:11):
Thank you for tuning in to Pay It Forward, a podcast by J.P. Morgan Payments for J.P. Morgan Payments. If you're interested in exploring career opportunities at J.P. Morgan Payments, please visit jpmorgan.com/careers and enter the word payments for relevant opportunities. Stay up to date on the latest stories, technology, and people powering payments by subscribing on Spotify, Apple Podcasts, or your favorite podcast platform. Keep pushing boundaries, and we'll keep moving the payments world forward one transaction at a time.
Sources:
1: https://www.jpmorganchase.com/ir/annual-report/2025/ar-ceo-letters