From Vision to Value: Execute an AI Strategy with Purpose

Transcript

Virginia Heyburn (00:03.509)
Hi everyone and welcome back to another episode of FinTech Unleashed where we explore the latest technologies and strategies shaping the future of the financial services industry. My name is Virginia Habern. I'm the Director of Research, Insights and Advocacy at EngageFI. Today we're diving into a topic that's redefining how banks and credit unions think about their strategy, their operations and the experience that they want to deliver to their customers and members. It's artificial intelligence. AI is everywhere right now. It's in the news headlines. It's in vendor pitches. It's in boardroom conversations. The potential of AI is very exciting, and that's why we're talking about it so much. But the path forward is far from simple. What does it really mean to build an AI strategy that's grounded in sound governance?

Robust data practices, and ultimately long-term value, and not just technology for the sake of technology. My guest today is going to unpack all of this for us. My colleague, James White, is back on the podcast to share his wisdom. James is our principal strategist, and he's a recognized expert in AI strategy and innovation for financial institutions. We are going to go deep. We're going to talk about why AI is harder than you think, why no institution should ever go it alone and how to separate real capability from the hype that's in the marketplace right now, now that we're suddenly flooded with AI powered everything. Let's dive right in.

Virginia Heyburn (01:38.891)
James, welcome back to the podcast.

James S White (01:41.286)
Well, thanks for having me.

Virginia Heyburn (01:43.617)
It's wonderful to have you again. I know we recently had a terrific conversation on CRM, obviously very closely related to AI, and today we want to go deeper on the artificial intelligence front.

James S White (01:55.106)
looking forward to it. I've been excited about this for a couple weeks now.

Virginia Heyburn (01:59.775)
Yeah, it's fun. Always fun to talk to you. Can you give our listeners an idea of your background just in case anybody is on who didn't listen to our last podcast?

James S White (02:10.763)
Yes, absolutely. So my name is James White, as Virginia mentioned. I'm a principal strategist here at EngageFI. And my focus right now is helping institutions bridge strategy with execution around AI, CRM, data, and customer member engagement. So everyone's chasing technology these days. But the real differentiator is how you deepen relationships, not just automate them. So it's really a balance of high tech and high touch.

Virginia Heyburn (02:39.553)
So, and you're in demand, James. We keep you very busy and out on the road with financial institutions. I'm glad you had time for us today. Let me ask you something. There's this sense that artificial intelligence has suddenly become incredibly urgent, but so many of the core technologies that go into artificial intelligence, they've actually been around for many, many years. What has changed in the last?

James S White (02:47.277)
Thank

Virginia Heyburn (03:05.377)
12 to 18 months, it makes this moment so different for banks and credit unions.

James S White (03:10.637)
Absolutely. As you mentioned, AI isn't new. What's new is the access to AI. So large language models and open APIs, they used to only live in R &D labs. But for the first time, community banks and credit unions can really apply AI and not just vaporware. So the leap has been usability. We've moved from predictive analytics to generative intelligence. And that can explain

create and adapt in real time. So the promise is not robots running banking, it's humans running smarter banks.

Virginia Heyburn (03:51.158)
And it's also this ability now for smaller financial institutions to compete directly with large financial institutions that do have the deep pockets to pay for very sophisticated technology. What I'm hearing you say is that sophisticated technology is now imminently available to a smaller financial institution.

James S White (04:11.469)
Absolutely. And it's just trying to navigate because the first mistake is everybody thinks that AI is a software purchase and it's not. So it is functionality that has been embedded in platforms as well as agents that are available to you all through ChatGPT or Copilot, know, Claude, some of the others. So the other real big mistake that institutions are making is

just trying to build it themselves without understanding what the compliance requirements are. So regulatory requirements are very gray right now. So making sure that your auditors are going to feel comfortable with the way that you're using it and the way that you're leveraging data. it is really great opportunity for us, but we are at the beginning stages. making sure that we develop a strong strategy and understand how we roll it out in a very methodical way.

Virginia Heyburn (05:10.389)
Yeah, and want to talk more about that, but just to level set, James, there are different types of AI. We've talked about AI that's been around for a while. You mentioned generative AI. Then there's the whole strategy around agentic AI. Can you just quickly, for our listeners, level set, what are the differences, and are they equally accessible to the smaller institutions, as you mentioned before?

James S White (05:35.234)
Yeah, so first off, I'd like to say that I wish that whoever was creating these AI models or types leveraged marketing because the words are so close together. It's so easy to get confused. generative AI is creating outputs that allow you to interact better. So if you think about building emails or content or doing research for you.

Agenic AI is actually able to do an action. So that's if you're thinking about connecting to workflows and being able to make decisions real time. those are the big differences is generative AI is what has been around for a long time. Think of chat GPT as an example, which is probably the most prominent. And agentic AI is what's coming out and being more accessible every day. But

What that's allowing you to do is take actions, leveraging AI that are going to learn from themselves and be able to get better and better.

Virginia Heyburn (06:40.385)
And it's really important to understand those differences. I think when we talk about where financial institutions should start, and we'll talk about that more later, but it's getting fluent. It's kind of a first step is really understanding what it is we're talking about here. What is AI? What are the different types? What meaningful benefits can they deliver to a bank or a credit union of any size?

James S White (07:04.109)
Absolutely. And we as an industry flow with trends. And so there's been many trends over the years. Omnichannel experience was one for quite a while. There was CX strategies or customer member experience strategies. Now two of the big ones are AI and CRM. So everybody tries to jump on, but it's so much more complex than it seems.

and understanding what the vendors are offering and how it's different and how it's going to apply to your organization can be very complex. And there's so many AI companies that are just starting up right now as well that want to go after the banking industry because it seems like a large market and they just don't understand banking enough.

Virginia Heyburn (07:46.902)
Mm-hmm.

Virginia Heyburn (07:54.112)
Yes, and so then as a result, building and executing an AI strategy is really not that simple. You say this over and over again, and we see a lot of banks and credit unions stumble on their AI journey and then course correct. What are some of the biggest mistakes you've seen banks and credit unions make, and why should they never try to do this alone?

James S White (08:18.005)
No, that is a great question. So the first thing an organization should have is an AI policy. So understanding how your users can leverage it, what types of data can be put in into the AI, making sure that it's configured so that it doesn't leave your organization, that there's no PII or MPI data that's being included. That's the first. The second.

A big gap I see is updating your vendor and procurement processes. So making sure that you understand what vendors are doing, making sure that if it's touching a customer, remember that it doesn't have any inherent biases or couldn't have any inherent biases. So there needs to be additional items added to your vendor process to ensure that when you're selecting something with AI, that it's going to be a compliant product.

Virginia Heyburn (09:12.147)
And you often emphasize that we can't just think of artificial intelligence as a technology consideration. There really does have to be this overarching strategy around the business. Can you talk a little bit more about that in terms of what problems financial institutions are looking to solve and where artificial intelligence can really fit the bill?

James S White (09:12.621)
you

James S White (09:35.704)
For sure. So it starts with intent. So what do you want AI to improve? Do you want it to improve efficiency, reduce risk, personalization, growth? Identify where your data lives and how it's governed and who owns it. And then most AI failures trace back to unclear ownership. If your data is fragmented, your insights will be fragmented. It's the old adage, garbage in, garbage out. So strategy means knowing your questions.

before you automate your answers.

Virginia Heyburn (10:08.651)
I love that line, that's really key. And then before you even go down the road of evaluating a single tool or a platform, that strategy piece, the ownership piece, the data piece, that really has to be in place before you start selecting these tools, correct?

James S White (10:25.649)
for sure. And I've helped organizations actually create separate data sets just to be used for AI. That way it can be done in a secure way that's a strong normalized data set that you can ensure is going to teach the AI the right way. But then even when you do that, it's all about managing it and maintaining it. So it's not, I do it once and forget about it.

Virginia Heyburn (10:53.685)
Yes, so you get your data in order, you have a strategy, you have a governance model in place, and now you go about really talking to the vendors, as you mentioned before. And nowadays, you look at the marketing materials, James, I it says AI powered somewhere in there. And to some extent, it really is AI powered, to some extent, maybe not. But from your perspective, let's start with the applications. Let's start with not the solutions, but what

James S White (10:55.607)
you

Virginia Heyburn (11:22.471)
What AI can be used for in financial services? Think loans, think deposits, think customers. How can leaders cut through the hype to understand what's really under the hood of these solutions?

James S White (11:34.304)
Absolutely. So the first thing you want to do is forget the flashy stuff. The power is in personalization, fraud detection, predictive engagement, and just operational workflows. So the next generation of AI, it's not going to be able to score credit, but it will be able to score intent. So it'll be able to tell you which depositor is likely to move funds in the next month, or which borrower has had a life event.

and mentally ready to enter the market. you really want to, from a vendor perspective, try to get through the noise. Understand, have they just connected to chat GPT as an example, which a lot of them have? And if so, how are they keeping that data secure? Are they ensuring that the data is not going offshore? And if it is, how are you going to, from your data strategy, ensure that nothing gets into that model?

that has MPI or PII data. And then once you understand if they're building their own models, how are those models being created? Is that something you're going to have to do? Or do they have a set number of options that you can pick from? As an example, when you're looking at chat bots or voice AI agents would be a great example. Do they have options for you to choose from? Can you choose from this 100 selection pick list?

Or you have to go through all of your calls at a contact center and understand how you want to communicate. So there's a lot to it as far as understanding what AI they're offering, what are the data requirements, and where is that data being stored.

Virginia Heyburn (13:19.009)
And you mentioned, James, probably 15 questions that should be asked, and there are probably 100. And I think this goes back to what you had said earlier that going it alone with artificial intelligence, complex, new, impactful, high risk, going it alone is just not a good idea. asking these right questions requires a skill set that is very different when we compare it to, say, evaluating a digital banking solution. Is that fair?

James S White (13:19.604)
Thank

James S White (13:48.722)
yeah, absolutely. Well, and a lot of organizations use buzzwords like responsible AI and things like that. Well, responsible AI does not mean compliant AI. And so you really want to be able to understand how their team is structured, how their team is learning because AI is at its worst spot right now that it will ever be. It's getting better every minute, every hour, every day.

And so we as buyers or users of AI have to stay up on how it's evolving, but the vendors need to even more so. So understanding the investment that they're making, is this something that they are offering just to check a box so that they can say that they have AI on a marketing slick? Or is this a major strategy for the organization where they're really evolving the organization and their platform? What kind of migrations process?

their bias mitigation processes, how's the framework working, all those things that your regulators are gonna start to ask once they get more comfortable with

Virginia Heyburn (15:01.601)
And when you sit in these vendor evaluations, James, I'm guessing you've had your fair share of those and you see these vendors presenting their artificial intelligence capabilities, be it embedded AI or their outright, you know, just AI vendors, vendors that didn't exist a couple of years ago because the market just wasn't here yet, the cottage industry of AI vendors.

James S White (15:03.853)
Thank you.

Virginia Heyburn (15:25.129)
What are the red flags that you're seeing? What do you get most concerned about when you see the vendor presentations today in front of banks and credit unions that are really trying to step up to the plate and gain this knowledge that's so essential?

James S White (15:39.65)
Yeah, so the first one is kind of funny, but the first one is, do they really know banking? And you can really tell if the organization knows banking just by the way their slides look. If you're a credit union and it doesn't say member, it says customer. That obviously shows that they aren't familiar with the credit union industry. Do they have things that are just named a little bit off? Primary account holder is a great example.

We in banking know exactly what that is. But if they're calling something else a primary account holder besides what we know it to be, that is a red flag. The other is if they're promising immediate ROI. So true AI takes calibration. It's going to take some time. That ROI is going to compound over time. It's not like just scratching off a lottery ticket and you either win or lose. So you want to make sure that you're

really factoring that in. is crucial. Used to, ROI was just something that a platform used to sell. And now the leaders at the financial institutions are being held accountable to those ROIs. And so if you're going to sign up for an ROI, you want to make sure that you feel comfortable on how that's going to get there. there's a lot of research out there right now. think the Boston Consulting Group, I know the Boston Consulting Group.

released a paper that said only 10 % of the organizations right now, financial institutions have seen the ROI that they were expecting. And so a lot of that has to do with not just the platform, but actually the organization itself, just like what we're talking about, making sure that you have a good strategy, making sure that you know what data and what you're training it, making sure that you have great policies.

policies for governance because there's so much that goes into it all the way from data to culture, making sure that your employees feel comfortable using it and that they're embracing it, all those things. mean, so AI is going to transfer and transition the whole industry and really the whole world. I saw a presentation this week just talking about how there have been so many different

James S White (18:05.121)
key items that have occurred in recent history. One of them is reusable rockets. When you think about the safety, I mean, think about the space race and now AI is going to do it for financial institutions, but the whole world. And we've got to embrace it, but there's a lot there that we need to make sure that we're checking the box on.

Virginia Heyburn (18:28.351)
and checking the box and how does a bank or credit union know, James, if their data infrastructure is ready for the AI solution that they're looking to invest in and how do they know that their governance model is mature enough for that AI solution that they're evaluating?

James S White (18:47.189)
Yes, well, that's a great question. So the first thing I would do is just assume that it's not ready. One of the best ways, and this is not any fun, but one of the best ways an organization can be close to ready is having recently done a core conversion because of all that data cleanup that needs to occur before the core conversion. So I'm always a great example, James. Is there a James, a Jim, a Jimmy?

Is Jimmy my son or is it me all those kinds of things need to be? cleaned up and so You can't have data be the afterthought it has to to be consistent data hygiene and ultimately owners for each of those areas of data Because how data moves through the systems AI feeds on context not chaos and so clean data is super important

Virginia Heyburn (19:46.24)
And I'm thinking, you know, the core conversion, that makes so much sense to me. And James, I've actually not heard that said in that way in the marketplace, that that's a really great time to be doing this. If you're in the middle of your contract, you've got three, four years until you are gonna either stay on your system or convert to a new system. It really does pay to start looking at that a little bit earlier, considering the competitive dynamics.

that we face as an industry. When it comes to AI, we can't afford to wait three or four years.

James S White (20:18.151)
Absolutely not. And we need to be going through as an industry and treating data differently. So I always tell people that financial institutions over the years have treated data as a liability. And fintechs and neo banks, they all treat it as an asset. And so we spend so much time being concerned about someone getting access to the data that shouldn't that we don't focus on.

making sure that we have good data and clean data and that we can leverage it ourselves to interact with our customers or members in a better way or understand the financial health or wellness of our customers or members to be able to support them through their life. There's so much that's there that's available. But a good data governance framework is important and we need to make sure that we

are ready to leverage it because we've got to to compete and that doesn't mean big brother that just means consumers expect us to understand them well enough to to make offers or have interactions that feels like they that we know them or we have a desire to know

Virginia Heyburn (21:38.55)
And I'm glad you brought that up because this new generation, Generation Z, very soon the largest generation, they really do have entirely different expectations of a financial institution. They're not necessarily looking for a transactional relationship. They are looking for an advice-driven relationship. And that's gonna require banks and credit unions to have pristine data and put that pristine data to use in...

wellness services that are available on the device of choice for these young members and customers.

James S White (22:09.911)
could not agree more. I could actually have a whole other podcast talking about what I call the golden record, getting all of the silos within the organization, the data combined so that, one, you have a unified message that's going to the customer member, as well as understanding exactly what products and services and interactions that customer member have with the organization.

Virginia Heyburn (22:36.405)
I've been in this industry for a long time as well and I remember 20 years ago, us sitting around in groups saying, you know, we just need better data in this industry. If we had better data, we could do so many cool things. And now you have just, of course, this urgency around it. It's really gratifying to see that the industry is now in a position where we do have to do that cleanup, that long overdue upgrade of our data philosophy within this industry.

that is going to, I think, just yield so much benefit and allow small institutions to compete with large institutions in fintechs.

James S White (23:10.909)
I could not agree more. you one of the problems we have as an organization is there's not one source of record. We all think of core being that source of record, but there are a lot of things that reside outside of core. Credit cards, as an example, sold mortgages as another leads from different platforms. So you really have to think about it as getting your LOS is connected, your card system's connected.

even in, I call ancillary products, insurance and wealth, you know, connected into a unified data source.

Virginia Heyburn (23:47.182)
That makes a ton of sense. we have to find something to disagree about, James, because we've just been agreeing with each other this whole podcast. And so let me ask you a question that maybe is a little bit controversial, because I see a lot of disagreement. And it's about whether or not AI is going to replace people in the industry, replace tasks for sure. How do you see artificial intelligence replacing roles within banking and credit unions, but also

know, reshaping employment in our industry.

James S White (24:20.471)
So that's a great question. actually spoke at a conference this week about this topic specifically. So AI is not going to replace bankers. It's going to replace tasks. So the new differentiator is empathy powered by insight. So trust and empathy are all coming from humans. They're not coming from technology. So being able to take that additional capacity and train those employees

to have different, more meaningful conversations. So we've spent so much time over the years teaching our employees how to handle the transactions themselves that oftentimes we haven't taught them how to engage and have meaningful discussions with our customers or members. so there are definitely roles that are going to evolve, but I don't believe that it's going to replace employees. But it is very important that the organizations

develop a training strategy to one, understand how roles are going to evolve and to help lead those employees and train those employees to be able to have the newer updated skills. You think of a loan officer with predictive tools now becomes a financial coach or a contact center rep now becomes a relationship manager.

Virginia Heyburn (25:44.363)
Mm-hmm.

James S White (25:47.563)
What AI will eliminate is human inefficiency. So it's not going to replace humans, but it will eliminate human inefficiency.

Virginia Heyburn (25:57.644)
I was talking to a CEO yesterday and I think it's very interesting in the context of AI. She was making the point that graduates from college today are ready for so much more than graduates from college say 25 years ago. They're really looking to step into a much more I guess complex role than say entry level. And what we see in banking and credit unions take the contact center, right?

These entry-level positions, they may actually be reimagined, but that's gonna be to the benefit of the people coming out of universities and training for those entry-level positions. The positions are changing, but so too are the people that are looking to enter the workforce.

James S White (26:40.661)
I could not agree more. the one big thing that we have to do is make sure that our culture, which is typically change-inverse, doesn't keep us from being able to evolve into what the new financial institutions are going to look like, because we want to be the ones who are acquiring not being acquired. And it's important that we can evolve.

Virginia Heyburn (27:04.767)
And do you feel like we also have to leave some strategic room for this idea that we just don't know what we don't know? And what I mean by that is, James, you look at the news, again, headlines, And you see these data centers that are being built, being planned, just enormous consumption of electricity. It makes me wonder sometimes, they're building for something that they don't want to tell us about. How do you feel about having that element of the unknown?

as part of a working strategic model in a bank or credit union.

James S White (27:37.932)
Well, it's, I agree with you completely. Actually, I saw where China was building data centers under the ocean so that they could keep them cool. yeah, it is really, it's very exciting, but nerve wracking, you know, to your point. So one of the things that we've always done is create, you know, three, seven tenure long-term plans. And I think we're going to have to be able to pivot more because if someone tells you,

what AI as an example is going to look like five years from now, then you don't believe them because nobody knows. I don't think Sam Altman knows. And so you got to make sure that you can pivot and evolve. And that's just not something that we all are used to. We're all used to doing the same things and getting the same results. And if we're not careful, we'll be left behind by those who can't pivot.

Virginia Heyburn (28:35.637)
And that's a red flag too in those vendor conversations, right? If you're talking to a vendor and they present themselves as so confident that they know what's gonna happen in five years, then in my mind, they're not demonstrating that ability to pivot or that baked in cultural acceptance of the fact that we have to pivot.

James S White (28:54.317)
I agree. So I've been in the industry for a long time, and I know you have as well. We used to see 12-month roadmaps by organizations constantly. Well, what I've started seeing now is with Fintax and newer technology, you're looking at quarter-long roadmaps. And so that is opening them up to be able to pivot. It does make it a little bit more difficult for us as

Virginia Heyburn (29:13.547)
James S White (29:23.351)
customers or users because we want to know what's rolling out by the end of the year or next year and it's very difficult for them to commit because it's changing and evolving.

Virginia Heyburn (29:34.818)
That's happening very fast, but there's also a risk associated with rushing, right? When you're in the seat of a bank or credit union, chief information officer, chief executive officer, the feeling is we have to do something in AI, we've got to understand it before we do anything, we've got to clean up the data, but hold up, that's gonna take too long, let's go in quicker before we've really fully assessed the risk. What's the...

What's the real urgency to do something with artificial intelligence before the competition does? And what's the danger of rushing it?

James S White (30:10.144)
Absolutely. So rushing your AI without governance is like wiring a house without turning off the fuse box. So the danger isn't just inefficiency, it's liability. So AI will magnify your strengths, but it also magnify your weaknesses. And so there are ways to roll out AI. One of the things that I would encourage individuals to do is lay out impact versus risk. So there are some real low risk ways that you can roll out.

roll out AI that are going to make a huge impact for your organization. One of the ones I talk about a lot is just creating a copilot agent that has all of your processes, policies, and procedures in it so that an employee that doesn't know how to do something can ask that agent and get information easily back. Well, that is a low risk, high impact way.

rolling out some agentic AI or anything that touches your customers or members without having a policy in place and governance in place is too high risk from my perspective.

Virginia Heyburn (31:15.113)
And are new roles required within banks and credit unions to be able to put those governance models into place? Do we need more resources for that?

James S White (31:26.354)
I think so to some extent, we definitely need a AI governance committee, so that you have leaders from multiple departments, because each of those departments are going to have competing priorities. And you've got to make sure that you're making those decisions as an organization and standing behind it. But some of these platforms that are really doing

high-end agent AI, you're going to need admins for those to keep them up and maintain them. I mean, even with the copilot agent that I just mentioned, as your policies and procedures change, you're going to have to keep those updated. Otherwise, they won't be teaching the employees how to do it wrong or if your technology changes, all those things. there are going to need to be more employees from an administration perspective.

I always, this is a little off color, but I always say one throat to choke. So you need to have somebody at your organization that is going to stay up on the regulatory requirements for AI, the new technologies out there so that you make sure that at least you have somebody who can help you sift through all of the noise.

Virginia Heyburn (32:42.901)
And is there also the onus on marketing or generally the leadership team to communicate to customers and members that artificial intelligence is making its way into the organization, particularly into how customers and members are engaged?

James S White (33:02.22)
Yeah, absolutely. So I think there's a big thing about how customers and members are engaging and understanding how that's changing and evolving. Because not just how they're engaging, their expectations are evolving as well. And so I call it Amazonification. Amazon has changed the way consumers expect to be engaged with. They want it fast. They want you to

predict things that they're going to want or need. And so those things are very important. And the days of us just opening the doors and growing by 20%, those are over. You we've got to really get back to relationship banking and leveraging high tech, high touch, know, technology and human to separate ourselves because otherwise the younger demographics just aren't as loyal.

as they have been over the years. They don't know Sarah, the branch manager, because they go to school together or their kids go to school. You just don't have that. So now you've got to build that trust through those in every interaction. Every interaction should be treated like gold from my perspective because you don't know how many years ago.

Virginia Heyburn (34:20.513)
And from my perspective too, mean especially this next generation, Gen Z, let's give them credit. They're probably a lot more familiar with artificial intelligence than we really give them credit for. as we sit around worried about how do we let them know, hey, this engagement is powered by AI or supported by AI. Meanwhile, mean Gen Z, they've used Spotify, Amazon, all kinds of retail.

James S White (34:46.702)
you

Virginia Heyburn (34:48.085)
applications that have been predicting their needs for years, right? They're using Apple Intelligence. You know, they can take a picture of a tree and immediately be told what it is, how to take care of it, et cetera. There are just so many use cases. They're using ChatGPT like crazy. I mean, I'm using it too like crazy. just love it. James, I don't know about you. So we're using this. I mean, we go to the doctor's office and the tools that are being used are using artificial intelligence to give us.

James S White (35:08.151)
Yes.

Virginia Heyburn (35:14.837)
better care and consumers are really aware of the benefits of AI, I think they're going to take that message in a really positive way as long as it's transparent.

James S White (35:27.136)
I could not agree more. there's actually a lot of consumer research out there that speaks to how consumers are embracing AI. And there's a lot of fear out there from a institution perspective, how are my customers or members going to feel about AI. And don't get me wrong, there's definitely going to be some who don't like it. But there are also some who don't like talking to your employees. So you never can please anybody.

or everybody. So you want to make sure that you're embracing it because there is so much to be gained. And to your point, the younger demographics, they are very used to it. I mean, a lot of these bigger players have been leveraging AI for years. Apple, Facebook, I know if you ever have a conversation with someone and then you go on Facebook two hours later and there's ads pop up.

That's that's AI and that's been going on for for many many years. So

Virginia Heyburn (36:30.641)
couple of questions for you James to wrap up our conversation. It's just been fantastic. Now I want to challenge you to look out into our industry five years from now. What is one change that AI could make to our industry that most leaders aren't even thinking about right now but should be?

James S White (36:52.044)
Yeah, I think that it's probably going to be predictive, proactive predictive banking. helping customers feel like that my bank or credit union knows me is going to be table stakes. so being able to support your consumers during major life events, which is where when somebody typically will make a major financial purchase and being able to

to know that and communicate with them without them having to reach out. So I think that truly integrated predictive banking is where we're going to be proactive. Not just next best cross-seller, next best action, but I'm talking about omni-channel, being able to be there to support your customers or members when they need you most.

Virginia Heyburn (37:43.392)
I love that, the bank of one, a personal banker in everybody's pocket, right? And then my final question, if you could give one piece of advice to a financial services CEO right now who feels both excited and overwhelmed by AI, what would it be and why?

James S White (37:47.03)
Yeah, absolutely.

James S White (38:02.38)
Yeah, so I would say start with curiosity, not fear. So you don't have to know everything. You just need to know where to begin. So start with a small pilot that solves a real business problem. It could easily be just that copilot agent that I mentioned earlier. That's something that you could build in 30 minutes if you have your policies and processes and just see how that business problem of high turnover of frontline staff and the training of

them help just by having that agent available. So take the business problem and just measure it ruthlessly because AI rewards momentum, not perfection. And so just make small strides. You and I were talking about how much we use chat GPT. I didn't start using chat GPT.

really until a couple years ago playing around with it. I didn't know what it was about. And all of a sudden now I'm using it throughout my day all day. So it just takes a little bit, just one step forward in a small low risk way.

Virginia Heyburn (39:11.009)
That's really great perspective, great insight. James, I use chat GPT for so many things specific to my own life. Like give me a homework plan for a seven year old that is survivable for this mom right here. And it's amazing what it spits out. It's just amazing. Journaling questions, just all kinds of things. You you wanna lose 10 pounds and...

two and a half days, it has a plan for all of us. All of this is possible with Chachi PT. Now some of it doesn't work so great, which is why we're still around, right?

James S White (39:45.167)
Exactly. You still have to use the human side and some of it's going to be wrong. You have to be able to discern that for sure and don't just use it blanketly. But I have built agents for myself for all different types of things for school and homework, just as you mentioned. For church, I do some church stuff. mean, part of some not for profits that I have agents for. And you just

teach it specifically what you want it to know and it's absolutely amazing what it can provide and it does provide more and more correct information than incorrect information over time.

Virginia Heyburn (40:28.267)
And like you said, better every day, right?

James S White (40:30.318)
Better every day. So I wish I could be better every day. You know, I had a football coach that would always say 1 % better every day.

Virginia Heyburn (40:38.465)
Well the nice thing is we do have more time thanks to artificial intelligence and that's something that banks and credit unions can really put to use to convert some of that time that they find into sound strategy and being much more competitive with fintechs and with their larger competitors.

James S White (40:58.454)
Absolutely. And one last thing that, sorry, just came to me. Financial institutions employees are probably using chat GPT, just like you and I said. And if you do not have a policy in governance in place, you could be putting yourself at risk because if NEPII or NPI data gets out there, that could be really catastrophic. So you want to make sure that even if you're not planning on implementing AI.

that you have a good governance policy in place because your employees are using it today.

Virginia Heyburn (41:30.143)
Yes, yes and I love what you said earlier, know, responsible curiosity over fear. I even wrote that down. That's really what stuck out for me. I've learned a ton in this conversation, James. Thank you so much.

James S White (41:43.03)
Yeah, thank you. Thanks for having me.

Virginia Heyburn (41:45.536)
We appreciate you spending time with us today on FinTech Unleashed. You know, what I love so much about this conversation today is that it reframes AI from being a shiny new technology to a disciplined strategic capability. I thought James laid that out really nicely. And for banks and credit unions, success with AI is not gonna come from chasing tools. It's going to come from aligning the right partners, governance, data, and leadership mindset to make artificial intelligence work responsibly.

and effectively. So again, a huge thank you to you, James, for sharing your AI expertise and to all of you for listening into this episode of FinTech Unleashed. If you enjoyed the episode, follow the show wherever you listen to podcasts and share it with a colleague who's thinking about where AI fits in their institution's future. And as always, look out for the next episode of FinTech Unleashed by following EngageFI on LinkedIn. Thanks for listening. We'll see you next time.