In this episode of Banking on Community, hosts Saxon Prater and Tara Schultz sit down with Kate Drew (Head of Research at CCG Catalyst) to dive into the real‑world impact of AI on community banks. Together, they explore the realities and possibilities, sharing practical ways institutions can boost internal efficiency today and lay the groundwork for future innovations, all while staying strategic and risk‑aware.
Kate serves up clear examples of how even small institutions are already tapping into AI, unpacks challenges like hallucination risk, and spotlights success stories that prove you don’t need a billion-dollar budget to get started.
Whether you’re experimenting with AI or just wondering where to begin, this episode delivers approachable insights and a healthy dose of encouragement for community FIs looking to stay ahead in a fast‑moving digital landscape.
Transcript
Saxon Prater (SP): Welcome to Banking on Community, the podcast where we share stories, insight, and solutions shaping the future of local finance. I’m Saxon Prater, PR and communications manager here at CSI.
Tara Schultz (TS): And I’m Tara Schultz, SVP of Strategic Insights and Industry Relations at CSI. I am very excited about our guest today, someone who I have the pleasure of picking her brain often and working with her on a variety of topics. Kate Drew, who leads the research wing of CCGA boutique consulting firm serving the financial services sector. Kate, welcome to Banking on Community.
Kate Drew (KD): Thanks so much for having me. I’m excited to be here.
TS: Of course. Why don’t we start by you telling us a little bit more about you and your background and your focus at CCG, and for bonus points, toss in what you love to do for fun. Besides, of course, writing with em dashes and deep financial services research, because those don’t count.
KD: Uh, well, like you said, uh, I, I lead research at CCG Catalyst. We serve primarily community and regional banks, and we help our clients with everything from business and technology strategy through to the implementation of those strategies. And my job is really to stay on top of everything that’s happening in banking, which is both fun and not easy, uh, so that I can communicate that out to the market and to our clients, um, and help kind of keep everybody on, on top of what’s going on. Um, we work with a lot of community banks, like I mentioned, so I’m really excited to be here with you today and I’m looking forward to our conversation.
Uh, for fun, I, I love to travel and I love to eat, and my husband and I actually took a year to travel before we got married. We were on every continent except Antarctica, and we were gone for an entire year. Um, and I, I have a 13 month old now, and everybody told me that when he was born, you know, good thing you did that when you could and all of that, but we’re trying to do it with him. We just got back from Paris, so fingers crossed, you know, he’ll be a little travel buddy too.
TS: I love that. So much fun. So much fun.
So we were talking about the next topics for our discussion here on Banking on Community. And I think it’ll be no surprise to our audience. So all of our conversations and, and the data on what is absolutely top of mind for our listeners and they’re seeking more information on, and that’s, that’s ai. Um, when you look back at our CSI Banking Priorities Survey, 83% we’re concerned about the potential of AI in banking. And 43% of respondents listed AI as one of the top three technology investment priorities or really priorities for the year.
Um, obviously we’ve seen a multitude of big news on this front over just the last 24 hours, and that’s kind of becoming a norm. So we’ve seen Anthropic releasing Claude for financial services and so many other things. So who better to pull in, uh, than someone I personally know has and, and is really deep researching this space. And you’re guiding financial institutions in this space today around AI, around the hype, the real, the oops, the downsides and the future. So Saxon, what do you say we dive in?
SP: Yeah. Uh, I’m all for it. And Kate, again, welcome to the show.
Um, we talk a lot at kind of a very high level, sort of more sort of a speculative angle on ai. It’s something that we’ve, Tara and I have actually talked about quite a bit. But I’m really excited for you to be able to kind of, you, you have dug into a lot of this so you can give us a little bit more, more detail. So, uh, I’ll, I’ll kind of just start there on some of the research that you’ve done. We hear a lot about AI in, in the media about probably too much, honestly. Has anything surprised you as you’re sort of digging into actual data?
KD: I don’t know if I would say it, it challenged my previous assumptions or, or truly surprised me. But one of the things that I have realized in exploring this space is that it is different from every other area that I cover when it comes to innovation and financial services. And that’s because as consultants or technology providers or industry observers, you know, we are often focused on areas of innovation that are specific to banking. So whether that’s banking as a service or open banking, um, or fraud, even, you know, developments in the fraud space. These are, these are specific to banks, right? But what I found with AI that’s really important is that this technology or portfolio of technologies represents an industry-agnostic leap forward.
So what does that, what does that mean? It means that we have to look at it really differently than we do all of these other areas of innovation, because it’s not happening just within our industry. And I feel like that was really important for me in, in kind of as I’ve been learning more because, because I mean truly like we’re all learning, right? Like even those, those of us that have built expertise in AI over the last few years, we’re still learning even, you know, those who are building Claude and Anthropic, those who are building OpenAI, these people are still learning too, right? Because this is very, very, very early. And I think one of the things that, that I have really come to realize is that we have to think about, put this in the context of the wider world. And once you do that, you realize that we are in the prologue. This is very, very early.
So the use cases that you’re gonna see might be a little bit boring in the beginning. It might be, you know, your employees using Microsoft Copilot. That’s okay, that’s okay because we are all at the beginning of this journey, not in banking, but as humans, right? You know, no matter how that sounds. That’s, that’s the truth of it, right? We are, we are at the beginning of this journey as humans. So whether you’re in banking or not, you really have to get comfortable with it on a personal level too, no matter what industry you work in. And so I think that that’s really the thing that kind of hit me the hardest because it’s just so different from all of the other areas that I talk about and all of the other areas that I’ve researched.
And, and I’ll give you another example of that. When we recently published a report on generative AI in banking, and normally when I’m doing research for my reports and I’m talking to technology providers in the space, so those who, you know, have that expertise to be, you know, building the future, they are not only wildly optimistic, but they believe that they can do anything. I can do this, I can do this, I can do this. And you really have to dig down to figure out, you know, where we are. Exactly. The technology providers I interviewed for this report told me over and over again, we’re not there yet. We’re not there yet. We’re not there yet. And that is very different. It’s very, very different from normally, normally what I hear, which tells me that, that this entire movement is very different from anything that we’ve experienced in the past.
TS: Yeah. And I think you make a really important point about it being truly industry agnostic. It’s, it’s simply about efficiency and effectiveness across the entire enterprise, no matter what type of enterprise that is. Um, but the transformation opportunity is truly massive.
And, and what I love about this, and it’s been a bit surprising because you don’t see this all the time, is some of the smallest financial institutions making big headway here. And I think it’s really all about baby steps. Doesn’t matter if they’re 200 million or 10 billion, it’s where they are willing to really start out with those small use cases, like you said, boring, but really important procedure development, you know, streamlining and speeding up those processes. They get a win, they celebrate that, and they move on to the next most pressing or repeated process within their, their banker credit union. Um, so it, it really, that’s a really, really important point about this being surprising impact industry wide. But so, so far beyond that, but even the smallest players are jumping in and tackling the biggest problems that they have with this. So it’s great to see. Let’s dive into use cases and some of the use cases, some of the impact that, that you’re seeing in the industry.
KD: Well, it’s funny, one of the things that you just mentioned was procedure development. Um, and one of the banks I talked to for our recent report, Citizens First Bank is doing exactly that. And I, I love this example for a few reasons.
So the bank began using generative AI to write procedures. It took a process that used to take one to two hours down to 10 to 20 minutes. It’s very straightforward. It was a very targeted decision to use this technology for this purpose. Um, and like you said, it’s small, right? But it has a big impact and it made a really big impact for this particular team and they did really well with it.
And I think the reason that this is such a good example is because they didn’t stop there. They used the learnings from that use case. First of all, they were applying the technology to a problem that they had as opposed to looking for problems that technology could solve, which is always what you wanna do. And then they used the learnings and successes from that instance to then go on and expand to different use cases. So now they’re using it to summarize reports, for example. And they’re starting to put together an AI steering committee so that they can explore other ways to use the technology.
I mean, this is really the way to do it, right? Is you start small with a real problem, you get comfortable, you get your employees comfortable, and then you also get your employees trained. And then those employees can kind of act as ambassadors to go, especially in a small institution, to go through the bank and help train everybody else on how to use the technology and, and, you know, act almost as champions, right? For the benefits and, and you know, take some of the fear away. Because whenever we talk about AI, there’s always going to be excitement and fear at the same time, especially if you start talking about automating work. Um, so really like, it’s not just about starting small, but it’s about starting small and then allowing that to proliferate in really smart and strategic ways.
Um, so that’s, you know, one great use case that I think really ties into a, a wider use case, which is, which is content development, right? If we’re talking about generative AI specifically, that’s what it’s really good at. It’s really good at understanding human language and then producing content out of human language. So think marketing, development of procedures, things like that. Anything that you can kind of, you know, would take a long time to create, it’s gonna be very useful there. You’re always going to need to have a human set of eyes on it, right? Because it’s, it’s not ready for prime time.
Um, but there’s, we’re seeing a lot happening in the content development space also, and I would put this under the content development umbrella is training. There are a number of platforms out there that are using generative AI to help financial institutions consolidate their training tools, consolidate all of that material, and make it a lot easier to train up new employees.
Uh, another area, and this is tied closely to content development, is knowledge management. So we’re seeing, I mentioned Copilot earlier, that’s sort of like a basic use case, right? Like you can just train up all of your employees on Copilot and you’ve gotten started with AI. That’s great. You’re doing well. Um, but taking that a step further, we’re seeing some institutions begin to create these sort of supercharged intranets where they build their own sort of customized Copilots that are fed the bank’s data and then become, you know, very, very trained up. The, the, the chatbots essentially are trained up on the bank’s data so that they can talk to employees and surface information specific to the bank and help employees move a lot faster. Um, we call it, I usually call it a supercharged intranet, because that’s essentially what it is. It’s a Copilot that is specific to you.
Um, and then the third sort of big bucket that I would put this in that we’re, that I would say we’re not quite there yet, but we’re gonna get there, is is customer service. And anything that, you know, kind of touches customer interactions and that’s where everybody wants to go, right? Because other industries are going there really fast. But for now, if you’re gonna have any kind of external facing chatbot or any external interaction with users, usually there’s gonna be a human in the loop.
But there’s a ton of potential there too. And I’ve seen a number of, of community banks starting really small with that, where they are sort of using traditional AI tools. And we can talk about this a little bit too, if you want, sort of how to break down the different subsets of AI, but they’re using traditional AI tools to create chatbots that will eventually lay the foundation later for when generative AI is ready.
SP: So you mentioned a supercharged intranet, and I bet that also indirectly helps with customer service too, because you could, you could scan through a lot of your documentation and information a lot faster. Do these institutions have specific guardrails around what the, the generative AI is accessing so that it’s just within their network and sort of existing ecosystem?
KD: Yeah, that’s really important. There are a couple of different ways that you can do this. One way is to essentially create, uh, a layer that prevents the original model—so say it’s GPT 4—from being trained on any of the bank’s data, right? So you’re essentially abstracting away the original model. Um, I think that’s probably the most typical way.
And then I’ve seen a couple of FinTech providers actually that are offering financial services specific, um, either knowledge management or essentially internal chatbot capabilities. Like they help you get up and running with that. And the way that they do it is by deploying individual instances for each bank, um, so that your, so that your data is secure. But obviously that’s a, that’s a major, major concern for, for anyone kind of getting involved in this space.
And you really have to make sure that you have a strategy in place before you jump in so that you know that you’re doing it the right way.
TS: I love that. So one, one additional point on, um, your Citizens First example too, and, and you referenced an AI committee. And I think what they’re doing really right there is they’re aligning the entire organization and prioritizing problems as such across so that, you know, not department by department, they’re trying to fight their own fires and most important problems, but they’re working as a team, they’re working as aligned with limited resources to really prioritize those and not making it an IT problem, but really centering it around the business problems they need to focus on.
So just wanted to, to, um, double click on that a bit and um, and ask you next, you know, we’ve talked a little bit about this in the past. Um, obviously some of this space is real and some of it is still overrated and overhyped. So, from a research perspective, how are you differentiating the real versus the hype?
KD: I think about this a lot, and I would say, I’m not sure real versus hype is how I would lay it out. I think it’s more, I think of, of us as more on, on a, on a timeline that we’re progressing through. And I think where we kind of get stuck in the hype is when you see people on LinkedIn or companies talking about us being a lot further along on that timeline than we are. So I don’t think it means that we’re not gonna get there, but we’re, we’re not there yet.
And so the way that I think about this today is that if you are talking about any kind of traditional machine learning, um, or NLP, we’re there, right? Those, those technologies are established in banking. You can use them, you can gain efficiency through them. If we talk about sort of the next phase in generative AI, there’s a lot of opportunity there too. Um, you know, like we were talking about before, internally chatbots, content development, things like that: all real, all happening today. It’s when we start to get into those more customer facing use cases where you see people talking about using AI in loan underwriting or using forms of agentic AI to begin to make decisions for customers, that’s where we start to kind of bleed into this hype cycle of it’s very exciting and it will happen most likely, but we’re just not there yet.
And I think the reason for that is, is really the technology itself, and this is why I keep saying it’s so important to remember that this technology is industry agnostic. Hallucination rates for LLMs are between 1% and 30%. That’s a huge swing, right? So because of that, in highly regulated industries, putting that technology in in front of a customer is extremely risky and banks are risk averse. So today we’re finding ways to do that with a human in the loop, right? So what I would say for community bankers who are interested in exploring this kind of technology for customer service is look for opportunities, tools that keep a human in the loop and be very wary of anything that seems to promise the ability to, you know, deploy anything agentic without that component. And you, you’re probably not even going to be likely to see much of that, even if it’s talked about when you actually dig deeper. Because even technology providers know that we can’t be doing that right now.
TS: Yeah. It’s about augmenting not replacing.
And I think, you know, one more bit of guidance is to stay really focused on the economic impact that that tool is, is truly making for them. And, and have true KPIs around what they’re achieving with, whether they’re talking to a FinTech that’s leveraging AI to solve a problem, or whether they’re doing that with, with internal tools themselves. I think that’s key and probably something you discuss with financial institutions all the time.
KD: Yeah, and I mean, I think one of the things that I hear a lot is this question of what’s real and, and what’s not. You know, it’s, it’s not like I, I’m not surprised by that question. I get it all the time. I think it is really more of a question of, of where we are on this journey, right? And we’re just so early, right?
So I, you know, every time I open LinkedIn, I’m seeing infographics of AI agents in financial services, and it’s when I see those infographics that I’m thinking to myself, yeah, this is hype. Not because it’s not gonna happen, but because it’s so far down the path and, and again, things are accelerating quickly, you know, hopefully we will get there, but if you have, you know, an LLM hallucinating even above 5%, you’re gonna struggle, you know, as a financial institution to get comfortable putting that in front of customers.
So that, that’s where I would say, you know, the greatest hype is, is probably around talking about, you know, agentic and things that really, things that really combine machine learning with generative AI.
SP: And I think thinking about it like a timeline, that’s, that’s a great way to put it. Like I’m now kind of visualizing it in a sort of continuum moving forward. What, what do you think realistically next couple years as these are getting more sophisticated, that banks are gonna be able to do a little bit more easily that they’re not already?
KD: I think that one thing that we’re gonna see happen quite quickly and probably in the next year or so, is bank employees starting to become much more proficient with AI and, and much more comfortable with it. And once that happens, we’ll be able to do a lot more. And the reason for that is because if, either of you use AI in your work or in your daily life, one of the things I’m sure you know is that you have to validate everything that it’s giving you. Whether it’s, uh, you’re using a public LLM or whether it is one that’s, you know, built internally, you still have to validate everything that it’s giving you.
And so the reason that you, you have to start with smaller use cases and, and ones that you know, have good return, but still you have to start small is because you have to train everybody on the technology to understand how to validate what they’re getting. Once you are confident that your employees can do that effectively and, and really are comfortable doing that, then you can begin to use it for other things. You can begin to use it to assist agents in the call center, right? You can be able to use it to assist relationship matter managers in making recommendations with their clients. You can start to use it with customers. I mean, I wouldn’t unleash it with customers probably in the next year or so. I think that’s gonna take a lot longer, but you can start using it in some of these ways that not only increase efficiencies, but also augment experiences.
SP: Yeah, that makes sense.
TS: I like that.
SP: Me too.
TS: So what are the biggest obstacles that really banks are facing when adopting generative AI technologies that, that you’re seeing experiencing, discussing with financial institutions today?
KD: Education is a big one. I think understanding that AI is you’re, you’re gonna be afraid of anything that you don’t truly understand. I think the exciting, you know, examples of community banks that we are talking about using this technology, it’s because they approached it in a way that allowed them to get comfortable, right? And they did the work to educate themselves. That’s really, really important because you’re gonna be wary of it if you don’t understand it, if you don’t understand what the opportunities are, if you don’t understand how to measure real risk around it, if you don’t understand that, you can start small. I think that is, that is a huge obstacle.
Um, again, not just for bankers, but for everyone understanding it is, is important. And when I say that, I mean, and I kind of alluded to this earlier, AI isn’t one technology, it’s a portfolio of technology. So you have established AI tools like NLP machine learning, and then, then you have these newer tools like generative AI, right? And when we talk about agent AI, that’s essentially combining machine learning with generative AI. So we are, you know, experiencing a major leap forward in the field of AI, which is why it feels like everybody is talking about it, but some of these technologies have existed for a really long time.
It’s important to understand that not only from a risk perspective, but also because again, if you go back to your strategy and your problems, you don’t wanna be looking for ways to use AI. You want to be looking at your problems and your strategy and what you’re trying to do and what you’re trying to achieve, and making sure that you understand the tools out there in order to map the right tools to the right problem. So I think as long as you feel like you have a really good understanding, and your leadership team and your board have a really good understanding of, you know, what the, what the technologies that fall under this umbrella are and what they do, then when you, you know, are kind of working through those problems and solutioning, you’ll be able to, to map the, the right solution to the right problem. Similar to Citizens First Bank when they identified that they could use it for writing procedures.
TS: Yeah, I think you make a really good point about the explainability aspect. And I think as the use cases continue to evolve, I think the explainability of the very base use cases are key because you’re understanding the technology and how it’s deriving what it’s giving you. Um, but then when you look at the use cases expanding, uh, from a regulatory perspective, from a loan pricing or credit worthiness perspective, like that gets a lot riskier when you look at the explainability risks and the need to be able to explain that to a regulator.
Um, but I would also say, and I’m curious if you agree on this or disagree, but the data awareness aspect is, you know, is key in having that complete and holistic data picture, um, from a financial institution’s perspective, from the access and the strategy for that to all be foundation to provide the right insight and guidance and next best product recommendations. You know, I feel like that’s pretty foundational.
And are you hearing that, are you having the same conversations with financial institutions that have not yet really centered in on the data strategy itself prior to jumping into, you know, more of the harder use cases leveraging AI?
KD: Absolutely. I agree with you completely. I think that that is also a, a huge obstacle for a lot of institutions because you can get started with AI without doing that with some of these smaller use cases that we’ve been discussing. But as the, the whole space kind of evolves, and as the industry adopts the technology, you know, more deeply and, and everything becomes a little bit more integrated and you have more advanced use cases, you’re gonna need to get your data in order.
Um, and you know, we, we’ve been talking about this with clients for, for a long time, even before this, you know, that getting, having a good solid data foundation can enable you to do so many things because it provides you with insight. And that’s really what you need to differentiate, right? I say this all the time that, you know, all banks, especially community banks, they’re about 90% the same and 10% different. And that 10% is so important because it is your secret sauce and a lot of your secret sauce is probably in your data.
So whether you plan to use the most advanced AI tools out there or not, or whether or not they solve your problems or not, having a really good understanding of everything that’s happening within the bank and with your customers is going to help you.
And I would just, if I could just maybe follow up on your question about obstacles, there’s probably a couple of others that are worth discussing.
I mean, we talked about hallucination, that’s a big one. Um, that’s going to come down over time. Um, and it’s gonna make more external use cases, um, more achievable and accessible for financial institutions. But that is, you know, something that needs to happen in, in the AI industry, in the field of AI and not necessarily in the field of banking.
And I think probably the other thing that is, that is worth pointing out is just getting started. Like, it’s so easy to get analysis paralysis around anything. And I think that that can be a huge hurdle for, for any organization today thinking about this. So, you know, if we could go back to kind of what we were talking about earlier around Copilot, just get started. Because the truth is, is that if you don’t, chances are your employees are using AI tools, whether you know it or not, and you are not participating in how that’s happening, right?
So one of the first things that I usually tell banks that are asking me about this is, you know, get an AI policy in place that’s just basic, you know, 101, get it in place so you have it, you can always iterate on it, but make sure at a, at a base level you understand how your employees are using it, because then you’ll be able to participate in the benefits and you’ll be able to mitigate a lot of the risk.
SP: Can you give an example of what an AI policy might look like? Obviously it’s gonna depend upon the bank or, or credit union, what have you, but what sorts of things would they include?
KD: Yeah, I mean, honestly, like it really does depend. It, it can really start with just foundationally governing how employees are using AI tools at the bank. Also, it should govern how you expect vendors to use AI and how you manage third party risk around that as well. And then, you know, eventually kind of mapping out the rules of the road for your own AI use as, as a bank. Um, but as, as you kind of grow in your strategic vision for AI and your business strategy, that policy is gonna have to grow and evolve too.
SP: Makes sense.
Well, Kate, you’ve given us a lot to think about. You’ve, thank you for providing all of your insight and, and just expertise in this area. Um, I’ll give you kind of like a, a final word. Any other thing that you would like to say to community bankers, smaller financial institutions that are hoping to stay competitive? Um, would you have any final words, suggestions, advice, whatever it happens to be?
KD: I mean, I think my advice is, you know, and, and I’m, I’m gonna sound like a broken record now, but it’s, it’s get started. You know, there are, and look at the examples. You know, the, the report that that we published recently has, has a number of them in there. Look at the examples of other smaller institutions and what they’re doing because it will be inspiring to you. Um, their problems are gonna be different from your problems. Um, but just looking at sort of how they’re thinking about it and how they’re approaching it in really risk mitigated ways, I think could be usually, usually helpful.
And again, like making sure that your employees know how they should be using the technology and, you know, leveraging things like, like Copilot inside of Microsoft, like that’s gonna do so much to help the bank as a whole to, you know, be ready for whatever the next wave is, which is probably coming sooner than we even realize. I think, you know, that’s, that’s probably like my biggest piece of advice.
And then, you know, to Tara’s point earlier, once you have any kind of foundation in place, then you can start to really think about how you both at the same time tackle problems using the technology by department, but then bring the entire bank along with you for the journey. And I think thinking through how you connect those points as you get started and, and making sure that you have that in mind as you’re getting started, is gonna be really important for building foundations for the future. Because this future, and even though I keep saying, you know, we’re in the prologue, we’re we’re moving quickly, so the future is probably coming before we know it.
TS: I think that’s so well stated.
And, and looking at it from a small financial institution perspective, many may not have the foundational expertise or specialty on AI, but I think the example given earlier is proof that they didn’t either, but they got started, like Kate said. And I think that’s very, very important.
But, but many of us and many of the financial institutions have been leveraging foundational machine learning for, for years via partners. They may not know it, um, but like CSI and the fraud landscape and, and they’ll rely on, uh, partners like CSI to continue to accelerate the real time fraud detection with partnerships, with integrations with better protection.
Um, but that doesn’t mean that they can’t make solo progress because many FIs have similar priorities, but every FI has unique challenges in the way that they work, in the way that they operate, in the way that they staff, you know, their talent varies, their processes vary.
Um, so they can still make progress even outside of their partner, but it’s very fair to be asking the questions of your business partners, of how are you helping me? What are you thinking? What is your future? What does that mean for our financial institution from an economic impact standpoint? Because that, that definitely matters. And, and it’s a very, very valid question.
SP: And I would also add to bring it back to the very beginning of this conversation when Kate said that this is an industry agnostic transformation. I think that’s also, at least for me, a kind of validating thing where you can see that all these industries are figuring this out. And you can also, even though you have different rules, regulations, et cetera, you can also learn things from other, other industries, like as they are kind of figuring out how, how to apply these gen AI tools.
KD: It’s probably like one of the few areas that you can think of, right? When it comes to innovation in banking that is also impacting everyone’s personal life, right? Yeah. So if you think about it that way, it kind of helps ground the conversation in like, wow, like this really is bigger, bigger than just banking. So certainly we can’t stick our heads in the sand, but like maybe we could get really excited about it if we can get comfortable with the risk.
TS: Yeah. Kate, fun fact: So my 13-year-old boy, um, when he used to want to look up something or ask me a question, he’ll say, “search it up, search it up.” And now it is literally GPT is a verb around here. He says, “GPT it.” I’m like, “what did you just say?” And that’s, that’s his browser now. Like when he is looking for something or looking to learn about something on his phone, he pulls up ChatGPT versus Chrome or Google. So it’s just a, a complete dynamic of you’re shopping for something and you’re on GPT. It’s crazy to think, you know, at that age already. That is kind of their baseline and that will be—go ahead.
KD: That’s also like a good point too. Like for them this will be Google. So, if we think, you know, even further into the future about banking that generation, it’s gonna look, it’s gonna look very different. I mean, we talk all the time about, you know, banking Gen Z and, and, and banking younger generations, but if we think even further forward to the generations for which ChatGPT is native, you know, what’s, what’s that gonna look like? I don’t have the answer, but I think it’s an interesting research question.
SP: Yeah. It’s a whole new world: search engine optimization, marketing, all of that is changing and evolving and it’s, it’s exciting. It could be a little scary to your earlier point, that it’s both these things are kind of happening concurrently, but um, but as long as you keep the human in the loop, as long as you’re able to like weed out those hallucinations, I think think that’s, that’s key.
KD: Yeah. I mean that’s, that’s definitely key.
I think some of the, the larger fintech companies have already kind of learned that the hard way where they’ve like rushed to, to release AI-driven or generative AI-driven chatbots. And then there were, you know, um, Redditors essentially who were able to kind of get the chatbot to hallucinate and say things that it really shouldn’t. I mean, they were trying really hard, um, but still if you’re an actually regulated bank, that simply can’t happen, so…
TS: Well, Kate, this has been amazing. You are obviously deeply researching this space and staying on top of it, so you are always a blast to talk with about what’s happening here, what’s going to happen, what’s real, and, and what’s next. So really appreciate you being on with us. I look forward to seeing you soon in New York City and Saxon, I’ll pass it to you to wrap us up.
SP: Sure. As you said, that’s a wrap for this episode of Banking on Community. Another big thank you to Kate Drew from CCG Catalyst for joining us. And thank you all for tuning in.
If you have a topic you’d like us to explore in future episodes, let us know on LinkedIn. You can also learn more about who we are and what we do on csiweb.com. Hey, while you’re there, uh, check out our Content Hub. We post new food for thought on banking and banking technology all the time. And you can also find past episodes of this podcast and we’ll be back soon.
Until then, keep banking on community.