Hello, everyone. Welcome to the Interface dot ai webinar this month. And this month, we’ll have our distinguished guest, Credit Union of America. I’m very excited to introduce Austin here shortly. But to just sort of give you an idea of what we’ll be discussing today, we’re going to talk about empowering frontline teams with AI, the impact and insights from Credit Union of America and their employee AI strategy.

But before we get to that exciting part, I’ve got a little agenda to run through with you.

Today, what you’re gonna hear from us is a little bit about Interface dot ai. We’ll introduce ourselves, and we’ll show you a little bit about some of the products and services you’ve no doubt heard about. Then we’ll kick off to Austin for our case study for Credit Union of America. And then finally, we’ll go over the evolution of AI and online banking go to market AI operator demo.

And then we’ll end with q and a. So please stay till the end and do put questions into the chat. Alright. I’m gonna go ahead and kick us off.

I’m Amber McCollum. I’m the vice president of sales and client sales here at Interface dot a I. And with us is our distinguished guest, Austin Hammond, and he’s the employee employee engagement lead for Credit Union of America. But I think how I need to introduce him today is he is the youngest and most accomplished panelist we’ve had.

And then finally, last but not least, our CEO and founder, Shrinivas Njay, who goes by Shri. So with that, Shri, I think I’m gonna turn it over to you.

Thank you, Amber. You know? And, Austin, appreciate you joining.

And, you know, congratulation being the youngest and most accomplished, you know, creating a leader in the webinar with us today.

You know, I’m gonna start off real quick, introduce a little bit about Interface dot a I, what we do as a company, and show you a couple of demos of our products so that you are able to understand our product offering and follow along the conversation with Austin later part of their discussion here. So as a company, we’ve been around for eleven years. We’re the longest standing AI, you know, partner out there.

You know, we’ve been doing this before. AI was cool. Right? So we’re about two hundred people strong. We are a QSO. We serve over hundred financial institutions with three hundred and fifty plus AI deployments.

You know, we have been featured in Gartner as the best AI for credit unions and community banks, Kuna, Allied Solution. We recently won best banking AI award at American Banker.

Little bit about myself, I go by Sri. I’m the cofounder CEO.

As Amber just introduced me as well. I actually come from a credit union family myself.

Grew up in a family where my father started a credit union, scaled it up to a couple of billion, helped him through the technology strategy a little bit. But I pursued my professional career and got significant exposure to AI at Microsoft, being part of the Microsoft AI Research Lab.

You know, pretty much bought those personal and professional experiences together to create Interface dot ai. Bruce Kim, my cofounder, CTO, you know, he was a pretty successful entrepreneur before we even started Interface dot ai. He actually built the world’s first billing SaaS company. And he he scaled it up to several tens of millions of dollars in revenue and joined me to start Interface AI.

The rest is history. Right? So that’s the founding team, and we’ve come a long way. We’re two hundred people strong now.

And this is you know,

we probably are the craziest partner to dream most ambitiously in AI. Right? And, know, pretty much and we’re also executing pretty well on that. So if you see on the left side where we where we are today present, which is our AI is this one AI brain across all the channels.

Either you want to use it for your employees or you wanna use it for your members or potential members, you have one AI brain that you could lean on. Our AI has a ability to upsell cross sell drive business, not just efficiency.

Efficiency is given.

We also have what we call intelligent RPA where AI has ability to or autonomously, you know, complete task by navigating through browser based applications, you know, without a human intervention. And by end of this year, we are building an AI native online mobile banking that is going live with the First Credit Union, replacing a major online banking provider. So it’s fully AI native.

Super excited about that. But what in the future, where we are going is, you know, we are going to let our AI enable the members to add multiple financial institutions so they can bank across all of them through this one bank GPT like assistant, which will then drive proactive financial wellness tips and drive financial goals.

We are crazy enough to dream big, which is this bank GPT. If we execute in this direction, we feel that it can actually, you know, eventually promise a universal basic income.

Right? So that’s that’s the big goal we have as an organization. And, know, we’ve taken a very different approach. We are first principle thinkers.

You know, if you look at a typical option in the market you have, which is on the left side, typically, you see the AI that helps on the chat and voice.

But usually, about twenty, thirty percent of your member interactions happens on chat and voice, but another sixty, seventy percent happens on your online mobile banking, which is not really AI first at all. But our approach is with the AI native online mobile banking, we are able to make sure hundred percent of your self-service is AI native. Right? So that’s something unique to us, you know, something really excited. It the AI, the the bank GPT vision we have is not just for the members or cuss customers of the financial institutions. It’s also for the employee.

And when we bring it together, this is kind of how our holistic vision looks like for BankGPT, which is your system of records, like core banking, credit card processing system, knowledge systems.

All of those kind of has a a common AI brain that powers your members and employees, right, through a chat GPT like interface, which is multimodal. Right?

You know, pretty much every time you make your employee AI better, your member AI gets better and vice versa. Right? So that’s the power of that one AI brain.

You know, that’s kind of the vision we have. And this brain can always make sure the interactions are insightful and automated, revenue generating, personalized, and secure. Right? So we’re gonna see some of this in action today very shortly here.

So AI has many flavors to it. Right? So we all broadly call it AI, but there’s a lot of flavors to it. Right?

So this is kind of the spectrum. Right?

So and depending on who you’re talking to, there may be anywhere in this spectrum, but the the the common terminology is AI. Right?

So the this the left side of the spectrum is AI that is highly scripted.

You know, your members know it. You know it while you’re maintaining it as a financial institution. It’s painful. Right?

And then there’s a partial generative AI, which is kind of partial AI. And there is generative AI based AIs, which is which is good in answering questions from your website and things like that, but they’re not really good in completing task. That’s where agentic AI kind of stands out with the AIs that can autonomously complete task, autonomously understand knowledge and fetch knowledge and and things like that or use intelligent RPA kind of tools to automate task that even when you don’t have APIs. Right?

That’s agentic AI. Right? So it’s a a whole new category by itself, but often, you know, that’s also called AI. So so that’s kind of what we see across the spectrum.

Right? So and as a company, we offer many solutions, but this is kind of a good snapshot for you what we do.

You know, on the member side, we have AI that can take up inbound inbound coming calls to your contact center and can automate forty, fifty percent of the calls and and continue to grow from there.

And there’s also chat AI, which kinda goes on your digital assets, like website, online application, online mobile banking to make sure there is a really intelligent AI companion assisting through interactions in these digital assets.

Because today, navigating these digital assets is very cumbersome, and they’re also not insightful and driving you towards a business outcome.

AI can do that in a significant way.

If someone is looking for an auto loan and reading through a description or trying to apply for an auto loan application, our AI works as a companion to make that process easy, automate the entire application process and things like that. That’s chat AI.

Then we have employee AI, which is gonna be the big focus for this conversation today, where we’re gonna show you how employees in financial institution like frontline staff and branch and call center could use AI and augment their workflows and accomplish more task and improve collective productivity.

Right? So and then recently, we launched a brand new offering, smart collection.

Basically, it’s an AI that has ability to do outbound call, text, and email.

It’s fully agentic, multichannel outreach for improving early delinquency. And, you know, OLB GPT, is our AI that replaces online mobile banking, which is going live with the first credit union end of this year.

Other products have been adopted by Tensor Financial Institution. And then finally, the AI operator is basically an AI that, you know, removes the need of CCaaS solution and live chat systems by seamlessly embedding human into AI conversation. Right? So it’s the world is changing. Right? It’s definitely with this technology, and we are kind of in the forefront of that.

You know? So we across all of these products, we have lots and lots of integrations as well as fraud prevention. Right? So to make sure each of the AI products has very strong guardrails, they don’t hallucinate and make things up.

They’re purpose built for, you know, banking, which we call bank GPT platform, purpose built for credence and banks with this very strong guardrails. So you get the benefits and the upsides of AI without the downsides. Right? So which is very important to us.

So that’s kind of a quick overview of our solutions there.

You know, real quick, we we’re gonna jump into our employee AI demonstration here to really kind of understand how, you know, Austin and team are able to leverage such a technology. We are gonna show you a little bit more comprehensive demo of employee AI. Austin and team may not be using all of those modules yet.

So but at least you’ll get an idea how it is supporting their day to day operations and things like which will I’m I’m I’m excited to hear from him here shortly. But let’s let us give you a quick sneak peek of our employee AI here.

Our employee AI that you’re seeing here is designed to help all all functions within the financial institution.

Your frontline staff or in branch and contact center or your your employees having ITHR related questions or your lending team needing to complete some task back office automation, including your executives. So let us show you how our employee AI supports all of them, and then we’ll jump right into hearing from Austin right after the demo here. I’m gonna invite my colleague, John, to help me out with the demo.

Alright. So to get started with the employee AI, what we’re gonna do is look at this from the frontline staff perspective. So as in a frontline staff, I may be in the contact center. I’m gonna go ahead and look at calls that are being sent to me.

These could be calls that are being sent directly to me, through a a menu system or through a automated AI agent. As you can see, when this call came in, it was our the AI is listening in, already making recommendations, and pulling up articles for me. The issue on this is the custom the members having issues with their online banking. So it’s gone ahead and pulled up this document for me to be able to reference the procedures and policies.

But as I’m talking to the member, I realized that they’re having a problem with their password, and we need to reset their online banking password.

So I can go ahead and do a quick search for resetting the online banking password. It’s contextually gonna follow what’s been going on in this conversation and automatically go ahead and search the knowledge article that’s already been open, and it scrolls and highlights the exact procedure or policy that I need to follow in order to help this particular member. So I’ve gone ahead and helped that member, and I’m gonna move on to the next conversation.

In this next conversation, again, there’s a a suggested action that is we need to lock this card. So we can go ahead and, in a chat GPT like interface, go ahead and engage with AI to actually carry out tasks on our behalf. So rather than actually knowing, how to do something, I’m just gonna go ahead and tell AI what to do. And in this case, I’m gonna head go ahead and lock this card for this member because they’re having an issue with it.

You can see through our integrations with the systems across the entire platform being one AI brain, it’s gone ahead and locked that card for me. And then if I wanna do any further, conversations with this member, I’m actually rather than navigating through all the screens and doing all the work, letting AI do that while while I can build rapport with this member. And maybe I get to the point where they might wanna sign up or apply for a new credit card. So we can say apply for credit card.

And this is gonna, then allow me to help them sign up, again, leveraging the AI to actually go ahead and do the work for me. I through that rapport that I built with the member, I can suggest, you know, maybe you want a rewards and cash back card.

And so this is gonna go out and search the cards that we have available, and then I can suggest that card to the member and go ahead and help with that application in this conversation.

Going ahead to the next conversation, what we have here is a potential fraud alert that is being brought to my attention. It’s making suggestions about how to handle this particular situation, again, using my text and just telling AI to go out and do this on my behalf. I can actually just type in what needs to be done, and I don’t have to spend time, actually, you know, again, going through the screens trying to, navigate all the different systems. If you think about this from your new hires or when you’re onboarding individuals, the time it takes to learn all the different systems of where to go, you know, lock a card or where to go open a ticket, all of that is being taken out of the equation, and it’s just a a chat communication.

This is then AI is then going to follow the process and the procedures that you already have designed, and AI is really great at actually following those procedures. It’s not gonna miss steps. It’s going to follow the exact process that you define for it every time so you have that consistent output no matter what that process is that you’re triggering.

So, John, one one more thing I wanted to highlight, on on if you go back to the call real quick. So the, the right side having that green box, So, basically, it kind of shows you the moment an agent picks up the call.

The the AI has already gone ahead, did, analysis on the risk of the caller. It does, you know, caller ID forensics where we do, like, thirty different checks, including calling back the number to see if there is a is going a voice mail or a busy tone. And if you can see on the right side in the details span, John, if you can highlight on the authentication, it even shows you at the bottom on the authentication what all authentication has passed.

Registered phone numbers have been checked and and things like that. So it it automatically verifies for the fraudster and and make sure your staff is kinda focused on executing to resolve the issue rather than going through complicated process. This ensures not only the member experience is better, but consistent fraud, you know, the rules being executed to ensure the you know, you’re preventing the fraud. Right?

You know, you’re not even leaving human mistakes to kinda leading to more fraud. Sometimes the the some of the tests that AI is doing to detect fraud is also not even possible for a human to do. Right? Like, it’s some some parameters are just test, like, caller ID forensics.

You know, AI is best at it. Right? So, so that’s helpful to to prevent fraud.

Thanks, Trey.

The other area that this is really helpful is if you think about having a coach for your team that’s handling frontline conversations, a lot of times, they’ll have to rely on, colleagues or asking questions or just trying to figure out what to do, how to help a member. This tool is now listening in into conversations and understanding what the intents are, what members are talking about, and then actually making those recommendations. So here, you can see one of those recommendations of exactly what the, frontline staff should actually respond with. And this is highlighting that there’s upsell opportunities.

So in many of the care organizations, it it’s hard to switch from being wearing that care hat to putting a sales hat on and making those upsell cross sells. So by taking advantage of AI to navigate that for them and not have to switch that persona, they can then see those opportunities brought forward to them to just make an impact on that customer, that member, and helping them with their financial goals.

Yeah. And and I think what is fascinating here, for for me is kind of, taking a customer who called to inquire about a missing auto loan payment and not only handling that right away, you know, but turning those interactions into a an opportunity to improve their financial well-being or offer them some products and services that could get help them get there. So that’s a massive shift from, you know, just transactionally handling why they called versus, you know, taking care of that efficiently, but, you know, then using that interaction opportunity to help them further. Right?

And if you see the buttons below, which is a suggested actions, AI is even predicting the next step you as a a frontline staff would take. Those are all actions. You say payment schedule or account history, it’ll connect to the core, pull it up, the account history as well. Right?

So that’s so you don’t you’re often, our frontline staff has, like, three three screens, you know, twenty different applications open, constantly navigating between them, you know, but this helps reducing all of that to one screen, right, and having conversation with AI where AI does the rest behind the scene. Right? So that’s kind of very powerful and helps them to have more meaningful conversation to create more value.

Yeah. And to the earlier point that Sri was making in regards to risk and fraud, you can see in examples where there is detected risk of fraud, AI is gonna clearly highlight that and make that frontline staff aware that this could be a fraudster. It could be a situation that, you need to be on high guard to ensure that nothing nefarious is taking place and taking advantage of all the information that’s being fed into the agent workspace.

And fraud is a big concern in the industry today. Right? So, you know, there is lots of different kinds of fraud. But, basically, people calling I mean, the one of the top kinds of fraud is, you know, compromised credentials.

Right? Like, you know, usually, you know, social secre security number, member number, all of the things that typically used to verify members are all available in the Internet for pretty much everyone. So that’s that’s very difficult and causing a lot of increase of fraud even post pandemic and things like that. Here, we’re we’re using AI to evaluate parameters that are beyond those basic things.

Right? So it could be even it can even trigger a a face ID, fingerprint ID, voice voice biometrics, or it could be, like, you know, caller ID forensics. So it’s using parameters that’s hard to, you know, break in, Right? And then ensure those verifications are helping prevent fraud in a significant way.

As you can see on the left side, John, scroll down. You can you can see the blue buttons. You can trigger voice device biometrics, which is like face ID, fingerprint ID, and make sure, they’re able to, use that and verify, themselves before you execute transaction.

Now transitioning out of the contact center and going back into somebody that may be working in the branch or branch manager, Really, this is applicable to everyone in the organization to be able to take advantage of and surface up knowledge and be able to get answers to questions, but it can even do much more than that. So as it may be a branch manager, I might wanna be looking at what’s going on, on my day to day. And so I’m coming in in the morning kinda using AI to help me come up with what do I need to accomplish this morning. And so it’s gonna go out and look up, you know, loan applications, wire transfers for me to be able to identify where I need to spend my time and and how can I resolve issues for members quickly and and get them on their way so that way, they can accomplish their financial tasks that they’re looking to accomplish?

As you can see, again, just using AI to chat through the process and and help navigate these particular, wire transfers and trying to identify where there’s problems, where there’s issues that I can resolve, I can just go ahead and have this conversation. And AI is gonna highlight all of these for me, and it’s gonna help me get through all of this without actually having to do a lot of the heavy list lifting. It’s leveraging AI to go in and accomplish tasks on my behalf, that that agentic experience.

The the last thing I’ll touch on just to show you one other example of that is, you know, in the contact center, you also might have leadership that wants to know what’s going on and what they can do to optimize the team and take advantage of how to how to better empower them, whether that’s through AI or or additional coaching. So being able to go and and analyze conversations, now that all the conversations are routing through AI, it can actually give you sentiment.

It can give you analysis. It can give you details, along with the analytics. But rather than, you know, have to go in and navigate analytics, why not just ask questions about the data? So that way, you can figure out what’s going on and where you the organization can be much more strategic in providing a better member experience or solving the challenges or issues that members are facing quicker. And and so being able to, again, create that conversation and and have that with AI as your your really your partner in tackling your day to day.

Yeah. I I love that. Like, you know, having a pulse on our members, right, is a big thing. Right?

Like, knowing what is causing frustration to our members. You know, we always do surveys once a quarter or once a year, but it’s very reactive. Right? So but AI lets you kind of be proactive about any member frustration and things like that, which kind of goes a long way in tackling that much faster.

Right? So, yeah, this is, pretty cool. So with that, we, like to, learn more from Austin, how he’s using employee AI on on his end and what plans he has to continue to, you know, you know, and, like, improve the workflows and things like that. I wanna hand it back to Amber and, Austin, for, learning more about CEO of America’s, employee AI journey.

Awesome. Thank you, Shri. This is my favorite part of our webinars always is to feature some of our amazing customers and their case studies as they use our products and services. So before I introduce you, Austin Hammond, I want to talk a little bit about Credit Union of America.

Credit Union of America was founded in nineteen thirty five in Wichita, Kansas, and we hear it’s not terrible weather in Wichita today, but it could be a little bit warmer.

One of the things I thought was super interesting to learn about the Credit Union is that it was born in the Great Depression. And I thought that was really interesting because it really feeds into your mission about stability and community centered financial support, and that’s a legacy that you have even today. And I’m pretty excited and proud to say humbly that we share that with you as well.

Today, you have over ninety thousand members, and you’re one point five billion plus in assets.

Welcome to friendly philosophy. It’s something you guys talk about all the time. And with that, I wanted to welcome you in a friendly way to our webinar today. So, Austin Hammond, thank you for coming.

Of course. Thank you for having me.

You bet. So the next part of our program is a little bit of q and a. I’ll be asking you, Austin, some questions about Credit Union of America and how you use the service. And then I’ll also be switching back over to Sherry and asking him some of those same questions as well. So let’s get started. Austin, how is Credit Union of America’s business strategy delivered more effectively with AI in general?

Great question. So I’ll start with the in general piece. Really, we’re focused on I’ll share a little bit about my role, I suppose. So my title is employee engagement lead, and what that effectively means is just making sure that our employees are, living their best work lives and are engaged as possible with, each other, their work, and their members ultimately because that’s what we’re focused on.

You know, if there’s a lot of other credit union folk in here, which there’s union folk hosting the call, that’s super important to us as member member experience. And, you know, we all view employees and members kind of in the same wavelength, and that’s what we’re after. So I we try to take a really humanistic approach to AI. So we’re not just looking at AI tech to enhance and improve efficiencies.

Right? Cut out, that one on one time that people crave. We’re looking at it to just enhance the workplace and make people feel more empowered to help their members, help each other, just really ease those pains. So, again, really just an emphasis on what can it do for us as opposed to the fears, you know, that constantly come up as, like, what is AI gonna take away from us?

So that’s that’s our approach there.

I love that. So yep. There you go. Anything else you wanna talk about kind of on, any of these, little Post it notes?

Yeah.

So I wanna start with that middle one because I think that glad you did.

Yeah. I think that would resonate with most people.

As far as, like, how it’s impacted our, strategy, employee retention engagement is definitely a huge one.

Sri actually came to us directly, which I don’t wanna set him up to, be doing that for every single rollout he does. But this was he came to us shortly post COVID. I I believe, it was, like, twenty twenty two when he first visited us. And, you know, those times, it was the great resignation. Right? It was tough to keep people on board, keep people engaged.

So that was something we were really trying to improve, and we saw, you know, we saw employee empowerment as the main way to do that. And we wanted to empower people by giving them information.

So, you know, some things we’ve seen, our member support department specifically, we’ve is historically and in the industry has huge turnover rates. Right? We’ve during COVID, I think we started around, gosh, like, eighty to ninety percent, and we’ve dropped it all the way down into the twenties, at this point. And, you know, I can’t attest that all to the AI and the systems we’ve implemented, of course, but I know they’ve played an impact. I’m grateful to have a really progressive leadership who’s looking at those technologies and, is also just really focused on the employees. So we make sure they’re happy, and the leadership team in that department specifically is also excellent.

Another big piece is, yeah, knowledge management.

And I’ll talk a little bit more about this later, I believe, but documentation is scattered, at the credit union.

You know, so much information was coming through email.

We have an Internet that we keep updated that I try to manage, but it’s, you know, it’s not everybody it’s not very easy to access that information compared to, like, ChatGPT like software where you’re going in and asking the questions, you know, what you’re looking or it helps you find what you’re looking for easier. So getting our documents into that system and making it more available, has really gone a long way to just making it more visible, making sure we keep it up to date more frequently because just more people have access to it. And then training and onboarding, that’s been a huge part of implementation.

Our training team, I work closely with our training. She actually normally offices right next to me, and we work, hand in hand quite often. And Frontline Assistant is something I gave them immediate access to because, ultimately, they’re the kings and queens of training documentation and what that looks like and how it and how it’s shaped. So if they make changes, I wanna make sure they can make the changes in Frontline Assistant and have that readily available for people.

And it saves them so much time because if in the past, when you release a newly onboarded employee on the job, typically, they’re the people fielding the calls like, hey. I forgot how to do this. How do I do that? Instead, they can say, hey.

Did you try Frontline Assistant? And, we’re introducing that during onboarding, which, again, I’ll get to later, but saves it prevents a lot of those calls ahead of time by just them having that resource available.

This is awesome. Thank you so much for that. Shree, my next question is gonna be for you. Alright. Shree, what are the CEOs that you’ve been talking to lately saying about using AI in their business strategy?

I I I think it’s we’re pretty much in line with how Austin and CEO of America is kind of thinking about this. Right? So it’s a great tool for, you know, employee retention and right? So because being in frontline is the hardest job on the planet.

They need to know everything, and they put on this they’re on the spot so many times in a day. You know, a member walking in, they expect them to know everything or a member calling in. Right? So it’s the hardest job in the credit union.

You know, I think having an AI that arguments and provides that the job a little easier, you know, provides instant access to that information is extremely valuable.

You know, there’s just so much organizational information you need to know to, you know, make sure the member experience is great and consistent. Like, you know, Forrester data analysis about a couple of years back talking to about five thousand credit union community banking customers asking them, hey, you know, on different questions on the member experience and the, you know, all all different aspects.

And about thirty six percent of the five thousand members who were interviewed said they often get incorrect information from the staff or the financial institution, the bank card. Right? So and that leads them to a different direction and also so it’s very and an important problem to solve to ensure the member experience is really world class. Right? So and, you know, having these kind of tools at the fingertip, they can get instant access to information goes a long way.

I think this employee AI is is going to become extremely important. And and, you know, the this will be like the days where when we got this iPhone in two thousand nine or something. Right? Like and we were like, we’re so used to using touch touch screens now.

And if we if someone gives us a button screen a button phone, we’re like, how do I use this? Right? Right? So, I mean, at least one thing we would say is how the hell we use this.

Right? Like, so it you know, because you can’t zoom into a picture, you can’t see it. So it’s like that. Once you get used to employee AI and, like, the AI augmenting your day to day workflow, you’d really can’t can’t survive without it.

Right? Like, because it’s such a a great addition to the toolbox.

You know, one of the CEOs I was talking to recently, they’re kind of talking about how this is now organization wide that they’re even having their employees ask IT question, HR questions too. They’ve been very creative kind of roll it out. And and the one of the unique use cases that there’s a a top request in our feature board right now from our CEO is they want to use this to detect for check frauds.

Right? So they wanna hook it up to a a check check scanning machine, and they want AI to kind of tell them what checks are fraud. Like, there’s just so many things possible with employee AI. Right? So, you know, it’s it’s it’s it’s gonna be exciting. We’re just getting started, the technology.

Love it. Alright. I actually wanted to pause for minute and talk a little bit. Austin, you’re using a couple of different things with us. Maybe we could go to the next slide, we’ll talk a little bit about what those are.

So do you wanna say a little bit about this, Austin, or shall I?

Yeah. I can I can, tap in here? So Okay.

Yeah. So we, like I said, contacted Sree. He came to us, visited us personally, and we initially got set up with Voice AI. That was a bigger priority because, you know, we wanted to takes the load off of our member support department as quickly as possible because people during COVID, I think, realized that they needed to call in. And afterwards, when it was kind of fading away, they realized they probably just could be calling in instead for most instead of the same range. So people just kept doing that, and it just you know, it stayed busy. So we wanted to roll that out ASAP.

That and the employee AI software that we got set up with, which I’ve been calling frontline assistant, you’ll see that here, is was still kind of in development when we were talking about voice AI and getting that set up, which is huge for us because we were able to kind of be a part of that development in a sense. But, yeah, it rolled out shortly after, which and I’ll talk about that rollout process a little bit later.

Awesome. Well, yes, thank you for that. So it actually brings me to my next question. So, just a little bit deeper dive into the problems you guys were trying to solve when you adopted employee AI or frontline assistant. They’re kind of interchangeable.

Yes. Yeah. Sorry. No.

You’re fine.

I employee AI and frontline assistant I say frontline assistant a lot because as I’m also advertising it within the credit union. So I wanna make sure I’m on brand in a sense and making sure people know about it. But, anyways, so, yeah, first off, scattered inconsistent documentation. That was a big thing, and I’ve already kind of hinted at that.

But we did have an Internet, but we we rolled out Frontline Assistant before we even used SharePoint, for instance. Our Internet was through a provider, and a lot of our documentation was just stored on network drives. So, again, sometimes it was just hard to find what you’re looking for with that because it’s much easier to get what you need when you’re able to have a conversation as opposed to typing in keywords in a search bar. Right?

So that was a huge thing, and having frontline assistant have that documentation in one place was a boon for us. Staff retention, I’ve already really kinda talked about that, but that was huge. And, I like I said earlier, statistically, proven or at least has been a part of that equation that it has helped a lot, because, again, it was at an all time high post COVID, and I don’t think we’re alone with that. Agent empowerment and morale.

Agents, you know, they they were we really care about our members like every credit union does. It’s but it’s defeating, right, when you wanna help somebody, but you can’t find the answer, and that kinda goes back to staff retention. So just giving them those tools and, you know, feeling like they have the resources available and they can actually help the people on the phone with them when they’re asking for help is is huge. Right? We they don’t even like putting off the call and sending it to somebody else. They’d rather do it themselves. So that’s what we’re trying to do here.

And, ultimately, it helps with growth too because they may be able to take the time to learn more about other roles in the credit union and be able to expand their knowledge there.

The other things that cause chaos with our documentation is just change communication via email, which is something that still happens and I still have to kinda manage, you know, making sure I keep a close eye out on emails sent to the whole credit union for sales, promotions, things like that that I’m just trying to make sure are also in Frontline Assistant. That way, our staff is able to reference it later, because, you know, they’re working in their email inbox. They aren’t they don’t have a special place, or maybe they do. Some of them, I’m sure, do have special places where they store these post its and flyers, but, I want them to be able to focus on their work and just be able to access the information when they need it.

I love that. And your point on the agent empowerment morale, I think that really, really resonated with me.

You started even our conversation talking about how AI is actually powering that human interaction. So I really appreciate you calling that out for your call center folks. It’s amazing.

Alright, Shree. What’s your if I use the word hot take, I think you just can’t tell my kids I used it because it’s cringey, but what is your take on that? What do you see?

Yeah. I I think a couple of things I I would like to add to that is often I’m pretty sure Austin and team faced this challenge as well. It’s often there is this question about how good AI is when our data isn’t that great. Right? Like, you know, our documents isn’t that great. Right? So, you know, that is a common question we get asked and the common challenge industry kind of thinks about before putting employee AI, you know, in place.

So the you know, because the scatter when we have scattered documents from different places and all, sometimes outdated information and things like that, you know, some sometimes, AI is as good as the data you feed it. Alright? So there’s a cleanup process and things like that that goes into it.

But in in our earlier version of the product, we built out a capability where, you know, people have ability to you know, employees using can ability to give feedback, and the AI can come back out in the dashboard, tell you what is going well, what what what is kind of feedback, and you can go fix it and things like that. So you’re taking a little bit of a reactive approach. Right? So after the launch, what knowledge is missing, what knowledge is broken.

But, like, you know, the demand in the industry has been more proactive approach, which is kind of where we are going next as well.

In in the future, Austin, you will be able to connect these kind of the, you know, sources of information, which is, like, literally next quarter. And when you click a button, it’ll show you all the conflicting knowledge so you can quickly resolve it and clean it up before you push to AI. So AI itself is gonna clean it up. So that’s kind of one of the biggest ask and that that’s coming.

So, hopefully, that makes employer adoption even better and easier for the teams.

There’s nothing more fun than delegating tasks to an AI agent, I have to say.

Alright, Austin. This is an important question for us that we would love for you to share with the group today. What made you choose Interface dot ai as your partner?

Yeah. No. It’s a great question. Early on, when we were shopping for solutions, for this, Sri was very quick to hop on and make the personal connection and fly out and visit with us and show us the solution.

Again, I don’t want him I don’t wanna be upset with me and French.

For every credit union.

I’m putting delta points in his account right now.

But, this was, you know, this was a long time ago. Things I’m sure have changed and evolved, and they’ve grown. But that being said, it was a very tailored approach for us. And beyond that, after implementation, we were able to really help influence the direction of, the solution we felt.

Interface was excellent about receiving our feedback and iterating on it, and just really showing, like, hey. That’s that’s a great idea. Here’s we’ll take it back to the team. They revisit with us, and then it’s exactly what we asked for.

So it felt like, we were kind of partners in that sense. You know? We’re not just waiting for the update to come out and crossing our fingers or crossing our fingers that they take our feedback. A lot of what we shared was acted on, and, yeah, it was it was a fantastic experience.

Oh, we love hearing that, and that is our roots. And, it’s something that we’re holding true to even today and making a real effort to continue to hold true for that. So thank you for that. Shree, we’d love to hear from you. What do you see as the key reasons that credit unions are choosing us today?

Yeah. I I I think, know, continue continue we continue to collaborate with the financial institution. Right? And I think we’re coming up with new ways of collaboration, and we made a announcement to all of our customers last week how the collaboration is going to improve further using AI.

So but at the end of the day, creating that you know, the one of the reasons financial institutions like credit unions like c o CEO of America or others partner with us is they wanna be their design partners. Right? Like, they want to work with us as an extended team to solve the challenges to serve the members better. They’re on a great mission.

And and and, you know, they they they’re looking for partners like us to co design, co develop, and, you know, co grow together kind of approach.

Typically, there’s you know, the industry has a lot of options of big vendors and which is not that flexible to work with them, influence the road map, and things like that. But when when it comes to young startup like us, we take a different approaches with different culture. Right? Like, you know, you you’re obsessed about the customer and you go out of your way to work with them to co design, co develop.

Right? The the the intention is right. Right? So as as a matter of fact, last week, made an announcement where now going forward, we are gonna have a really tight collaboration with our customers.

So typically, what happens in a soft software development process when you’re, you know, working with your partners is you have something broken, you go create a ticket, and then it takes a while for product team to analyze and come back and build it. And, you know, it’s a long cycle. Right? So and and meanwhile, you had business problem problems that wasn’t solved.

Right? Like so but AI is changing that. And being an AI native company, we are thinking **** ** these problems. How do we collaborate more faster?

So we we introducing a new way of collaboration with our customers, what we call, you know, agentic development, which is our homegrown technology where, you know, very soon, Austin, if you were on the product town hall, we announced about this. But very soon, we are gonna be, in your Microsoft Teams, as a channel where, our AI and your team and our engineers will be in a a a channel together. So you can provide feedback and AI starts building it immediately. And then the engineer comes and make sure it is, you know, banking grade and then push us out so that we’re able to ensure we’re able to collaborate much faster.

The gone are the tickets. You have a idea. You know, you put it in your team’s channel and AI starts building it right away. Right?

And makes it available to in a shorter notice. So that, you know, that is something I’m very fascinated about, how we can collaborate with someone like Austin and team in the coming days in a more tighter way and more faster. Right? So I I think it all goes into the the the these kind of initiatives where we listen and kind of iterate and work hard to build better stuff.

Right? So, you know, grow together kind of mindset is the top reason, you know, people choose us even today.

I love that. I can’t help but harking back to the iPhone example you gave earlier, and I’m thinking of the pain of having to hit one number three times just to get to the letter I needed. So crazy now that we’ll be able to do that with AI just via Teams.

So, Austin, walk us through your implementation journey, please. We’d love to hear more about it.

Yeah. Absolutely. So, I was tasked with rolling out Frontline Assistant to our staff. And, pretty early on, the the first thing I did was reach out to pretty much every department individually and just say, hey.

If we get this type of product, this type of software, what would you like to see in it? What are some of the things that agents or our employees commonly reach out to you for, sometimes members as well, that maybe we could help answer so you could get fewer phone calls ultimately?

So we, you know, got some of those important pieces. We made sure the documentation was put together.

Sometimes, you know, you find out there was no documentation, and that was a bit of a a bit of a lift, but it was needed ultimately.

So you have to be prepared for that, and that’s that’s bound to happen. But once we actually got to the testing phase, did kind of an alpha approach where I brought in twenty to thirty people, and I had a lot of folks in our member support department or phone center, our people, our branches, just a few different departments here just to get a sample size, but primarily member facing individuals and ask them I created templates for note taking. So, you know, if you ask this question, what does it respond? How do you feel about that response? Does it give you a good document?

Just to make sure it was working for them, working for them properly. And, also, I wanted to hear about the questions that they were asking that they couldn’t get answered. So I was trying to fill those gaps as well, and all of it was necessary. Obviously, if there’s questions they’re asking that can’t get answered, then that goes just beyond rolling out a product like this. I mean, we have to make sure that documentation is there, but this made it easier than ever. As far as actually rolling it out, once we were happy after the alpha, then it was just kind of department by department. Of course, started with our member support team and our call center because they had the biggest need.

After that, we rolled it out to some other back office, departments who are eager to try it out. We have a department who handles all of our payment solutions. So ACH is wires. They often interact with members.

They wanted to try it out and then move it to our front operations, which is a much bigger space we have. That’s our biggest department by far is those on our branch operations. We rolled it out to them, and, yeah, that was essentially that. Once we got it rolled out to everybody, we enabled it.

So, it’s all single sign on. Nobody has to remember a password. And if you as soon as you’re employed, you’re ready to go.

As far as, like, adoption tactics and training onboarding, we make sure that day I think it’s not it’s not day one, literally, but it’s day two. I go in and introduce frontline assistant and, tell them what it can do for them. I share all the information they need to know about giving feedback on it. So if it can’t provide documentation or it gives them an answer that isn’t quite right, I, you know, give them exactly what they need to share those share that feedback with me.

But frontline assistant also is able to give me it’s called the next best questions feature. So, basically, it tells me what’s missing. So that’s already helpful. But sometimes you have people who have very unique questions that doesn’t get flagged there, and I just wanna make sure I get everything.

But, yeah, point is we start from day one, make sure everybody is interested in well, not interested in, but aware of it. You know, you you’re still gonna have folks who are hesitant of AI, but I really just wanna make sure they’re aware of it and, you know, really stress that sometimes when you’re new on in a role, like, we’re always here for you. You wanna you have people who are available to help you, but sometimes when you’re new, you wanna be able to find those solutions on your own and feel empowered, and that’s what this is here for. Like, there’s not much of a better feeling than, like, bring being pretty new at something and figuring it out on your own. Right? And this is part of that, and that’s what we really try to encourage.

Love that. As you hang up the call feeling like, nailed it. Had it. Yeah.

Yeah. I did it.

Right. I love that. So, Sri, you’ve done a few of these. We’ve done a few of these together. What’s your advice to credit unions on implementing AI projects?

You know, I I I think the approach sorry. I’m the approach Austin and team has taken is extremely thoughtful. Right? Like and I think it’s as much as, you know, they’re they’re they’re, like, having the impact and the value from these products.

And, you know, it’s it’s a it’s a a huge blessing to have partners like that. Right? Like, you know, who are having capable teams and who can execute such projects really well because technology, you know, is is is there. It works, but, like, know, it’s all in execution and adoption and such a thoughtful approach as are hearing so far what Austin’s sharing.

Again, you know, the the important takeaway in my mind is it’s it’s about a fundamental change in the mindset of implementing an AI project or a product versus a non AI product. Like, that’s that’s where the struggle often lies. Right? So, you know, in a if if you’re implementing, let’s say, online mobile banking, right, so you kind of do big heavy lifting during implementation, you know, ninety five percent of it.

And then afterwards, it’s really trying to make sure it’s working as you expect. Right?

But in a AI project, it’s a it’s a learning software. So it can keep growing, keep adding value. Those feedback loops becomes extremely crucial and being able to take those feedback loops and then execute on those. Right?

So because Austin hit the nail on the head. Like, if AI is not getting better and continues to answer more and more question every day, people are gonna adopt it. Right? It’s the same for your employees, same for your members.

Right? So they want a smart AI, right, that learns and grows. Right? So that that is the most important part of the struggle, which is the implementation never stops.

So in our mind, implementation of a software project is, oh, I set it up. I cannot forget about it. But in AI, it’s it’s an ongoing implementation.

And and, you know, it is effort, right, of course, but it’s also outsized value that comes out of it. Right? So, you know, every effort often is putting and his team is putting to add another one more document, you know, that could answer hundreds or thousands of times the same question to people.

There’s just so much efficiency and, you know, with that and right? So so that’s that’s kind of the major takeaways. Implementation never stops in AI. Right? So it’s a learning software. The change management continuous change management is really key. Making sure your project team project management teams think that way and plan that way is a key to success of any AI project.

That’s great advice.

Alright. Back to you, Austin. I have more questions for you. Kind of impact have you seen, and what’s the adoption?

Yeah. So first off, higher employee retention. That’s definitely been a big thing, especially for our call center impact as well. Again, people just feel empowered. Our we do internal engagement surveys, and those have been steadily rising over the past few years.

Again, especially after, like, COVID where there’s a natural dip in employee engagement just because times are tough. People are having a harder time just living their personal lives. Right? So they wanna be able to come to work and have less stress, and things have overall been better. We also set a record member satisfaction scores, just this past year, which was wonderful.

And, you know, getting things I can’t just say, like, oh, yeah. Employee AI, we added it and boom. It did that. There’s a there’s a million different pieces that go into it, but, it all ties into the strategy of just making sure our employees feel empowered, and the employee AI and frontline assistant has big impact on that. As far as, like, specifics, you know, agents on average are asking fifteen to twenty five questions a day.

We have really good engagement at the credit union. Last year, I started we were we’ve always been tracking how many questions get asked in a month, but I believe we started the year around, like, five hundred questions a month asked through AI and then ended the year with around nine hundred, and that was regularly for the past three or four months before that. So people are really engaging with it. And like Sri just mentioned, that’s the creates that feedback loop where as more people engage with it, it gets better.

And then word-of-mouth and more people will try it out, and then it gets continue gets better. And you can find those trickier questions that, those who were hesitant at first because they maybe have more experience, so they’re, a little less open to the idea of asking an AI question. They’ll ask those really complex questions that aren’t well documented so that I know what to fill those gaps with. So we’ve just improved our documentation overall and been able to just make our processes smoother as a result, of that.

Just, again, having visibility into what people are asking and making sure they are feeling more comfortable with that. And, yeah, that’s that’s that’s the kind of general impact that’s had on us.

I love your living, breathing strategy around this. It’s amazing.

Sri, I would love to ask you what you see as the impact of employee AI projects just across the industry.

I think, you know, it’s it’s it’s something we dug, but, like, know, we kind of use similar technology internally and everything too. And and and, you know, generally, the direction we see this is gonna make a huge difference is I know I I know there’s kind of crazy claims about, like, you know, small comp like, you know, fifty, hundred people companies could be a billion dollar company and things like that across across the industry. So I I think we are gonna be given a break from manual repetitive work as employees. Right?

So and then help think on a more strategic level and which could be about, you know, expressing their empathy to the customer on the the member on the call or taking a break and talking to to to them about their struggles of the day or how we can help. Because the the movement of credit union is people helping people. Right? Is is comes from, you know, the the the being able to form that connection and help in a deeper way.

Right? So it it’s not about helping them with balance inquiries. Right? You know, that’s what AI is taking away.

That giving you time back so you can live up to that people helping people philosophy truly. Right? So by, you know, that’s that’s that’s for me is the biggest impact. Right?

Like, you you get back to caring again. Right? It’s not a task you wanna complete. So let the AI do it.

Right? But let me take care of my member. Right? So that’s probably the biggest impact in helping us live live up to that people helping people philosophy a lot more.

Not your hustling and trying to find the document and answer what, you know, you don’t want member to get frustrated because they’re waiting too long on the call and, you know, those things are going away with AI. Right? Like, you can be peaceful, calm, AI is telling you what to do, and then you’re just caring enough about members. Right?

So that would take us to the kind of the that is very important with with this this generation and the era. Right? Like, you know, we living up to that philosophy more and more is important to continue to make sure the industry is growing. We’re all attracting more members and things like that.

Right?

So yeah.

Yep. Makes perfect sense. Okay. So, Austin, part of the reason the people are here is they wanna know about best practices from your point of view. So if you wouldn’t mind sharing, we’d love to hear it.

Yeah. Absolutely.

First off, you know, get executive and IT buy in.

I think it’s important to, of course, get their buy in if you’re going to get a product like this, which will provide a lot of value, but, obviously, it has its expenses. You know, it’s not free. You guys are running a business, but, it’s important to show them, you know, why this matters and really, again, take that humanistic approach to it, which is what a lot of our upper management did when we were approaching our executives about it. It’s like, we need our staff to feel more engaged. We want them to feel empowered, and that’s what we’re after here.

Obviously, IT security also needs to be involved. So start talking to them early on if you have an IT security department or if you have an IT department.

AI and the subject of that is on their minds constantly. I can guarantee it. So they’re not gonna be thrilled. They are knocking on the door to talk about AI, but it’s gonna be crucial if you’re gonna have success later on.

And Interface will do a fantastic job walking through any of those concerns that folks might have. Beyond that, once you actually get it, you know, like I said earlier, involve your frontline staff early on for testing. Make sure they’re engaged with it. Make sure they’re aware that this is happening because it’s for them.

Right? People people should know that you’re actively trying to improve their day to day work life by doing what Sri mentioned earlier and trying to reduce those those tedious tasks that don’t really build relationships. They’re just things that need to happen, you know, trying to cut those out.

So involve them. Don’t silo yourself. Reach out to departments right away.

The goal of a system like this is to unsilo. So don’t silo yourself and start working on this as a project that you do by yourself. Reach out to departments immediately and see, hey. What documentation would you like to see in here?

Guarantee you, you will find gaps in knowledge, and that is a part of the process. So, you know, be prepared to assist them with filling in those knowledge gaps. If it’s all there, wonderful. You may have to, work with them on organizing it so it fits seamlessly into the platform.

But you will likely have to help fill those gaps for your frontline staff who’s been involved in testing and those people who wanna try it later.

Again, rolling up by department by department is probably a good practice there. Just, make sure you have a small sample size. It’s functioning well for them. Get their feedback and improve it, just before you really move on to those big departments. Like, I really wanted to make sure this was as, working as well as possible for our before we moved it out to our branch land folks where, you know, we have over half of our workforce.

Because if it doesn’t land with them when it when it launches, then it might just be dead in the water. Right? And then you’ll have to kinda start from square one on, word-of-mouth. Like, hey.

Try this out. We’ve worked on it. It’s improved. You know, you wanna just make sure you have that documentation in there ready to go.

And then we also do engagement incentives. So this was something about I started about six or eight months in the platform. I wanted to continue to drive engagement.

And, you know, we had our base of users. It was steady. It was reliable. It was just growing a little bit, but I really wanted to ramp that up, because, again, it only gets better the more people that are using the platform. So we have a an internal rewards platform that basically allows you to get points that you can redeem for literal cash or gift cards, things like that. So, every month, I have a an award that goes out to the individual who asked the most questions.

It’s twenty dollars. We have an award that goes out to the person who provided feedback. So, like, if you when you ask a question in the system I don’t think we saw this, but when you ask a question, you can say how was it. It’ll ask you how was the question answer to, which if you’ve used any other generative AI software, you’re probably familiar with that.

It’ll ask how it responded, and you can give it a thumbs up or a thumbs down. If people give me detailed feedback on, you know, what was good or what was missing, then person who does that the most in a month will get twenty dollars. And then I have a random drawing, which I think is the biggest draw, is the biggest draw point for people. It’s just a lottery.

So if you log in and ask a question once during the month, then you’re in for that, and you can you can get easy money there. The idea being that those folks who are were hesitant to try it when we first launched it or just haven’t given it a shot yet, they’ll give it a shot, and then they’ll hopefully stick with it. That’s the idea there.

I love it.

Yeah. I I think that about covers it.

The followership internally is amazing. Love that strategy. Sri, what’s your take on best practices?

I think my my suggestion would be find your Austin.

That’s a great one.

Right. So within the credit credit union sorry. Yeah. You know, find someone who is passionate to drive something like this. Right? So there’s just so much thought went into that.

Even half of that effort would still be very impactful. Right? So I know Austin is kind of very thorough and very thoughtful in what they they did. And sometimes, you know, it’s hard for us to do all of it, but even, like, some key parts of those initiative, creating strong feedback loops is probably, I would say, the foundation. Make sure it keeps learning.

You know, keep looking at the dashboard, see what else is missing, and keep adding those. And and at at least make an announcement to the team, hey, you know, based on your conversations last one week, we AI just got better. Here are the questions it can answer now. Imagine Monday morning, they get that message every week.

You know, you will change everyone. Right? Like, you’ll change their perspective, and they’ll start using it for for sure. Right?

So I I think in in incentives and that lottery and all is a really fun thing to do. And and, you know, it translate backs back to how we are collaborating in in a new collaboration model with our customers like CEO of America, Austin, and everyone, which is, you know, it’s a similar philosophy where as we move to the AI driven development where they have Microsoft team channels, they can kind of literally provide feedback. And we wanna increase that feedback. So we created similar structure internally, which is rolling out to our customer soon, which is every customer provides feedback and cost and and the employees or the customer provides feedback.

We are creating the lottery system too. So to ensure now you have the full pipeline, you know, the credit union driving the adoption and feedback loop at the top end, but and then working with us as partners on on on the behind the scene to drive the product to get there. That completes the full cycle to be able to ensure the team’s collaborative working to deliver faster and faster. The best practice is, of course, Austin talked a lot about how employer option.

But from the partner engagement perspective, actively partnering with us to give those feedback just like Austin and team did. And they’re also already on a a pilot program for our next version of employee AI, which is very advanced and fully agentic. I’m I’m super excited about that to get their feedback and come back here in a few months to talk about, you know, how that made a difference. Right?

So, you know, you know, that the best practice and takeaway there is to collaborate with find a partner who is open to collaboration and then, you know, put in that effort to, make their collaboration with all the tools like AI. That collaboration could yield quite a bit of significant ROI. Right? So that’s kind of my recommendation there.

It’s very good advice. Alright, Austin. I think I just have one more for you. What’s your road map for employee AI? You’ve done such an amazing job of planning and laying it out for us so far. What’s next?

Yeah. So this year and recently, we’ve been moving those network drives I talked about earlier where we didn’t have anything on SharePoint. Our IT department in the background has been moving all of that to SharePoint. So my goal is to work with them and collaborate with other departments to centralize our policy and procedures and then put SharePoint in frontline assistant.

So integrate so that integration has been available for a while. But as a company, we were lagging behind on it because of our, love of network trust. So we now that we got that set up, I’m very excited about, really working and creating a a strong structure of documentation, and I’m excited to have frontline assistant help me with that, like Sri mentioned earlier, something that could, you know, tell me where things are being redundant. Right?

Like, documents are saying the same thing. So looking to do that.

Along with that, kinda goes hand in hand. I’m trying to get more departments who operate on the back end invest invested in frontline assistance. So, department I mentioned earlier, our payment solutions who focuses on wires, ACHs, they have a lot of sensitive internal procedures that, they don’t want everybody to have access to, and those kinds of things aren’t in frontline so now that frontline assistant has the capacity to have role, role based restrictions, which is, something that’s new, we’re working on implementing that and adding their sensitive procedures, making sure that only they have access to them and it works well for their department, because, again, it’s something that can help everybody. It’s just, making sure that they’re aware and that it works well for them.

And lastly, just really wanna grow adoption beyond, those departments and really try to get those subject med x subject matter experts on board and, the managers, the people who would never normally need to ask, policy procedure questions because they tend to be the people that get asked those questions.

I wanna get their questions in the platform because I wanna I wanna get those really complex things, you know, that, AI our system currently could not answer because the documentation doesn’t exist. So it’s really just about, documenting those complicated policy and procedures that are just sitting in the brain, and just exist because we’ve spread it through word-of-mouth. So really getting those really nuanced things down just, again, for the the backbone of our company just so it’s stronger in that sense and getting it all within, frontline assistance of the documentations available to everyone when suitable or, you know, available to only certain people when that makes sense. So those are the things I’m working on and things I’m excited for.

This is amazing, and thank you so much for sharing all that with us. You kind of dovetailed right into something that Sri is really fond of talking about, which is the crawl, walk, run approach for AI projects. Sri, do you wanna talk about that a bit?

Yeah.

That’s our favorite It is.

Approach.

Crawl, walk, and run.

Right? So because it could feel overwhelming. Even if you saw that demo, there’s just so much it can do. And you you kinda worry about, like, you know, where do I start?

What all what all needs to happen? Right? So I think taking one step at a time, like, you know, which is like what CEO of America and Austin team did, which is, like, starting with that, you know, policy and procedures. Right?

Like, with the AI is able to quickly help your frontline staff find those important answers instantly, accurately without making things up. Right?

So that’s, like, the first step.

You know, it from the IT perspective, that will be kind of a low hanging fruit. Right? So because usually, that is also a huge bottleneck for all of the credit unions. Like, you know, they want to do business teams wanna do it, but then IT doesn’t have bandwidth and things like that.

So at least starting with the crawl would be my my you know, recommendation crawl would be just start with policy and procedure. Right? Where, you you know, like, Austin and team did, they connect to the network drive. The AI could pick it up every front everything from there and start answering questions to your frontline staff or it could be SharePoint or any other system.

That that’s kinda low hanging fruit. And then my way of the walk would be, you know, I I think ensure there’s two things. Ensuring all departments are able to use it and and and then slowly kind of as part of the walk phase, you can expand it to transactions too, which is now instead of AI telling your staff how to do it, it can do it. Right?

So that kind of saves much more time and effort. Right? So that’s usually kind of the next phase and and things like that through integration to your core system, which is a little more IT in one. Right?

So I wouldn’t I wouldn’t even underestimate. I wouldn’t say little. It is IT needs to be involved for that. Well, to get to the walk phase.

Right? So core bank integration, LOS, credit card, debit card processing integration. So that even there, you can start with one system and kind of grow into other system.

Then the AI, instead of telling the staff how to do it, it’ll get it done for them through calling the APIs and things like that. Right? And the run phase would be now, okay, you kind of cleared off all the manual work for them. The AI is helping kind of take care of doing transactions when members asking or giving them instant access to information.

So you kind of freed up some time and kind of create some efficiency there so they could spend more with the members. And now the run phase could be, how can you leverage that time to have enable your staff to have meaningful conversation with your member? Right? Of course, like, connecting with them, making sure we’re understanding their challenges and and the how the day is going is great, but how do you translate that to a solution to them?

Right? Like, okay, if they say they’re having hard time, you know, paying credit card debt or, like, they’re having hard time to make next payment that is coming or they’re or it’s just a good situation where they’re, like, thinking about buying a car. They’re unsure how to do it or, you know, where to start, you know, can they afford it? So imagine now, the run phase could be, as you’re having that conversation, our, you know, product recommendation features and upsell, cross sell, financial insight features, which is kind of the run phase, which is what we call intelligent banking module.

So if you enable that, now as you as your staff is talking to your members, now it can start formulating solutions to their problems and put it in front of them so they could not only deeply connect with your members, but at at the end of the call, they can solve the problem. Right? So that’s, for me, is like a crawl, walk, and run kind of journey. Making that interaction from, hey, hold on a second and you’re doing something before AI.

Right? And and and to, you know, listening to your members and let AI do the repetitive task and then coming back and telling your members, here is where you can fall solve their financial goal or a financial problem. Right? So taking the your member ex member interaction with your staff from hold on a second, I’m looking it up to, here is how I can help you, you know, buy a car or pay off your credit card debt.

Right? Like, that’s real people helping people for me. Right? Like, meaning making that meaningful impact in members’ life every day is is kind of the run phase.

Right? So yeah. So, you know, start small.

There’s so much you can do.

Great advice. Well, Austin, thank you so much for joining us for the q and a today. We it’s been such a pleasure to have you. I’m gonna turn the rest of the presentation over to Shri.

Wonderful. There were a lot of questions, Austin.

You know, the team kept answering a few. There’s a few open.

Let me look at this. What makes interface employee facing interface better than using Microsoft three sixty five SharePoint agents for staff to gain access to information. Yeah. I think you you you’re talking about, like, even the copilot from Microsoft, like, you know, integrated with SharePoint, all of that. So it it but the the challenge is your inform your all of your documents are not only in your Microsoft products. Right? So then, you know, the the place where you get stuck with companies like Microsoft, large company, you know, amazing companies, but where you get stuck is they don’t purpose build this solution for our industry, which means, you know, they are not that accurate in responding to the questions as much as we do care because we are a regulated industry.

We ensure high degree of accuracy and guardrails. Right? So that’s number one. Number two is, you know, it it so if you look if you heard about the walk phase and the crawl crawl, walk, and run, we talked about how you can take it to the next level where instead of AI telling your staff how to do it, it can do it through API integration to your core or something like that.

That’s where kind of the limitation with something like Microsoft. They won’t go that far and or even coaching your agents how to, you know, upsell cross sell or provide the important financial insight to the members. It it they won’t go that far. Right?

It helps you with the crawl phase, you know, and then you’re stuck. Right? You’re not going to go far to do the walk and run phases we talked about because they’re not purpose built for our industry. Right?

So they lack all those integrations and all those model tuning and the trained AI system that can understand member offer suggestions and things like that. Right? I mean, Austin, you wanna add anything to that? Did you had ever kind of think have to think about, hey.

Can I use Microsoft Copilot or something versus this? How do you guys navigate that?

Definitely. I mean, once we’re a Microsoft company for sure.

And once they shoved Copilot into basically everything, within the past two years, it’s low hanging fruit to look at that. Right? We explored it. I think the main, you know, the main issue with it that we ran into is kind of something that we already mentioned is it’s not tailored for that. It’s not it’s not designed to do that. Right? It can, and it can fulfill that to a small degree, but it’s not gonna really flourish.

And it’s not as baked in as front line assistant. It’s not as dedicated to it. So Yeah.

It’s there, and it can function to a degree, but it’s not gonna work as well as you might want it to. Also, the feedback loops I mentioned earlier where it’s if it’s unable to answer a question, unless you were to maybe build an app with one of their bots, which would require some technical know how, you’re not gonna get that information on what’s not being answered. Whereas, interfaces solution will give you those reports on what information is missing, and you can update that.

Yeah. And even in, you know, in Austin’s case, like, they also use our voice AI on the member facing part. So when the member facing voice AI, when you call their contacts contact, you know, one eight hundred number or something there on the contact us page, you can experience that. If their AI is unable to answer that, it’ll be passing on the context back to the employee AI.

So so because all one AI brain, so it is able to now pass the context and help where AI left off for the member. So that that is also not possible, you know, in Microsoft Copilot and things like that. Right? So, yeah, that’s a good point about the dashboards and feedback loop, which is really important part of making AI project successful.

I didn’t know they don’t offer that, Austin. That’s a good in information for us.

It seems like there is lot of other questions already answered.

I think there is some one I see is probably specific to you.

Where within Credit Union of America site does the chatbot interact with members or potential members looking at the site? I went to check out their site. It’s spun for several minutes and connected to a live agent. Does the member have to be logged in in the online banking? I think you don’t use a chat AI, so I think we clarified that. But if you wanna add something.

Yeah. No. I I I did see that just because I was like, oh, no. They couldn’t get the help they needed for a second.

But I thought that they did connect with somebody. But, yeah, main thing there is we we currently don’t use interfaces, chat AI solution. And I think Jack was yeah. We’re correct.

We use Glia at the moment.

So yeah. Right now, it’s just all hands on. We have people dedicated for, chat interactions.

Wonderful.

Nice. Alright. I wanted to just say again, thank you so much to Austin and Credit Union of America for joining us today.

Austin’s always been very generous with his time when we have questions, and we wanna sit with them and hear more about how they use our products and services. So just really wanted to thank you again. And to everyone listening today and that will hear the recording, please give us a call. We’d love to talk to you as well about how Employee AI can help you.

So with that, I hope you have a fantastic day, a great weekend, and hopefully, we’ll see a lot of you at GAC this year. Yeah. We’re gonna have a booth. You can’t miss us and a phone booth as well, so come see us in the phone booth.

Perfect.

Awesome. Have a great day, everyone. See you.

Thank you.

Bye bye. Thanks, Austin.