Unpacking EFCU's AI Strategy: Driving Automation and Frontline Efficiency with Voice AI & Employee Co-Pilot

Welcome all. I’m sure everybody is, kind of following AI, the progress in AI very closely.

And the reason you probably joined this webinar is to figure out how you can take advantage of AI and apply it to your credit union and community bank.

And this webinar will give you very good insights on how you can harness the power of AI for your credit union and community banks. We are lucky to have Tyler Brooks, from EFCU.

They have recently deployed AI voice and, frontline AI. And he will kind of give us insight into why they decided to do something with AI, how, why did they select Interface dot AI, and now talk about the impact.

Obviously, they have been around using it for two months, so they’re very early in their journey. So it will be good to hear from Tyler, and I’ll quickly introduce him very soon. So the topic of the webinar is unpacking EFCU’s AI strategy, driving automation and augmentation, automation on the on the incoming calls and augmentation for the employees.

So here are the presenters.

So Tyler Brooks, as I mentioned, he’s joining us from EFCU.

He’s a VP of projects and innovation.

Tyler, maybe you can quickly say hi and introduce yourself.

Yeah. Like Jack said, I’m the VP of projects and innovation for EFCU Financial.

Been with, EFCU for eleven years now and in the credit union industry for, almost fifteen this year.

Worked on the operation side of things and just kind of, moved up over the years. But, yeah, yeah, projects and innovation is what I’m into now. It’s fun times.

Yeah.

Okay. And then the next presenter is, Srinivas.

He is gonna join us soon unless Shree is here.

So Shree is the founder and the CEO of Interface dot ai, and he’s, kind of one of the most visionary person I have worked with. And, he’s kind of mission driven founder, comes from the background of, Credit Union, in India.

And, he has built this company over this ten years, with his vision. And we had done many, many interesting things. But as I say, we’re just getting started. And he will show us what is coming in the near future with the agent.

I’m looking very much forward to it. And my name is Jack Chawla. I’m, VP of product marketing. I’ve been lucky to work with, a lot of mega trends in the industry over the last twenty, twenty five years, starting with, when email was a big thing, and then, of course, iPhone and texting and so on and so forth.

And I’m really, really excited about what is feasible with AI. It will be a revolutionary technology, and we’ll learn more both from Tyler Brooks, who is a practitioner, and the visionary, Huddl Shree. So we’ll get they’ll give us a idea of how AI is actually helping right now and what is coming, very soon, which, credit union and community banks can take advantage of.

Okay. So this is kind of the overall agenda. I’ll quickly talk about interface dot ai so you guys have a framework on what we are offering. And then we’ll dig deeper with Tyler on the case study and see, why they decided to do something with AI, why did they select interface, and how what is the impact.

And then she will come in, and he will talk about where the AI is going. And, of course, I’m I don’t wanna steal, Shree’s thunder, but one of the big things in evolution of AI is agentic AI. And then we’ll talk about we’ll show you the demo of what agentic AI really means in the context of the use cases which are important to credit unions and community banks. And then we’ll do a q and a.

We’ll obviously do q and a at the end, but please feel free to type in your questions as we are going along. You will see q and a kind of an icon at the bottom of your user interface on Zoom, and you can start asking questions. And we have, some of our colleagues like Harry, Kate, and Jay who are who will answer the question as they go go along. And some other juicy questions, we’ll, save it for as for, like, q and a, especially the ones we have for Tyler, and we’ll answer that at the end of the call.

If you’ve missed any questions, we’ll answer them after the webinar.

Okay. So before we get started, Kate, if you can kick in the quiz, let’s get a feel for where people are on their AI journey.

Okay. So you guys see the poll there? We’ll give you about thirty minutes thirty seconds, not thirty minutes, to fill out the poll, and then we’ll show you the results. And that also gives us some good idea of, what people are, where they are in their journey.

Looks like the answers are coming in. Wow.

Okay. So in terms of the solution deployed, amazing. Seventy seven percent haven’t done anything with AI voice, AI chat.

Great. So you’ll learn a lot on how it is actually helping the customers.

Have you currently deploy live chat?

Sixty four percent have said yes.

Thirty six percent have said no.

Okay. That’s kind of math doesn’t math is not matting in terms of people who are deployed chat and they have done none of the above. Anyway, so fair amount of people who have kind of done something with AI chat, not much with AI voice, it looks like.

And then, a few fair amount of people want to upgrade their phone system. So we will talk about how AI can AI voice can actually help in that kind of, project.

And then people planning to upgrade, about thirty one percent, are looking to upgrade. So okay. That’s great data. Thank you so much. So I’m gonna stop sharing the results, and let’s move on.

Okay. So a little bit about Interface dot AI. We have been in the business almost for ten years. We’ll compete over ten years today.

And, as I mentioned, Sri is one of the cofounder and, he had a vision on how AI can help, credit unions and community banks. So we are kinda mission driven, just focused on credit unions and community banks.

We are probably the only AI vendor focused on credit union community bank which has been featured by Gartner. We work with, the associations you are familiar with, like CUNA and your State Leagues.

If you’re a light solution customer, you’ll be happy to know that we are the only AI vendor they work with. So if you are interested, then reach out to Allied Solution or us, and we can jointly work with you. And then, in terms of, kind of the vision we are bringing to the market, we have been rated as, the best in shows number of time by Finovate.

Very tough competition in AI as you know, and we are grateful that we have been able to kind of show that we are doing a lot of visionary stuff, visionary and practical at the same time.

And we have about hundred credit unions and community banks using us right now.

And I think our number of live customers we have is the highest in the industry, and, I’ll talk about how that is an advantage in terms of our data.

And the solutions we are providing so if you are in charge of, contact center, obviously, voice is your biggest channel. So we provide you AI voice agents, which help you automate the incoming calls.

And, of course, when you automate many, kind of, ordinary calls, like what is my account balance, what is the routing number, and so on, you save time, and that time can be used for more complex issue. And then also, as you can kind of imagine, the the wait time will go down. The people who are waiting for complex answer don’t have to wait in the queue, in front behind the people who ask simple question. So the abandonment rate goes down. Right? So this is, AI voice is a big kind of a solution for the context and the people.

Then if you are responsible for digital transformation, then AI copolytes for the website and mobile apps is what we offer.

AI copilots for the website is obviously, you can think of it as AI chat, but we go above and beyond where, of course, you can do member service, but you can do a lot of interesting thing from marketing customer experience perspective, education perspective, financial planning perspective. So there are a lot of things which can be done by, this, AI Copilot.

So if you are especially if you are a CMO trying to figure out how to attract young audience, look at Copilot as a as a as a wonderful tool to go beyond member service and do all these other use cases as well for you guys.

Then AI Copilot for frontline employees is, kind of a companion. So automation and augmentation. And AI Copilot for frontline employees are designed to augment your frontline employees. The frontline employees, the two big populations are sitting in your contact center, or they’re sitting in a branch.

So we have solution for them as well. And then we have Spear Nexus, which is our agentic solution. We’ll talk more about this, when we see the demo from our cofounder, Sri. So across this hundred customers, we are serving sixteen million plus members every day.

And then we are doing one point five million conversation daily. Right? So this is kind of important because we are working with so much data. And as you know, like, in the self driving, the more data you have, the more you learn.

Similarly, the more interactions we have, more we are learning, and our AI gets smarter every day. And we are a QSO as well. So as we said, we are highly dedicated and mission driven, focusing on credit union community banks. And, you’ll find that we are not a general horizontal place, kind of serving this industry as well.

We are hundred percent focused on this industry.

So this is kind of a market texture of our solution. We talked about our solution, voice AI, employee AI, chat AI. The main thing I wanna point out is this is sitting on a agentic banking platform.

So this is important because this agentic banking platform is actually becomes a one brain. And, AI is, you kind of basically work with it in, at the platform level, and then it gets exposed to all the different solutions we have. Right? So it becomes, like, in a way, like, if you are technical, write ones, run everywhere. It’s kind of a similar concept.

Kind of make your AI work in one place and then apply to many, many different use cases.

And then over ten years, we have done forty plus integrations.

We are integrated with all the common codes and so on and so forth. And then we have, in the recent years, we are also building the fraud solution on top of our platform. So caller ID verification, biometrics, and all this is built in. So you don’t have to look at another vendor to provide that is kinda right built in and provides you to make sure that, you’re preventing fraud not only in on the AI automation, but when you transfer to the human, all the authentication and the caller ID verification is done.

So this becomes much more elegant solution than simply, like, a horizontal AI that people don’t even know what what goes on the FI. And this is all the knowledge we have gained and applied it, to our platform. So this all this is all the learnings we have got working with hundred plus credit unions, and this is kind of built in. And when you come in as a new customer, you are getting all these advantages there that the learning we got from data, the typical intents people are using, they’re already built in.

So you get a very fast start. Right? So this is the reason, we are able to have Tyler join in only two months into the journey and being happy enough that he’s willing to talk to you guys. So in two months, we are able to get him to a certain amount of automation.

We’ll talk about it. And, it’s a smooth enough journey that there are no major issues which, you might find with working with a non, with a with a vendor who is not focused on this particular industry. So, Tyler, maybe you can tell us a little bit more about, your EFCU Financial, what kind of members you’re serving and so on and so forth?

Yeah. We’re EFCU Financial. We’ve been serving members since nineteen thirty four.

We’re right at one point two billion asset size, making us the second largest credit union in the greater Baton Rouge area.

And we currently have around sixty eight thousand members that we serve.

We originally started out as Exxon’s credit union. So we still retain a lot of, ExxonMobil employees, as a base of our membership.

But we are community chartered, so, we serve nine parishes, in and around the Baton Rouge area.

Great.

Okay. So let’s start off with the big question.

So, Tyler, what is the EFCO AI strategy?

Tell us about how did you kind of think about AI and, the projects you thought which will help. So really want to kind of get a feel for, how do you strategize around AI.

Yeah. Well, I mean, we’re just getting started in our AI journey.

Interface AI was the first form that we rolled out so far here at the credit union. We decided to tackle voice AI, for automation purposes, helping to automate some of our calls that are coming into our contact center. And then we also, did the frontline AI assistant, like you mentioned earlier for augmentation, just enhancing what’s already there, helping our team members that are on the frontline when they have questions from members, having one database that has all of our training materials in it. The AI can search through and help them find the answers in a quick manner. So that’s kind of the start. I think we’ll, you know, we’ll talk about it later, I know, in other slides. But we have other things on the horizons for from an AI perspective, but we were super excited to get this rolled out and to help our team and help our members.

Yeah. Tyler, maybe you can give a little bit insight on how your organization is configured to kind of think about AI. Is there a, like, a working group around AI? Or how do you guys make a decision on what AI projects to take on?

That’s like conversations that we’re having now. This kinda originally started on the operation side.

Myself, our COO, our VP of contact center, kinda got together and started, looking for solutions to help or it all originated from, like, the contact center and how we can, help them, and then that kind of branched out into the frontline as well. But those conversations started on with the operations team. I think in the future, like, we may have some sort of, like, AI, committee internally.

But now we’re talking about AI policies that we need to put in place for the organization as we start developing or getting more, technology in the AI space. And then from there, you know, who’s gonna be responsible to vet these future vendors, and, you know, making sure that we’re using it responsibly internally.

Got it. Got it. So this was a no brainer project you guys got started, and then now you have a, building on kind of a process around how do you relate other AI projects. Great. Okay. So we’ll talk about the road map, and, Tyler will share, what other AI projects he is looking at.

So from the strategy to product, the strategy was automation and augmentation to to execute on the strategy. Tyler’s team basically have deployed AI voice, which is the automation on the phone side. And the frontline assistant, which is basically the AI for employees, which can answer, the questions in the call center and branches.

And we have special treat. Tyler is actually gonna do a demo of their implementation on frontline so you guys can actually see what kind of questions he’s asked, what kind of questions can be asked in the frontline, and how it helps the employees.

Okay. So let’s double click on AI voice, which is again, as a reminder, call automation.

And as we talked, the voice is the biggest channel which you need to kind of hand, tackle.

Lot of issues with, hiring, retaining, the abandonment rate and so on. And AI obviously can help. So we’ll dig deeper into this.

If you guys wanna try out the EV, if I’m pronouncing it right, which is the name of the AI voice agent, for EFCU, You guys can call this number and give it a try. If you’re not a member, you’ll basically experience it on the before the authentication. If you’re a member, you can actually authenticate and see, how it is able to handle the transactions as well.

Okay. So the next question for you, Tyler, what were the problems you were trying to solve for your members, and for your own organization?

Yeah. So, like, whenever we first started having conversations and, like, looking for alternative solutions, you know, originally, we were discussing, like, you know, do we need to expand our contact center? And then we started looking for other things that are in the the marketplace. But, ultimately, the goal was to kinda reduce our abandonment rate, and our extended hold times. Like, a lot of, at the time, probably a year or so ago, we our abandoned rate was kinda creeping up. It was higher than industry standards.

And so at that point, we were, you know, trying to figure out ways to get that number down as well as, like, the whole time, you know, members don’t wanna be on hold for ten plus minutes. So how can we, alleviate some of that? We also utilize a third party contact center for overflow and after hour calls, which can be pretty costly, whenever we have a lot of calls overflowing to them. And so whenever our, hold times were high, a lot of calls were getting rolled over to our third party contact center. So we were trying to figure out how we can minimize the usage there.

We also wanted some additional protection for fraud, which Interface AI has, you know, different, things available for that, like the one time passcode and the authentication, for different types of transactions. So we wanted some added layers of verification for the members.

Ultimately, like, just automation of some of the simple tasks. You know, a lot of members will call in, stay on hold for three or four minutes just to get a balance on their account. We wanted to free up our our reps to be able to answer some more complex, questions.

You know, the automated teller, while it is, available, it’s maybe not as advanced as things in the marketplace, today. So we want to find a different solution for that and try to drive, to to automate more of those simple tasks.

And then also, like, it gave us the option for routing calls, quicker to to get, like, a a resolution faster. So we were able to, like with rolling out Interface AI, because the member has to tell, like, why they are calling in, we can kinda direct those calls in a faster manner versus just having a standard IVR, which typically ends up getting to our contact center anyway, and then they have to route the call.

So, you know, with a a smarter solution, you’re able to get your resolutions faster.

Alright. So in the call center lingo, you are able to really have a very high first call resolution because you’re able to kinda do a bull’s eye routing to the exactly the right expert.

Right? Okay. So how did you go about the partner selection process, and where do you look for partners? Like, how did you discover who the right partner is? And And then, of course, thank you for selecting Interface dot ai. And then why did you select Interface dot ai?

Well, like I said, originally, conversations were being had about, like, you know, do we expand our contact center, add more team members? Do we overflow to our branches instead of using our third party contact center?

But one of the local credit unions here in Baton Rouge, went live with Interface AI. So that’s kinda how we learned about it, and we started doing some research around it, asking questions.

From there, we were intrigued by the solution, but we wanted to see what other vendors are out in the marketplace.

So we re researched, like, kind of the big names that are out there, scheduled some demos along with Interface AI.

Ultimately, we we selected Interface AI to be our partner just because of the we we really liked that it was a financial industry focus. We felt like, you know, because you were built for credit unions and community banks, the solution was gonna be much better and stronger and a better experience for our members.

Also found that, like, the AI voice, side of things was further developed than some of the other vendors at the time that we did our demos.

So we felt like it was a a stronger solution overall to, to implement.

And then we also felt like the team at Interface AI just understood what we needed as a credit union. They understood how credit unions operated. They understood what our members needed, what we needed on our side. So we felt, you know, very comfortable partnering, with Interface AI.

Great. Thank you so much. Yeah. Given our focus on the industry for ten years, we really understand this industry and the intense come kind of people and so on as we talk at the at the top.

So it’s, grateful to hear that our strategy is actually working for the credit unions and community banks. Okay. So could you walk us through the implementation journey? Obviously, AI is not the simplest of the project you can take on.

So I wanted to understand how you, what was your kind of the journey?

Yeah. It was a I I don’t think we knew exactly what we were getting into.

But as we got, like, the first part of the workbook that we were like, woah. Like, this is a lot.

It’s definitely tedious.

It’s cumbersome. There’s a lot that’s involved with it, but that’s a good thing too. You know, this is impacting your members directly. This is going to be a service disruption if you don’t get it right.

So the more, the better. Even though it’s really intense, especially, like, the testing, scripts were very specific. We had to go through, I I wanna say it was over a hundred pages of scripts, that we had to go through and test. We kinda divvied it up amongst our team members, you know, that were part of the project team, to where we could get through them all.

But at at the end of it, we felt very confident in rolling out the solution that we had tested everything, that we had thought of everything.

Throughout the process, we did have, like, a change in our project manager.

But, overall, that still wasn’t a bad it didn’t cause an upset. We felt like our project manager was really there when we needed the assistance, made sure to talk to us in our terms. You know, we were on the operation side of things kinda driving this project. It’s not like, our IT team was driving it. So, like, we didn’t understand all the IT or the technical lingo. They would put it in our terms to make sure that we really understood what we needed to accomplish and just made sure that we stayed on task throughout.

We are still learning and still correcting experiences.

You know, the initial experiences that you roll out when you’re filling out the workbook. You think you know the answer or you think you wanted this call to get routed here, but then whenever you go through testing or you go live with it, you realize, oh, no. It would be better if we routed calls to here instead of here. Let’s make this change. So the Interface team’s still working with us to make some changes and adjustments just to make the experience even better for our members.

The ticketing system was really organized, very transparent, a lot of comments going back and forth on the tickets that we had open, and, Interface definitely made sure that our top priority items were corrected prior to us going live.

And I will say that I really appreciated the Interface team that they were very receptive to our feedback, whether that was good or bad. We had a couple times where we were like, no. We do not like how this is. This is not gonna work for us. We need to find a solution.

And they were like, okay. Let let’s talk about it. Let’s try to find something that’s gonna work for the both of us. So I really appreciate that.

Great. And did you go to hire AI specialist for this project, Tyler?

No. No. We used our internal team that we already had.

Okay. So, I mean, this is one of the specialty of interface. Like, we are really a managed service. So all we need is, business people on this on the other side, to make it work. You don’t need any technical people to be added to the project. So okay.

Great.

So this is, great to hear. Fairly complex project.

You’ll be happy to hear, Tyler, that with the generative AI, this will become much easier.

The tedious work will kind of reduce to a great extent, and we’ll talk more about it when kind of she talks about generative AI and agent tech AI.

Okay. So, of course, we have been live for sixty days or so, but, let’s still see what has been the impact so far, with this AI voice. And what is the adoption? Like, there are two sides to the coin. One is, kind of you you might have the greatest technology, but you really need their members to adopt it. They’re, they’re kind of used to IVR. They need to kind of change how they kind of interact with your company.

So also talk about, kind of how are the members reacting to it.

Yeah. So just to give you, like, some background on all, like, our credit union, we have a very lean contact center. We’re we’re a lean credit union overall. So, like, as far as reps in our contact center, we only have seven reps, and then we have, like, a supervisor and then a manager over it.

They’re servicing our sixty eight thousand members, and they’re receiving around twenty one thousand calls per month.

This was prior to us going live with Eevi. So now that we’re only live for a couple months, we see, like, according to, like, the analytics that we’re able to pull, we see automation at over fifty percent automation of calls. Some of that can be partially automated. Some of it can be fully automated, but there is, you know, a piece of the call that is being automated, prior to it reaching our contact center.

We’ve also seen thirty percent reduction in calls going to our contact center, and we do have confidence that that number will continue to grow as we develop new experiences, for our members.

One of the big things is we definitely saw a reduction to our third party contact center. The calls that were being overflowed or going after hours to that contact center.

Just to give you some numbers, in January, we had thirty nine hundred calls.

February is when we went live. We went live February eleventh. So the month of February, which is already a short month, the calls had dropped to around twenty six hundred.

March was down to sixteen hundred, which is actually the lowest that we’ve had in the last two years of calls.

And then so far this month, we’re on track to be around fifteen, sixteen hundred, as well. So we’ve definitely seen a reduction in, calls to our third party call call center for sure.

Also, like, the calls that are actually getting to our contact center because members are still you know, as this is a new solution, some of the members are still asking, like, they won’t speak to a representative, transfer me to a representative. But of those calls that are getting to our representatives, it’s not we definitely still see a decrease in our call queue prior to going live. We had calls that would be, on average, like, fifteen to twenty calls in queue at a time. Now we’re in the single digits. It’s something that’s manageable for our team that’s in our contact center.

And, we’re also seeing, like, with the analytics that we get on the back end, we can see what are some of the next best experiences for us to roll out for our members. So we can see why they’re calling in. We can see when they’re calling in, what are they asking when they call in, and then try to automate some further experiences from there. So that’s something that’s kinda on our agenda to do is to kinda roll out some of the other supported experiences, to automate even more calls.

Great.

Great.

I have a lot of questions, but I’ll move on in in the interest of time. So this is the automation level, forty percent to fifty three percent. Obviously, we’ll continue working with Tyler to move it higher as we learn, the issues which we are not handling well right now. Okay. So what are the best practices? What do you recommend people take on the similar project? What can they do to make sure that, they have a smooth journey?

I definitely think that, like, the marketing strategy is really important.

Interface AI gives you a a a good recommendation of what you should do from a marketing standpoint, and I really think you should take it up. Take take them up on that and, like, try to, you know, outline a plan for your marketing ahead of time. This is a member disruption.

Members, especially if you’re coming from just a standard IVR, members want to know this kind of stuff ahead of time, so just make sure you get that marketing information out. Let them know what your, AI assistant’s gonna be able to tackle, what they’re able to do, to whether members are, prepared prior to you going, live with it.

Think outside of the box. Don’t recreate your IVR. Just because it’s the way that you’re currently doing it doesn’t mean you have to keep doing it that way.

It’s a really flexible solution. There’s a lot of different experiences that can be made. So just kinda think outside the box when it comes to that.

Really think about the member experience. Like, you know, when you’re testing, think about how it sounds. Do you like it? Does the flow make sense if you were the member?

Pay attention to all the fine details. We found some, like, little small like, the way that the bot said this, we didn’t like that it said you know, how it said it or how it read this. Just pay attention to even the small details to make sure it’s a, you know, a a good represent representation for your credit union.

And just, ultimately, just make sure that it’s best assigned for your credit union, when you’re looking at the experiences. You don’t have to just take the suggested. You can go back to the interface team and say, you know, hey.

We don’t do it this way. This is how we do it. Can we make some adjustments?

And they’ll work with you.

Perfect. Perfect. Perfect. Okay. So now let’s move on to the frontline assistant.

So frontline assistant is kinda new concept to many, attendees.

So Tyler has gracefully agreed to do a quick demo of their front end assistant. So let me stop sharing. And, Tyler, maybe you can share your screen and show us a quick demo. So front end assistant to refresh your memory is for your employees, call center, and branches. And Tyler will show you how their call center employees and branches are actually using this product.

Yeah. So, we have both, you know, this this one system, both our contact center and branch use the same one. So some of the questions could be, you know, the same type of questions that they’re asking. But I’ll just give you some examples, What What it’s gonna do is just search our knowledge depository that we have here, and then we are able to click on the document here, to to see any kind of resources that we have surrounding, travel exceptions within within our knowledge depository. And then we ask our team to help with training it, by, you know, responding whether this was the accurate information that was given, or if they were searching for something else, they are able to provide us feedback to where we can continue to update the product.

Yeah. So one thing I wanna point out is when the question was asked, the right document came up, and we highlighted where the answer is coming from. Right? So it gives you you have confidence that the answer is actually coming from the the appropriate document, and the answer is right. So this kind of is a double check to make sure that the answer is right.

So what are our debit card limits?

And it’ll give you, like, the little summary right here so you can read this. But if you wanna see the document, then you just click view, and it opens up in this separate pane here. And it’ll take you to, like Jack said, the highlighted section of what our different limits are for debit cards or whatever the question is. And like I said, you can reply with feedback to it to make sure that it’s giving you the accurate information.

Yeah. And, Tyler, these are the documents you uploaded into our solution. Right?

Yes. These are our internal documents that we’ve uploaded here.

Yeah. And by the way, we can while Tyler is asking the question, you can also point it to your SharePoint or whatever your document sources, or you can decide to upload or you can do both so we can kind of, work with any documents you have.

Either you upload it to our system or integrating. Okay. So the next question is what type of checking account do you offer?

So, Tyler, you mentioned that these are the questions being asked by the contact center employees and also the branch. Right? When somebody walks to the branch, the employee doesn’t know the answer. They can simply type and get the answers immediately.

Yeah. It’s the same concept.

So, like, if someone came into a branch and, you know, wanted to what are I’ll do I’ll change Yeah.

So while Carlos is typing so the way we price the solution, you don’t need to worry about the seed price. Right? So you can deploy to anybody, like those typical CCaaS solution, which is seed based. You’re limited to only the contact center people. So this frontliner solution can be deployed to any of your employees.

Yeah. It’s, it’s really nice for sure. Answers right here at your fingertips. Quick and easy.

Okay, Tyler. Really appreciate you showing the demo. Let me share my screen, and let’s continue.

Okay. We saw the demo.

What are the problems you’re trying to solve with this? Obviously, the answer is pretty obvious, but very quickly, if you can tell us.

In the interest of time, please keep it short.

Turn yeah. Yeah. No. You’re fine. Turnover on the front line, always happening. So training’s always ongoing.

We just wanted a a place that we could have all of our training materials built in, an AI assistant that could help search and find answers quickly for our team.

Okay. Thank you. Thank you. Thank you. Okay. Then what kind of impact has it at? Obviously, only sixty days, but let’s see what the impact has been so far and the adoption.

Like I said, it’s still early. We’ve received positive feedback from our team. They love how quick they’re able to get the answers.

We’re gonna continue to promote it to our team to make sure that they’re adopting it and continuing to add new training materials, to further develop it. But, like, last month, we had six hundred and sixty five requests within, that system. So we thought that was a pretty good number.

Yeah. So, like, you have basically given them the chat GPT equivalent to your employees.

Okay. So the best practices for deploying frontline as well?

Just make sure you have up to date training materials.

Our team specifically wants, like, how to. So just maybe have some how to documents, promoted internally, make sure your managers are promoting it to their team members, and just keep everything up to date.

Well, got it. Okay.

Then, quickly, let’s talk about the future road map. As you mentioned, you’re looking at other AI projects as well. So let’s, understand how you are thinking about a AI road map.

Yeah. So we’re about to embark on our core and digital banking conversion. So that’s super exciting. I’m gonna be super busy. But we are looking at AI in the lending space, from a marketing standpoint as well, to, be able to do more target marketing, and then chat AI, in the future too.

Got it. Thank you, Tyler. I really appreciate your time. So let’s move on and talk about the evolution of AI. So, Sri, if you can, kind of give us your vision where AI is going and show us the agent decay demo.

Absolutely, Jack. So appreciate, Tyler for your demos and, and the feedback there. And and the impact is, wonderful to see what we are able to accomplish as a team together.

So, some great impact numbers there.

You know, AI is evolving really fast. Right? So, you know, in the last couple of years, we’ve probably seen onslaught of new capabilities.

And, we’re we as a company is continuously innovating how we can apply all of those technology into practical solutions that could drive value for banks and credit unions. Right? So, you know, largely, there are kind of three phases. If you put all the noise of AI into a, a, you know, simplified way to look at what’s really happening is, like, these are kind of three phases.

Right? There’s the past was conversational AI, what we used to call NLP, and the present is gen generative AI, and the future is agentic AI. Right? So, but, of course, there’s a lot of updates happening in each of those kind of, different technology sector.

But, really, converse conversation I was NLP, which was, you know, it was cumbersome to set up. It required a lot of manual effort to set things up. Even though we we made it little easy on our client by providing managed service, but still, we had to do a lot of work behind the scenes as well as our customers to some degree. But generally, AI kinda makes that easy, but generally as a technology is pretty much like, you know, understanding patterns from a large dataset, and recognizing, characteristics in the large data set and his ability to create novel data set that follows the same characteristics.

That’s what generative AI is. For example, if you, fed a million cat images, if you, ask a new cat image, is it a cat or a dog? It can tell you because it already has million cat images. But it can also create a a brand new cat image that you’ve never fed it in the past.

So it creates novel con novel content that kind of has those characteristics as being trained with. That’s what generative AI is. So that the the the generative AI now you can generate novel voice text, you know, and and things like that. Right?

So images and videos and what there’s explosion of technology there and the use cases.

So in a simple words, like, you know, if you’re, making a dish for, let’s say, the Thanksgiving, you wanna have a best recipe for, you know, for one of the dishes you’re cooking, generative AI can go to Internet, find the recipe for you. But agent AI, is kind of the next phase of evolution in the AI, which is, once you find the recipe, the agent AI can go order groceries on behalf of you going on amazon dot com or whole foods dot com and look at all the, preferences you had for brands or, you know, your, you know, dietary restrictions and kind of carefully shop and put all that in a shopping cart and get that delivered to you.

That’s agent AI. Right? So the world is about to change. We all thought, AI is going to change probably the blue color work more often, than but it turns out, you know, the white color work is getting disrupted, through AI as well.

So, you know, now how do we apply this technology in banking? That’s that’s where, the couple of slides here for, about the past, which is NLP or Conversation AI. What we did is, we embedded our AI into all channels of communication. Right?

Like, for example, if you’re on online mobile banking, we would have a AI chatbot. If you’re on if you’re calling, we would have had a AI voicebot to pick up the call.

You know, of course, if you’re a frontline staff, we had an AI for you. Where the AI worked as a first responder. Right? That was the past.

With generative AI, we can achieve a lot more. Right? So that’s the present. We call that phase unified, where you’re going to see a one AI brain evolve for the entire financial institution, and that’ll gonna be so powerful.

So we saw we’re achieving, fifty percent automation from, the impact metrics from, you know, Tyler where where recently.

That is going to go up to sixty to seventy percent. The call automation and things like that is gonna go up to sixty to seventy percent. Not only generative AI is gonna make this AI very powerful with, and and that makes it easy for them to launch with limited effort as well as it can be very powerful at doing significant automation, but it also reaches a point where some of the channels that we know as as in the past, would no longer require to exist. Right?

Like, so, for example, number one channel that will continue to play significantly less less role in the coming days is live chat. Live chat is gonna disappear in the coming days. Right? The AI is getting powerful enough that, you know, it is automating a a lot of them, but the rest that are remaining complex enough that people prefer to call, there is there is a very little room for live chat to play.

Right? So so that’s, you know, the the unified phase. We are seeing a one AI brain evolve, offers more impact and gets to a place some of your channels will no longer be relevant. Right?

That’s the journey we’re on currently. And the future is where we are going with agentic AI is we’ll reach a point where we’ll have a bank GPD for your customers and employees where you literally your customers or members who today already bank with multiple financial institution, An average number of bank accounts and, as someone living in the US, bank with us about somewhere between ten to twelve financial institutions.

So they’re already having extremely fragmented experience hopping between multiple financial institution tools, but, you know, with the next phase of evolution, which is what we call autonomous banking or bank GPT, your members or customers will be able to add all bank accounts in one bank GPT like AI assistant that you would have offered to them, and bank across multiple financial institution in the same place where the AI will start driving their financial wellness goal in an autonomous way. We are literally two to three years away from this. Right? Not too far away.

So that’s kind of pretty, exciting future to look out for. So what I’m gonna show you real quick if you, switch to the next slide, you know, pretty much we as a company have, built out solid, solutions, for both phases of AI, which is generative AI, agentic AI. Right? Today, you saw, some of the generative AI in action, which is on the employee side that, Taylor showed that is fully generative AI where they had to just upload documents or point to a knowledge base and it learned everything in a few minutes.

Right?

You know, in the future, you know, the agentic AI solution is gonna take that to a whole new level. And with the combination of the techno the solutions for both this technology evolution, we become the most capable AI banking platform. And I wanna take a couple of minutes to show you agent AI solutions.

If if you move to the next slide here real quick, which is Fear, Nexus, Orbit.

So, you know, all these three are our new solutions for agent AI era. Right? So and I I wanna spend a couple of minutes to show you this. So we’re gonna start with Arvid.

Arvid is a, a AI copilot, right? So, that sits in in your own, website, online mobile banking, as well as online applications. Right? Pretty much in all of your digital assets.

So we have created a hypothetical bank called interface bank. Don’t worry. We are not in the business of banking.

So you could see at the right bottom corner, you can see the AI. We’re gonna pop it open. We’re gonna ask a couple of questions, just to show you how today’s chatbot already works. Right? We’re gonna ask you what is the routing number.

Right? It quickly, shows you, what is the routing number and things like that. And, this is generative AI capability. Right?

Nothing new. You’ve seen probably these kind of solutions that in some of the financial institutions you go to. Now I’m gonna show you how agentic AI upgrade would look like. Right?

I’m gonna ask, you know, want to apply for credit card. Right?

And as we were looking at it, it asked you what additional features you need. I’m gonna say, want more savings.

Right?

And then, shows up and says, hey.

You know, do you want any other features or, you know, can we go can I go and do the research? I’m just gonna say not not anything right now. Go ahead.

It immediately takes over the entire, website on behalf of me. Start, you know, researching on behalf of me, looking through all the credit cards on the website, you know, pretty much doing a thorough analysis, what it could have taken you many minutes. Right? It does that by itself and comes back offers me, hey, you know what?

I did all the research. It is the best credit card that’s well suited for you. Right? And it says, you know, do you wanna apply?

I’m, you know, I’m gonna say, yes. What traditionally happens if you wanna apply for a credit card? Right? So you, pretty much, you know, up click apply online button, you go through a very cumbersome forms, you know, probably seven to eight different, forms.

You hop through one form to another to complete all the details.

So that’s what we do today. But how it’s gonna change in the agentic ARR? Let me show you. With the click of a button, things could be, very different. So I’m gonna say, let’s apply.

It says, you know what? I already looked at the entire application.

I want you to go and submit a few documents for me. I’ll upload the I’ll I’ll take care of, you know, applying for the entire application on behalf of you. So I’m gonna click, upload here real quick.

And I’m just going to upload a couple of sample documents, say upload.

It processes those documents and comes back and says, hey, I have all the information. Should I go and apply? I just say yes.

It reads all the information from those contents, starts applying on behalf of me and filling up the entire application form.

So you as a consumer sitting back, you know, relaxing and it, you know, it comes finish up the whole application process for you. So that’s the power of agent AI, you know. Pretty much, you know, with a click of a button, you’ll you’ll be safe to do another hundred clicks that you’re doing today. Right? So it’s pretty, powerful.

So let’s say rate. Right?

Yeah. Pretty much. But let’s say you’re online mobile banking. So I’m gonna log in here, and the AI will continue to be there for you on online mobile banking.

Right? So if you click and pop it open, right, so, you know, again, you can, ask any question. I’m gonna ask something you could you cannot do in online mobile banking today at all. Right? So I’m gonna say, hey.

How much I have been spending on streaming services?

Right?

Imagine trying to find this, on your online mobile banking today, which is, let’s say, your multiple accounts.

You have multiple statements for the accounts with multiple transactions.

You have to go through download those statements or scroll through multiple of these transactions and then find, you know, transactions that are related to streaming. Right? It it takes you forever to do that. Let me show you how agent AI can do that.

So I’m just gonna ask this question.

Comes back and says, hey. I’m gonna pull all this transaction, analyze and provide you a summary. Would you want me to go ahead? I’m just gonna say yes.

Again, my hands are off the keyboard.

It is actually going through the online mobile banking system, going through each account, different statements in each account, downloading them, going through transaction in each, and consolidating and summarizing me how much I’ve been spending on, you know, streaming services.

Right? And gives you a beautiful graph. Right?

So this is kind of the, you know, possibility with agentic AI, with just one of our products. I know we’re running out of time. I wanna make sure there’s time for, q and a.

We have other two products as well, as part of the follow-up to the webinar, recording, we can send you the demos of the other two products as well. Jack, back to you.

Yeah. Thank you, Sri.

So as you can see, agenting AI can solve a lot of problem. Right? You can finish the journey and, some of the issues we face in the industry, like people dropping off and so on will go away.

So then coming to the q and a, looks like, a lot of questions have already been answered.

Tyler has answered a lot of questions, while we are kind of talking. Let me see if there is any open question which you can answer.

Okay. So Cliff is asking, did you deploy to both website and mobile app? If yes, did you have any issues, challenges in supporting both platforms?

So Tyler’s actually not hasn’t deployed AI chat. So maybe, Sri, you can answer this question.

The challenges of deploying AI chat for website and mobile app.

Yeah. It’s it’s pretty much a, small JavaScript that you could embed on your website, and it automatically shows up both on your website and, in a mobile version of your website.

But if you need that inside your online mobile banking, you basically also take the same, couple of lines of JavaScript code and put it inside work with your online mobile banking vendor to add it. That’s that’s pretty much it.

Okay. So I’m I’m gonna read that question with Tyler who’s already answered. I’m not sure whether it goes to all the, attendees. So the question is, Tyler, how did you assist your client in adjusting to not working with a person? What demographics are easier to make the adjustment for?

Yeah. So with after it’s live, like, our contact center is the the biggest one promoting it. So whenever members are getting to our contact center reps, if it’s something that the AI could have assisted the member with, they’re making sure to let the members know, hey. Look. You wouldn’t have had to wait for us.

Our AI assistant’s able to take care of that for you. Make sure that whenever you, call in, you state what you’re calling for to where you’re appropriately routed, or they’ll just educate the members as they’re calling in, to, you know, help the member, but also help them with alleviating some of those calls that are coming through. But we also have, like, a a landing page on our website that’s dedicated to, stating what Eevee’s capabilities are, and then also sending out, like, marketing emails, social media posts just to remind members, what the AI could do. We did you know, obviously, the younger demographic, picks it up the easiest, but also some of the members that are used to an automated teller, are more willing to go through with, asking Eevee for, like, their balances or to do a transfer transactions.

And then some of the other members just wanna speak to a rep. But that’s fine. We’re still working on those members.

Okay. And I guess, one question which everybody can benefit from is, Tyler, the project team. What are the project team which are, kind of involved in implementing the solution?

We utilize, like I said, our operations team. So our chief operations officer, VP of contact center, and myself, we were, like, the core, project team. From there, we also had a couple members of IT that were involved for the integration piece. They worked with our, phone system, which was RingCentral, and then also our core, Fiserv XP two, to make sure that everything was integrated. But once they kinda got that connection up and running, they helped us with a few things throughout. But overall, it was, us three on the operation side that really saw the project through, and then we got some of our reps involved in the testing.

Okay.

Okay. So quite a few questions about what CCaaS been, you guys are integrated with, what core. Maybe, Sri, you’re gonna answer it in a generic way, like, what are those strategy around integrations?

Yeah. So, I mean, see, our strategy is to seamlessly plug in AI and provide immediate value regard not having to change any systems. Right?

So that’s kind of the philosophy we’ve been working with.

You know, regardless of the online online banking system, core banking system, you know, credit card processing system or, loan origination system you have, we have our solutions and products that plug in seamlessly.

So you don’t have to rip and replace anything. Right? So it just works off of that, in a seamless manner. So that’s been the philosophy and the goal of the company. So in you know, from day one, that’s why we went to Hire integrated practically with every single third party out there and stay neutral, in supporting, in in as seamlessly as possible.

Okay.

Great. So we are running out of time. Thank you. Thank you. Thank you very much for, joining us. And thanks, Shree, and thanks, Tyler, for giving you insights.