Meet the Industry's First Agentic Voice AI — Live in Credit Unions and Community Banks

Hello, everyone. I wanna welcome you and thank you for attending our webinar today. You’re meeting with the team here at Interface AI, and we’re so pleased that today we’re going to have you meet the industry’s first agentic AI.

I’m your host today for at least a minute. I’m Amber McCollum. I’m leading our West Coast team here at Interface AI, and I’m very pleased to turn this over, to Shri. Shri is the CEO here at Interface AI. And with that, I’d love to turn it over to you.

Thank you, Amber. Really appreciate it. Good morning. Good afternoon, everyone. Thank you for your time today.

We live in an exciting time, isn’t it? So, you know, AI is coming to help in many ways of all the challenges we as industry face.

I’m so excited to kind of, you know, talk about the the possibilities of agent AI and everything it has in house for us to help grow our creating and community banking businesses.

I’m super excited today. There is you know, there were over a hundred and thirty attendees in our webinar today across close to thirty states, about fifty three percent, you know, c level executives and and thirty two directors and fifteen percent practitioners.

You know, AI is a topic that interests everyone clearly. So we’re excited to, you know, kind of share our journey. So before we get started, here is our agenda for today. We’re going to quickly start off with introduction to interface AI, and we’re we’re we’re we’re going to talk about the advancements in AI and really understand how those advancements is going to help our industry, you know, and then we’re gonna show you some demos of, agentic AI, you know, in action, how it’s transforming member experience. And we’ll also hear from our visionaries who launched industry first agentic AI. We’ll, and we’ll do q and a. That’s the agenda for today.

You know, real quick, we are gonna do a quick poll just to understand where are you in the journey with the various AI solutions. There’s about five different questions that should be already on your screen. I would request you please take a few seconds to answer them. That would really help us to understand where you’re in the AI journey, and we can, you know, personalize this presentation and demos accordingly. This is a question about, what role you play in the financial institution, What is the current use of AI you’re considering? What are your goals, big concerns, and how familiar are you with some of the terms like generative AI, agentic AI?

Great. We’ll end the poll here. Great. These are the results from the poll.

You know, there’s about thirteen, of forty are c level executives who joined. Their VPs and IT team, risk team have joined, today.

How would you describe your current use of AI member customer experience?

Thirty five twenty six percent said just exploring, twenty percent small experiments, forty percent said live in one or two areas, and eleven percent said deployed across multichannel, and three percent said advanced working towards agentic autonomous use cases. Awesome. The third question, what is your primary goal for AI in next twelve to twenty four months? Thirty percent said improve self-service and satisfaction. Thirty seven percent said reduce operation operating cost and improve efficiency.

Seven percent and three in twenty three. Twenty three percent said build an overall AI strategy. Not sure where to start. Wonderful. It’s a great place you’ve come. Right? So then question four here, what is your biggest concern about using AI technology?

Twenty percent said losing the human touch.

Absolutely makes sense. And twenty seven percent said data privacy and security.

Twenty percent says compliance, legal risk. Twenty percent said cost complexity, change management.

Great. We’ll talk a lot about change management today, from our visionaries on the call. And the, fifth question, how familiar do you feel with the terms like generative and agent AI?

Seven percent said brand new. Twenty three percent said somewhat familiar. Forty three percent said comfortable can explain to colleagues. Twenty seven percent said very familiar.

Already working with this concept. Wonderful. We have a a well educated audience who are trying to understand, you already know where to apply AI. They are concerned about, some of the regulated risk and the change management, want to know where AI could take them.

They wanna understand the road map. They probably know what to do right now already, which is great. Awesome. We’re gonna cover all those topics.

So appreciate participating in the poll.

So let’s

kick it off. The to start with a quick introduction about myself and the company. It always helps to know who you’re listening from. So, I go by Sri.

I’m the co founder CEO of Interface dot ai. I actually come from a family. Most of you may not know. My father started a Credit Union, scaled it up to a couple of billion.

I was, helping him through the digital strategy through during my college years. And I went on to work at Microsoft, part of their AI research lab on bing dot com. So when I found my calling, and I bought my personal and professional experience together, AI and credit unions and community banking experience to build interface AI with the entire company mission is to empower every financial institution like yourself to scale efficiently and help your customers achieve financial well-being. Right?

So that’s literally close to two hundred people in the organization wake up every day excited towards making contribution to that mission. Our vision as a company is to help you leapfrog from digital to intelligent banking.

You know, we have been, know, we’ve been doing this for a long time. We’re ten years in the business, longest standing AI vendor. We started doing AI before it was cool.

Right? So and we are featured in Gartner, Kuna, Allied, and as a best AI solution, including we won several awards at Finowate American Banker as the best AI award. We build a lot of things, but today, or rather to I mean, as a company, we only focus on using AI to improve your self-service. Because AI can do many things in the world today, but our specialty is to help Canadians and Canadians to improve their self-service for their members and customers.

That’s that’s the focus of what our capabilities are, and you’re gonna hear a lot about how far we made progress on those and also from our visionaries. Right? So, you know, in terms of AI, right, so there is quite a lot of noise out there in terms of what is possible. But this is a statement from Ilya, who is the cofounder of OpenAI, talking about, you may not take interest in AI, but AI will take interest in you.

Right? So the reason he kinda calls that out is in about a year from now, AI will reach the skill set of a graduate student in coding and mathematics.

Right? So, you know, pretty much a graduate level student capability. Right? So in a in a year. So and in two to three years, AI is going to become, achieve general artificial intelligence, as in it would be better or equal to the best expert in any field.

Right? In about five to eight years, there is a prediction that AI will reach super general intelligence.

That what that means is AI is gonna be better or equal to a group of us on on our knowledge and expertise, not just each of us. It will be better or equal to some of us. Right? So, that’s how fast things are evolving.

You know, AI is growing six times faster than Moore’s Law.

Moore’s Law made computers and mobiles happen, and AI is growing six times faster in in in its speed and capability and things like that. So there is, of course, a concern in terms of ensuring whether human capabilities will evolve as fast as AI can and the need for policy changes and things like that.

So what can we do with such powerful AI?

You know, if if you’re an optimist, you would think about there are a thousand challenges in our industry scaling and building a successful credit in our community bank. We are excited. Hey. We can use these tools to really make a huge difference in our businesses and thereby to the life of our members and customers.

If you’re a pessimist, you’re probably worried about, you know, that how this could impact the jobs and things like that. You know, I think the either side is not wrong, but I think we’re all on different sides or probably on we believe in a little bit both both sides.

So I am an optimist. So I look at what problems we can solve. We got so many problems getting our businesses. I’m thankful about AI. So our consumers are facing significant challenges today. Our members and our customers, you know, they have disjoint finances.

They none of I mean, our members and customers bank with ten to twelve different financial institutions today. You know, they don’t have a consolidated view of their entire finances today at all. Right? You have to be very savvy to build everything on an Excel and then you would have it.

There’s not a single three sixty degree view of our finances.

And two third of, US consumers live paycheck to paycheck. Fifty percent of our US consumers are living under five hundred dollars in savings.

Right? So it it’s it’s a crisis. Right? You know? And about seventeen percent of US consumers are considered financially literate because we we don’t get taught about how to manage our money in school.

Right? So and it’s a learned skill. Right? So and and consumer expectation is changing. Right?

They want it to be everything instant, proactive, personalized, you know, why don’t you tell me, right, kind of an attitude. Right? So they want coaching guidance, you know, given where they are in the journey and lot of stress they’re carrying.

So, yes, that’s why I’m thankful for AI if it can solve these problems.

Right?

Not only our members and customers have challenges, that AI could overcome in a significant way, but our institutions have a lot of challenges too.

You know, we have a lot of fragmented systems and data silos. There’s no single view for us too about our members and customers.

High operational costs and training needs.

And while the impact is low, sometimes we are spending a lot of time on routine tasks.

There is a significant challenge in attracting talent in our industry.

So, you know, we need to find ways to attract talent and kind of become cool again in some way. Right? And then raising customer expectation. Customers want twenty four seven support, and they are being spoiled by other fintechs and companies like Amazon, providing the best of the experience, for our customers and members.

We, unfortunately, have slow innovation cycle. It takes a long time for us to launch new products and test out something, experiment, and double down.

A limited personalization, our wallet share is, not growing as fast we all want. We all wanna be the primary financial institution to our members, and it’s challenging to get there.

Regulatory compliance pressures, you know, fraud risk, and then generational churn risk. You know, we are, looking for ways to attract younger members.

You know, there is so many challenges we all deal with every day and strategically trying to work towards them. You know, AI can help with a lot of these. Right? So that’s the rest of the presentation to really kind of show you products and road maps and where is industry heading to solve most of these problems using AI. Like, that’s why it’s so exciting time.

And and, you know, we we pray for this AI to get better to solve some of these challenges and eventually serve our members to achieve the financial well-being. Right? So but not all AI is equal.

Right? So especially when I say AI, I’m talking the context of just member experience, AI for member experience. Right? You know, that’s where our expertise is. So

not all AI is equal.

So there are different kinds of AI. The latest and more capable AI is agentic AI. But let’s look at different ones. Here, we created a simple framework for you to understand what this different kind of AI really means. The framework is we understand how AI under AI understands your member request or customer request.

And we also look at how AI creates the response that is going to share it with the member or a customer. So when we we’re gonna look at it in that basic framework, how it understands the member request and how it generates a response. Right?

Three, four years back, including interface AI, we were here. You know, with the technology, it was very hard to even build this, honestly, and we were here. Right? So it was understanding member request was extremely limited and poor, and response, was highly scripted. Right?

About two, two and a half years back, we got into a journey of, being a partial generative AI, technology, where member an AI understanding member request become very superior, but the responses were still highly scripted because we were worried about hallucination and, and the risk of security, risk of attacks and all of that.

And then about a year back, we moved on to fully generative AI where the member and and the same member request become very superior. At the same time, response becomes superior, but it was limited to informational queries only.

Right? Because, you know, what is the routing number? What are interest rates and things like that? Now we come to agentic AI today. Right? And you’re gonna listen from some of the visionaries who launched agentic AI.

They can share their experiences where the request have become far more superior in understanding even your, voice and tone and are you, unhappy. A lot of sensitive things beyond just the words.

And the response have become fully autonomous and can take action to drive a transaction, you know, complete end to end fully autonomously.

While previous generation, it was scripted, the transactional experiences. But in the agentic error, everything happens fully autonomously. Right? So with minimal direction, from you in, hey.

Do this. Don’t do this. Here are the constraints. Hey. I’ll figure it out the rest.

Right? So it’s become very powerful.

So what does that kind of technology enable us to do in member experience?

You know, we believe a typical approach of using such an agentic AI is something like this. You want to deploy this to voice and chat channels of yours, where typically about twenty, thirty percent of your member interactions actually go there. Right?

But, you know, sixty, seventy percent of our member customer interactions, which are happening online mobile banking, is not AI enabled at all.

But someone with a bold agentic vision, here is the path you could take, which is hundred percent of member interactions could be AI enabled. That’s what agentic AI the power of agentic AI makes it happen. So don’t be stuck in a trap of agentic AI being used in voice and chat only.

You know? Today, agentic AI is so powerful, that it is actually, limiting agentic AI to only using voice and chat. You won’t be able to use the full capability of agentic AI if you limit it to only voice and chat because it has ability to generate a video on the fly and use multimodality. Like, you can’t do any of that any of that in voice and chat. But if fully realized vision from your partner for agent AI would also take on building an AI native online mobile banking, thereby a hundred percent of your interaction could be AI powered. So that’s what agent AI could, you know, make it happen.

You know? And when such a powerful technology comes across, you will be able to create a one AI brain for organization wide. You know, the future we see with AgenTiC AI is, you know, your customers and employees are talking to one AI brain. That AI brain is insightful, automates everything you ask for.

It coaches them to, you know, sign up for new products or recommend new products in case of employees.

It could be like a chat GPD like interface, much better, and we’re gonna show that demo today.

It’ll be extremely personalized. It learns about you and your employees every day through their interaction, and it’s also gonna be secure. That’s the one AI brain, which all connects back to your system of record, like your core banking, LOS, online mall banking sorry, debit cards and RPA system and whatnot. Right? So that’s the future we see because of this immense capability of agentic AI, which has a ability to autonomously complete any task and guide you towards your financial well-being.

So but it’s not about one AI brain organization wide. It, you know, it also has a quite a bit of depth in what it’s able to do. And how we see the future is with agentic AI, it it may start off with one AI brain for the organization as we see. That’s just the beginning.

And AI could get to a place where it can increase your wallet share and financial well-being by coaching your coaching your members and customers through the through the interaction. Because if AI is able to complete something autonomously, which means you give it a goal, it figures out how to get there by itself. Right? And following your constraints and, you know, ensuring it’s not hallucinating, ensuring it is not creating any security risk.

But it has ability it has a brain of its own to navigate to get to the path, get to the goal you outlined. And the goal could be, hey, increase wallet share. And remember goal or customer goal could be increase financial well-being. AI will autonomously guide them.

And AI will also be able to take actions on a computer just like a human being and complete any task autonomously. We are no longer will be limited by, you know, siloed systems, which doesn’t talk to each other because we don’t have APIs. Our third parties don’t have APIs. Those days are gone with AI.

AI, agentic AI, can actually take actions autonomously on, you know, computer applications just like your human agents would do. Right? So it doesn’t matter if there’s no API. You can still have AI complete the task.

So that’s what’s possible with agentic AI.

AI agentic AI is soon going to replace the traditional online mobile banking and contact center systems. And we’re gonna show you demos of these today. It’s kinda mind blowing what agent AI could do in challenging the status quo, of point and click interfaces. Right? Who it’s like the days of, iPhone as a touch screen.

Right? You know, the the amazing thing happened with touch screen is we got used to its own naturalities. We don’t wanna go back to buttons. You can’t zoom a picture if you go there.

You can’t draw. Right? Similar to that in agentic AI, you know, if you can give it a goal and it’ll get you there, why would you need to figure out where to click in a in an online banking? It’s not needed.

It’ll do it. It’s an assistant. It’ll take care of it for you.

So that’ll be fundamental shift. People would never want to come back to the current reality. Right?

In the future, you know, AI will end up becoming a universal banking assistant.

So you could offer AI like a bank GPT kind of AI to your members that they could use on as an app or online or on a call, and then they could be able to add any other financial institution that they bank with within that universal banking assistant you offer to them.

Thereby removing the, you know, the silos and the the lack of three sixty degree view they don’t have across their finances can be solved through this universal banking assistant.

And we’ll show that in action as well. And AI will get to a place where it will dry it will be able to provide advice to drive financial goals, and very soon, it can be autonomously.

When AI can drive your financial goals autonomously, why not it can promise units of basic income?

You know, if we are worried about AI losing jobs, AI can also secure, you know, our finances too. That’s very much possible within the reach of today’s technology. Really exciting future.

You know, what we want to show you today is our agentic voice AI demo. This is demo of, you know, you know, as I shared with you, the technology has evolved through multiple layers of transition and capability.

There are some of our customers who are in the old version, some of the customers in the new version who joined us to talk about their experience. We’re gonna show you, you know, customers across, know, different credit unions, community banks across different level of maturity of their AI adult AI deployment today. Right? So from, partial generative AI to full generative AI to agent AI, we’re gonna take example of each Credit Union Community Bank in this in this level of adoption and show you the difference. Right? So I’m gonna, you know, bring in Elliot. He’s gonna start off with partial generative AI and then go into fully generative and then agent.

Right. So, Elliot?

So

the first demo, as Sri said, I’m gonna be doing on the partially generative AI.

So I’ve got a a real life sort of scenario that I’m gonna demo here where I’m in a car going to a dealership to buy another car. So we’re gonna call into Service First bot, and we’ll see how it handles my situation.

Thank you for calling Service First Federal Credit Union. Your call may be monitored or recorded for training and quality purposes.

Please note, Service First will not reach out to you via email, text, or phone call to request personal information.

Never provide your sensitive information, including but not limited to usernames or passwords to anyone who calls, emails, or texts. If you know the extension you are trying to reach or would like to use the staff directory, press the pound sign.

For further assistance, please dial zero or stay on the line.

Nine nine. Yep. Hi.

I’m I’m driving in my car to a dealership, and I was wondering if you could offer me a loan to close the deal today.

To make sure I understand your question correctly, did you mean any of the following? Or try asking your question again in another way. To utilize our dial by name feature or No.

I I don’t need any of that. I’m I’m just trying to to go to a dealership to to buy a car, and I just I need to get my loan closed today. Is that something you can help me with?

We are sorry to hear that you wanna close your account.

Let’s connect No.

No. That’s not what I asked. Could you could you just, tell me your your current vehicle loan rates?

Our rates are available at service one dot org under tools and resources.

You may also connect with a contact center agent I’m in my car at the moment, so I can’t go to your website.

Could you just tell me your nearest branch location instead?

I am currently on East fifth Street, and I’m heading towards Strawberry Avenue.

And that’s in Bloomsburg.

You can find a branch or ATM location by navigating to w w w dot service one dot one I’m in my car, so I I can’t really do that.

Alright. Thanks.

Okay. So the partially generative one, not a great experience in this situation. So the next

one we’re gonna try is the generative AIA.

Here. Sorry. An application error.

Maybe we’ll try it.

Hello. Thank you for calling Credit Union of Texas. For English, say English.

English.

Hello. My name is Sam. This call may be securely recorded and transcribed by us and our service providers for quality assurance and training purposes.

You can simply say things like check my balance.

Hi, Sam. I’m, driving in my car to a dealership, and I was wondering if you could offer me a loan to close the deal today.

Thank you for reaching out about auto loans while you’re at the dealership. Let me share information on Credit Union of Texas auto loan options to help you close your deal today. We offer fast, easy auto loans with competitive rates and flexible terms, perfect for closing your car deal at the dealership. You can apply online, get a quick decision, and even choose protection options like GAP and vehicle service contract.

That’s great. Could you tell me what your current vehicle loan rates are?

Let me share the current vehicle loan interest rates offered by Credit Union of Texas. Let me quickly verify your identity for security purposes. What is your member number?

I’m sorry. I I don’t have a membership with you. I’m I’m trying to get a new loan which would give me a membership with you. I just need to know what the loan rates are.

We offer very competitive auto loan rates and a fast, easy application process. Membership is required, but we welcome anyone in our community to join. You can apply online, visit a branch, schedule a Okay.

But can you just tell me what the loan rates are?

I’m sorry.

But I can only assist my own specific loan information and if I’m able to So I can’t visit your website.

So can you just help me get to, your nearest branch location? I’m currently on, Midtown Plaza in Kilgore, Texas.

Thank you for sharing your location. Let me provide information on Credit Union of Texas branches and services near Midtown Plaza in Kilgore, Texas. We have over twenty branch locations across DFW and East Texas, but there is not a branch listed directly in Kilgore. The closest branch to Midtown Plaza in Kilgore is in Gilmer at seven fifty five US Highway two seventy one offering full services and coin cashing machine.

I’m sorry. I Quickly Are you absolutely certain that there’s no branch in Kilgore, Texas? I could swear I’ve been to one there before.

Let me double check the available branch information for Kilgore. Texas to ensure accuracy regarding Credit Union of Texas locations. We do not have a branch listed in Kilgore, Texas.

The Credit Union of Texas has over twenty branch locations across Okay.

Great. Thanks.

So the generative one understood me a lot better, but it’s still not able to really assist with my issues here.

So now we’re

gonna try the fully agentic one.

Hi. I’m Joy, Navigator Credit Union’s digital assistant. If you prefer to have this conversation in Spanish, just speak to me in Spanish.

Hi, Joy.

May be recorded for quality assurance. How can I help you today?

Hi, Joy. I’m driving in my car to a dealership, and I was wondering if you could offer me a loan to close the deal today.

Please wait a moment.

Navigator Credit Union can help with auto loans for new and used cars, trucks, and SUVs, offering flexible terms up to seventy two months and rates starting as low as four point nine nine percent APR.

You can apply online or get prequalified at loans dot navigator c u dot org. If you’re a first time buyer, visiting a branch can give you extra support. Would you like me to help you start the application process?

So all that information sounds great.

I actually would like to go to a branch in person, and I’m currently driving in my car on Old Highway forty nine headed north towards Orange Grove Road. Could you tell me what the nearest branch is?

Please wait a moment.

The closest branch to Old Highway forty nine and Orange Grove Road is the Gulfport branch at eleven two eight one Highway forty nine in Gulfport.

They have drive up and deposit ATMs, safe deposit boxes, notary services, and more. I can text you the details for easy access. Standard message and data rates may apply.

Yeah. That’d be great. Could you text me the address information so I can pull up in Google Maps?

Please wait a moment.

I can text the Gulfport branch address to your mobile phone, but I can’t send it directly to Google Maps. You please provide a mobile number to receive the address via text?

No. It’s fine. Just text it to my phone, and and I’ll pull it up in Google Maps separately.

Give me just a second.

I sent the Gulfport branch address to your phone. What else can I help you with today?

Great. I have that. I’ll be in a few minutes. Thanks.

And so there you can see the Agetic AI. Even though I’m in the car, I get everything I need to go right into the branch and get the loan taken care of.

Wonderful. Thank you, Elliot. Appreciate it. All of the things you showed, you can actually go try it firsthand. If you go to Navigator Credit Union website, and you can call and experience firsthand.

You know, I have promised them to waive off this month, you know, invoice because we keep driving a lot more calls to them. So please try AI, but don’t use their agent time.

So thank you. But if you want loans, yes, please feel free to use their agent time. So so thank you for showing that, Elliot. Really appreciate it. Hope you saw the difference in partial generative AI, fully generative AI, and agentic AI. There is a day and night difference.

So, I’m just gonna go ahead and share my screen back here.

You know, we, we saw this in action.

You know, what we wanna, do now is kind of invite our visionaries who have taken, this step in launching industry first, agentic voice AI. Agentic voice AI, you know, doing this for a regulated industry like ours is extremely challenging. And, we spent two years of research to make this work, and, and we invited some of the alpha customers, and they were brave enough and trust trusted in our partnership to come in and launch this.

And they’ve been having a great experience. So I wanna invite Crystal and Tara to share their experience launching world’s first agentic AI for financial institution. Christa?

Thanks, Shree. Like Shree said, my name is Crystal Brown. I’m with Navigator Credit Union. I’m in the IT department. I should start by saying that Navigator did not set out to be the world’s first agentic AI in the financial services industry. We really just set out to solve a problem. We had an aging phone system, and we had an aging touch tone phone banking system.

And, both needed to go away. So we started looking for a phone system and and quickly realized we were gonna need something additional to fill that void for phone banking, and that’s where where Interface came in.

So we initially launched on v five, but we were in the right place at the right time, at the right point in our adoption journey to to make the leap to v six. So we launched on v five over the summer.

And the day after launch, I think we started working on v six.

We’ve been live on v six for probably about a month. We’ve we’ve managed to make it through hypercare, and we’re transitioning to more normal support. And it’s it’s become sort of business as usual here at at the credit union, and we’re so excited about it.

We named our bot Joy, as you guys heard. And so Joy has very quickly become one of the hardest working team members we have.

She’s available twenty four seven. She’s constantly working.

She allows self-service options that we never had before with our our old touch tone phone banking system.

You know, it’s not all rainbows. We’ve had some member adoption struggles that we’re working through. The interface team has been fantastic in helping us try to identify ways to increase adoption.

Typically, what we found is as long as we can get a member to actually try it, we’re we’re gonna win that member over. You know, our our adoption struggles are people that just aren’t willing to to give it a chance. So we’re making near constant tweaks to increase those numbers, and that does seem to be working.

You know, we went through partial generative and full generative and then the AI demos here a few minutes ago, and it really is just night and day difference on the back end as well. I mean, we we got to spend a little bit of time with the back end of v five.

And comparing that to to the credit union side, the back end of v six, it’s it’s phenomenal. The self-service options for the member are one thing, but self-service options for the credit union are are another thing entirely. And we’re able to make tweaks, push them to to UAT, have someone else test that, push them to production, you know, roll things back.

We really got a tremendous amount of credit union self-service capability with v six that is just kind of the icing on the cake. I don’t think we were expecting that, and it has been wonderful.

V six does sentiment scoring on the fly.

Joy rates herself on every call, and you can see it in the back end. You know, here’s why she scored herself low on this. Here’s what what it did. You know, there’s a as an overview of the calls.

If you’re trying to look and see what that call was about, why did it take half hour, you don’t have to read their transcript. You can tell everything Joy did. She just summarizes it for you. It’s it’s wonderful.

As we looked at sort of some of Shree’s slides, it talked about things we could do in the future and, you know, replacing contact centers and things like that. And when we didn’t set out to do that, we really feel like Joy is going to make our contact center so much more capable and give them so much time back in their day so that when they do take a call, it’s a call that someone truly needs help with, and they can spend they they have the time in their schedule to spend as much time as they need with that member and really solve the problem. Because for all those easy inquiries, Joy is very human like.

She can handle those. She can keep those members happy, and then we can focus on those those calls that need a little bit extra. You know? So we’ve been super excited.

Our next steps are gonna be integrating more vendors and and some more things through APIs with with Joy and increasing her capability and the member self-service capability.

And and we’re just I said it probably too many times on this slide, but we’re so excited with what Joy is gonna offer for our members.

On our project well, before I do that, so one of the things I tried to take notes during Shree’s poll that I had not seen the poll before. So people said that a human touch and then complexity losing the human touch and complexity were two of the things that really they were very concerned about. I think that was forty percent overall. In our experience, we didn’t lose a human touch because Joy is so human like, and she intelligently routes you, the member, when she needs to get you somewhere.

You’re not gonna sit there, wait on a contact center representative, and then have to have them transfer you to somebody else and just sort of be shuffled around. Joy does intelligent routing. She gets you where you need to go. She’ll ask you questions to identify that, and then she she presents that information to the agent if they’re reading the transcript.

And they they know what you need when they get you on the phone. You don’t have to repeat yourself. So we we have not lost that at all. And the complexity to set up AI, there is no scripting with AgenTik AI.

We didn’t have to do any of that. We didn’t have to do any mapping. We did a lot of testing. We did a lot of trying to break Joy and and see if we could, you know, lie to her or convince her to lie to us or bypass security.

That was kind of what our role was in the testing because and and in the setup, we didn’t have to do a traditional mapping anymore.

We had gone through that a little bit in v five, so we were very familiar with it at the time, and it was it was a much different product and project to jump from five to six.

So Tara Mitchell, who is also on the call with me, was our our project manager for our Interface project. So I’m gonna turn it over to her. She’s probably got way more details than I do, and she can kinda talk about what we went through. Tara?

Thank you, Crystal. Yes. The the Interface team was phenomenal to work with in getting us live.

I will duplicate or reiterate what Crystal said. The difference between v five and v six is night and day.

When they offered us v six, at first, I was like, oh my gosh. We’re gonna start over. We didn’t have to do that. Actually, the move to v six was actually to me easier than the v five project.

And that whole thing about losing that human touch, if you tell Joy or whatever you should choose to name your bot, my spouse recently passed away and they’re on my account. What do I need to do? She empathizes with you. She expresses condolences.

So it’s little things like that that really do still keep that human interaction in there. And as Crystal also said, it is allowing our representatives to truly dig in with those members when they do talk to them and assess needs.

We are participants in the member advantage, and it’s really letting our staff use those components of that program to really help them assess what our members need and dig a little deeper, where before when they were just having to answer the calls, they were almost more order takers. So it’s really exciting to see the growth and development with our staff that Joy is bringing in because it’s letting them have those deeper conversations with members.

Awesome. So so nice to hear about your experience, Tara. And, Crystal, thank you. You actually redefined the word human touch for me.

Like, you’re saying, if a human touch comes from AI, it’s still a human touch. Right? Like, is someone able to empathize things like that, and then get out of the way when they can help and bring someone who is more of an expert. That’s fascinating to hear.

Thanks a lot for sharing your journey with it’s been amazing to work with the team as well.

We my team loves you guys there. So thank you for all the support. And, you know, I

wanna introduce Josh, who is senior vice president of IT at Red Rocks.

They’re also, you know, launching the world’s first agent in voice AI here. Josh, we would love to hear from you.

Yeah. Thank you very much.

Glad to be here and, you know, share our experience and talk about, you know, our goals and vision for AI and why this is, being such a great fit for us. So we’re currently wrapping up, testing. We’re planning to go live, very soon with version six, which is the agentic AI.

Like navigator, we actually started our implementation on the previous version, when we got offered this opportunity.

You know, the this aligns with the experience we wanted. You know, we’ve I’d sat through several of, Interfaces webinars before, and we’ve seen their, vision for the product. And that aligns closely with our our internal vision for AI, which is, you know, AI is is a tool, right, that enables our staff and our members. You know, it it drives personalization, automation, and real time usage of of data and information. And, you know, this implementation is important for us, even though it was, you know, something new. Right? A member facing agentic AI application.

Because we believe that building AI maturity inside of the credit union is is essential for our long term success. It’s, you know, it’s a it’s a priority because, you know, it’s it is you know, as Shree mentioned, it’s the fastest growing technology we’ve we’ve seen and how it’s advancing. And the vision for this product and the implementation and the improvements that we see coming, you know, are are are a great start for us in in, you know, demonstrating that to our members, getting some comfort, so that we can really responsibly roll out, AI initiatives across our organization. That’s one thing that’s been a focus of ours is, we started with a a framework, a governance framework that we developed with our audit and risk committee. We outlined our intents of being you know, of of deploying, person centric AI tools, which is what we believe this is.

You know, it enhances the it enhances the experience of our members, of our staff. It provides instant knowledge.

And, you know, in the short term, this tool is providing some new functionality for us. So, you know, we don’t currently have a phone banking offer, and now we have a agentic phone banking offer that, you know, twenty four seven can post transactions, you know, move money inside of our members’ accounts. It can, you know, tell them about what’s happened on their account.

The experience has been really strong, you know, as we’ve gone through the testing, and and and we’re very pleased. We’re excited to, you know, roll this out shortly here. We’re just kind of on the you know, about to cross the finish line in getting that rolled out. So as far as, like, you know, the the the goals and what we think this will achieve, right, there’s the twenty four hour self-service banking, that it’s gonna give our members, which we don’t cover right now.

It’ll also reduce friction. And I think one of the key things, you know, to keep in mind about AI is if we can have it handle some of the, self-service items, the low hanging fruit, it really frees up time for our staff to have those meaningful relationship building interactions with our members. Right? That’s where we want our staff time focus is on the relationship development, figuring out what solutions we can do to help benefit our our members’ financial journey.

And, yeah, we’re our our whole team is very excited to roll this out, and that’s kind of where we’re at in our journey.

Wonderful. Thanks a lot, Josh. It means a lot to us to share your journey, and thank you for the partnership. I I truly believe the right partnership is where the partner’s roadmap drives your business, not your contribution and effort drives their business.

Right? So I I I I I am thankful that you see that. Right? Like, we we are aggressively pushing a road map and moving fast that you could see that it can drive your business and drive your goals and improving members’ financial well-being. Thank you for joining and sharing that with us. It means a lot to us.

So, you know, with that, we want to show you demo of how AgenTiC AI is going to transform or replace the current form of contact center software.

Crystal, I didn’t mean that AgentDK will replace the entire contact center staff. I meant AgentDK will replace the current contact center system, the software.

That’s what I meant. I am agree I agree with you. We can’t replace that. So we’re just talking about just the software part, which is the Genesys, Ciscos of the world or, like, you know, all the Avaya’s of the world. The way the contact center software works, Agent Digay is going to fully redefine that.

We are gonna show you a demo. I’m really excited because if fifty, sixty percent of the calls so an AI like Joy can automate, Then for the other fifty, sixty percent of calls that goes to staff, how can AI partner with the staff to seamlessly resolve the remaining four fifth forty, fifty percent of them so that AI adds value across a hundred percent of member interaction. Right? So, you

know, I’m gonna invite my colleague here to help us with the demo.

So the future of contact center system is gonna be very, very different. Right? So let’s take a minute here to unpack.

So what happens in a contact center system contact center system today? If some a member calls for doing a wire transfer, know, let’s say for whatever reason, if the credit union uses, you know, the, you know, the they uses the AI and AI is unable to do that and the AI transfers a call to a staff, What does our staff do? A staff, you know, pulls up a system, you know, tries to figure out how to start the wire transfer, and they’re probably ten steps. They go through all the ten steps to do it. Right?

But the future of contact center system is very different. As you can see on the screen, the agent, the human agent could come in and wait for them to task to be assigned.

And wherever AI gets stuck in the wire transfer, it’ll bring the human agent live. Right? So AI is already out of the ten steps. AI is probably already doing eight or nine steps.

Right? And for let’s say, for one step, it got stuck.

It in real time, it brings your staff into the call to unblock it.

So they don’t have to do ten steps. They have to do the only one step where AI got stuck. It turns out here, the AI got stuck because the probably the session got expired.

The the your staff can take control quickly and then log in and unblock the AI. And the AI will continue and comes back says, you know what? I took care of all the steps. Here is the system, FI system or something, whatever you want to use. It’s right there. I opened the right screen for you if you wanna use it.

But here is a summary on the right side. I’ve I’ve did all of this already.

Do you want to authorize this wire? It shows the entire transcript that has happened with the AI and the member until that point, and then comes here and says, do you wanna authorize this?

Your staff has to click a button. AI has done everything already. They click a button. It’s done.

AI sends them next task. Here, it seems like a member or a customer on the call has requested a callback, and they wanted to learn all their application status. For whatever reason, AI was not able to help. And AI said, okay, I’ll have someone call you. And then as soon as your staff become available, they were assigned with this task. They can click a button where they have this full summary.

They can click a button says call customer.

It’ll call them. They can have the conversation right here. On the left side, AI has pulled up the right software if they want to take control and navigate and take action. They don’t have to go do across ten different I mean, three different screens, ten different screens or applications.

AI brings the right application, opens the right screen, and keeps it ready for them. It seems like they’re done with the call. They’re able to get the application status. Then they sit back, relax.

AI sends them the next task. It seems like AI got stuck here in trying to increase the credit card limit.

Right? AI said, hey, you know what? I was able to do all of this. You know, I need their income statement is too old. So, you know, the your staff got involved and, click that button, said request latest income.

Because the member is saying it is already on the email of your staff. They already sent it. So you tell AI, give access for it to scan your emails and find that, statement and is able to take that up and say, you know what? I got everything now. Just click approve.

So then your staff come in, approve.

The request is done. So your member wouldn’t ever know that a human was involved behind the scene. They’re just talking to AI all along. So now AI sends them the next task.

Right? So, like, that’s kind of the future of, you know, call center software. It’s a human and AI working seamlessly at every step of the way and where human getting involved wherever they’re absolutely needed. Right?

So, you know, that’s pretty exciting future. You know, if you look at twenty years back, one staff could do one call at a time. And then ten years back, live chat systems came. And now one staff could do four to five chats at a time.

And now with agentic contact center system, one staff could be handling thirty, forty different calls and chats at a time because AI is doing eighty, ninety percent of them, And then your staff coming in wherever AI gets stuck to unblock it. Right? So that’s the future of contact center system that we see through agentic AI. I know we are coming close to end of this webinar.

We had a lot more to show you, and we plan to have a part two of this webinar with the amount of demand we’ve seen to make sure we’re able to see agentic online mobile banking system.

Imagine if AI if someone built an online mobile banking system today using all the advances of AI, what would it look like? What it can do? So we’ll be exciting to show that next time given we just got five minutes. So so thank you for your participation today. So couple of things we want to share with you before we end the presentation. Of course, we wanna take up some questions that are still not answered there. Give me one second.

One thing we wanted to share with you on this webinar is, I know you’re on the journey of figuring out the AI and the road map and everything. What we wanna offer you is if you drop us a line at sales at interface dot a I, you know, we would actually turn around and get you a custom agent AI in a couple of days so that you could actually use the AI like Joy and show it to your board and leadership and get the journey started. So it will know everything about your organization that’s publicly available.

So we should be able to get that set up to you in a couple of days. Just drop us a line. We’ll give you a custom agentic AI just for your financial institution. That, like, you you can experience it firsthand.

So I appreciate, everyone joining today. We’d, take a couple of questions here.

I see a lot of questions are already answered.

If there are any other questions, feel free to please, you know, put them on the q and a.

See about there were about ten questions. Those are all answered from the team. Hopefully, that was helpful, but we’re happy. Thank you for the feedback.

Some of you are saying great content. Thank you. It means a lot to us. We’ll do a part two very soon.

Let’s see the, agentic AI, how it transforms online world banking, and take us towards autonomous banking, very near in the future. Thank you all.