Learn How Dupaco Sets the Standards for Fraud Prevention and Personalized Member Experiences with AI

Welcome, everyone.

My name is Chris Bechtel. I’m the VP of marketing here at Interface AI, and I’m gonna provide a brief intro to today’s session. We have a whole lot of great content for you. So today’s topic, we are absolutely thrilled you you could join us to explore some of the most important topic shaping today’s financial services. As I’m sure you all know, the landscape today is constantly evolving. Fraudsters are getting smarter. Member expectations are higher than ever, and institutions need solutions that keep pace with all of these developments.

We are extremely excited to showcase how Dupaco Community Credit Union has become an industry leader in both combating fraud and delivering exceptional personalized member experiences, all with the power of Interface AI solutions. We’ll be discussing in detail how they made the journey from IVR to the latest advancements in voice AI built specifically for banking and credit unions.

On our panel today, we’re joined by three leaders in the financial services and credit union space beginning with, of course, Todd Link, an esteemed leader and industry veteran. Todd’s the chief risk officer at DuPaco Community Credit Union. With more than eleven years of tenure, he also serves on coop strategic cocreation council. He has been with DuPaco since twenty fourteen and has been a driving force behind DuPaco’s AI and fraud prevention implementation.

We’re super grateful to have Todd with us. Thank you so much for joining.

And of course, Charlie Peterson. He is the VP of strategic initiatives at Allied Solutions, which provides more than six thousand clients leading solutions across North America. Charlie has been at Allied for almost forty years. He looks too young for that, and comes with us, comes with a wealth of financial services and risk management experience. And finally, we have Srini Vasanjay, the CEO and cofounder of Interface AI with us. Sri hails from a credit union family where he helped finance helped the financial institution really scale from inception to several billion dollars in assets. He has over a decade of experience in artificial intelligence, including spearheading AI initiatives during his time at Microsoft.

Sri started Interface AI with a mission to really revolutionize the financial industry through AI. Recently, Sri was prominently featured on the cover of Forbes magazine just last year for steering interface’s remarkable growth. Under his guidance, we have achieved significant milestone of becoming the most valuable agentic AI company in financial services. This is one of those hot buzzwords you’ve heard, agentic AI.

And we are really leading at the forefront of that. And to help us get a better standing, really a better understanding of your current and future plans, I did wanna run a quick poll because there is a whole lot of content that we have today. Here

is our agenda.

We have a whole panel discussion. We’ll be talking about the evolution of AI. We’ll have a product demo so you can see exactly what we’re talking about. It’s one of the best ways, of course, for you to learn. And, of course, we’re gonna go through a detailed case study on what we call the three legged stool of how we’re helping improve member experience and reduce fraud.

But real quickly, just to get get a sense of where each of you are on your journey, today, I’m gonna run a quick poll here. You’ll be able to see this on your screen.

And if you can please take a moment to complete, the poll, I’ll give you all a moment to do your best. This is multiple choice, so you can choose more than one. But we’re, you know, curious to see, you know, what projects is your FI actively considering.

And this way, it’ll help us sort of tune the some of the content to today’s webinar towards your interest areas, and allow us to give you some better insights and knowledge.

I’ll give you all just a couple minutes more, to fill it out. Really appreciate everybody’s time. It’s super insightful.

Looks like the largest category here, of projects is AI assist augmentation for member service representatives, followed by voice and or device biometrics, are the two biggest interests.

So thanks everyone. That’s fantastic.

Excellent.

Well, I appreciate all of your input and feedback. And

the final component is really just as a reminder about Interface AI established in twenty fifteen. Really, the key component of interfaces at our sole focus is financial institutions.

For almost ten years, we’ve been building conversational experiences for credit unions and community banks exclusively. Today, we’re trusted by more than a hundred financial institutions to power better member experiences. That means sixteen million consumers, one and a half million conversations every day, recognized by numerous organizations for our leadership, including Gartner, CUNA, of course, Allied and twenty three state credit union leagues. We also won an award at Finovate. We are really an AI first company. So, again, thank you all for joining us. I look forward to sharing more of this content with all of you.

Sri, take it away.

Thank you, Chris. Really appreciate it.

You know, there is so much to talk today. Let’s, get started. I’m really excited to, share a couple of, big updates and, and also, you know, go through Dupaco’s journey. It’s been an amazing journey. You know, five years ago, actually about six years ago now, I wrote a blog titled the new enterprise interface, highlighting how the way we interact with computers, which had remained unchanged for decades, was about to undergo a radical transformation.

In the blog, I highlighted how over the last twenty, thirty years, we have been learning how to work with computers.

But the next twenty, thirty years will be about, you know, computers learning how to work with us. Right? I I predicted that the current methods of how as humans we interact with machines would cease to exist, paving the way for the new enterprise interface powered by AI. Fast forward six years, Satya Nadella, this Microsoft CEO declared SaaS is dead.

Whether you see the glass half full or half empty, the meaning is same. The era of AI driven transformation is here. Having this conviction early on as a company allowed us to build a pioneer pioneering AI company, putting us on the map as a leading force in AI for banking. Today, I’m thrilled to share where this vision is taking us, more importantly, how it will create more value for you and your members and customers.

Before we deep dive on our vision, it is important to understand how AI as a technology is evolving. Right?

So, pretty much this is kind of how we see, you know, kind of trying to simplify all the things going on in AI, you know, past, present, and future. The past has been what we call conversational AI.

These are chatbots, you know, pretty much, you know, you know, little bit, you know, kind of fixed in how they work, not too flexible. It takes a while to implement it, lot of manual efforts and things like that. That’s kind of what we call it the, you know, conversational AI, which is of the past. And now present is generative AI.

These are, AI technology that have ability to seamlessly understand how humans are speaking, generate very cohesive human like responses, you know, without manual efforts that had to be happened in a conversational era. Generative AI makes it even easy to implement some of these technology.

The whole experience of the user becomes very fluid, you know, and kind of, you know, the overall value and automation of, you know, incoming calls and chats continue to increase. Right? So the member experience becomes significantly better with generative AI. That’s kind of where the technologies with kind of the present.

And the future where it’s heading is agentic AI. Agentic AI is, you know, is is is gonna drive what we call autonomous agents. Like, where it has ability to navigate tools, in in in the computer, the tools and softwares and system just like a human would do. Right?

You know, navigate those systems to complete a task where you can give a goal to the AI. It’ll take probably five, ten, fifteen minutes for complicated ones or a simple task could be just a couple of seconds. And it will actually, you know, navigate systems like a human would do and complete the task on behalf of the request requesting person. That’s the future autonomous agents where the technology itself is heading. And understanding how this technology is evolving is really important as we dig into our vision because very aligned, with the technology evolution. Right? Our vision, again, the past has been, through conversational AI.

We’ve been embedding AI in all service channels within a bank or a credit union. You had online mobile banking, you had website, you have text, solutions, your contact center IVR, our AI has been getting embedded into these channels working as a first responder. Right?

And as the generative AI era is, coming upon us, you know, we we are moving into the second phase of this journey, in in our, product and vision, which is what we call unified. Where the AI with the power of generative AI is capable of, handling more and more service requests that are coming in. You know, you know, in the in the past era of conversational AI, our customers have anywhere achieved between thirty, forty, fifty percent automation with the with the with the present, which is generative AI, their automation of incoming calls and chat using AI would go up to, like, sixty, seventy, eighty percent. Right? So and, this unified era, you know, where it’s not only about improved automation, but the whole member experience becomes very fluid, very flexible. It’s almost feels like talking to a human being.

And then these systems will continue to, evolve to even enhance those interactions, not just about task completion, but also driving upsell cross sell, providing financial insights, and driving kind of to some degree, financial wellness, in in this era of unified.

And in as you exit this era and move into the future, which we’ll talk about in a bit, we see some of the channels you invested in, like contact center, online mobile banking, they will significantly diminish in the role they play in banking. Right? So to an extent, you would be replacing, your online mobile banking to a, AI agent. That that’s a possible as we reach the end of this unified era. And the future, what we call autonomous, this is kind of a whole next level. With the agentic AI evolution in the future, this is where, that technology could be applied to create more value for financial institutions.

And what what we see the future is what we call autonomous banking, which is more like a bank GPT, where your members and customers can add multiple banks and credit unions they’re banking with, including, non banking mortgages or auto loan brokerage, firms, accounts in the same AI agent and transact across all of them through one AI system. Right? And it has a built autonomy, drive financial wellness goals. You say, hey, ten years I wanna save, college fee for my kids.

It has ability to reorganize your finances and save and drive you up to get there, in an autonomous way. This, you know, interface AI vision of these three phases is very well aligned with the technology evolution. And we’re proud to say some of the things that we imagine when we create this company in six years back is coming to fruition now with evolution technology. So and these systems, even today, we have number of banking integrations.

Many we couldn’t be listing on this one slide, but may many of the major systems we already have integrations out of the box, either you’re looking at integration of different core systems, online banking systems, a loan origination systems. We have all those integrations ready to go on day one. And with that, I would like to take few moments to show a demonstration of a product so that, you know, you understand kind of the products that Dupaco has already launched, and we wanna demonstrate that. And then when we go into case study, you’re able to resonate better.

One

of the products Dupaco has launched is VoiceAI, which is basically an AI powered phone banking, where it receives hundred percent of the incoming calls and has ability to triage the call, understand the question, validate the caller for the fraud, and then fulfill the request. If not, hand off the call to, do a warm hand off to call to an agent. Right? It does all of this.

Right? And we’re gonna show a quick demo of the same. And this, module, which is like voice AI, it comes in two different kind of a crawl, walk, and run kind of package.

You know, stage one is where the AI answers informational transactional questions, which is nothing but in front of the pin behind the pin questions.

And, you know, with an add on as part of stage two, you can evolve this, AI Voice AI to be able to do upsell process during the conversation.

Thereby, you have an opportunity with the Voice AI to transform a contact center from being a cost center to more of a revenue center. And that’s the direction AI could take this. And so let’s say a quick demo. I’m gonna invite my colleague, Harry here, to, dial up Jessica and show us a demo of, how a member interaction or con consumer interaction would work with an AI.

Hari, could you help with that? Yep. Thank you, Sri. Let me quickly share my screen.

Okay.

So I’m gonna make the call here.

Hi. I’m Jessica, interface digital assistant. Please wait while I read the instructions. This call may be recorded for quality assurance.

For best experience, avoid using speaker mode.

How may I help you today?

Hi, Jessica. I’m driving around, and I need to make a deposit. Are there any branches open?

Yes. We are open. Our Main Street branch only has drive through capability today.

Do you want me to send you driving directions to the nearest branch?

Yes. That would be great.

Great. Please provide me the ZIP code or city name.

I’m in Raleigh, North Carolina.

Thanks.

Looks like there are a few branches that are open in your location.

The branches are open from nine AM to five PM, Monday to Friday, and ten AM to two PM on Saturday.

I just sent you a text with driving directions, and you can always talk to me whenever you need help. I’m available twenty four by seven.

Is there anything else I can help you with?

Yeah. I just started a new business, and it’s not doing well. I don’t think I’m gonna make payroll. Can you help me skip a loan payment?

I can surely help you with skipping a payment. I have sent you a text with the link to the application to skip a payment along with the instructions.

Please click on the link and follow the instructions.

Please note that the application should be filled at least five days before the payment due date.

To assist small businesses like yours, we have removed the thirty five dollar fee associated with skipping a payment. Do you wanna know which loans are eligible to skip a payment?

No. But, can you tell me the balance in my checking account?

Sure. I will help you with your account balance.

I see that you already have registered for device biometric.

Would you like to verify using your device?

Yes. Let’s verify using my device.

Great. I’ve sent you a link to your phone. Click on it so I can verify your biometrics.

Thank you. I verified your device.

The balance in your checking account ending with two two three four is two thousand three hundred and thirty seven dollars and fifty five cents.

Also, I want to remind you that if you maintain a minimum balance of three thousand dollars, you will be eligible for an additional one percent cashback.

Do you wanna know the balance of your savings account?

No. But, I’m traveling to South Africa next week. Can you put a travel notice on my account I mean, my card?

I can help you set a travel notice. I’ve noted that you’re traveling, and I have successfully set a travel notice on your card. We’ll make sure your transactions run smoothly. By the way, you’re traveling a lot, and I recommend you switch to our new credit card with no foreign transaction fee and many more benefits. Do you wanna know more?

No. My wife is looking to buy a car. Can you help me with car insurance?

Sure. Happy to do that. Give me a moment. I will transfer the call to our specialist, John.

That’s great, Harry. Thank you for that. And what you’re seeing on the screen that Harry is showing is once the you know, there’s a hundred percent calls are picked up by this voice AI assistant. Right? And, you know, fifty, sixty percent of calls, it has ability to automate. As you saw, many questions it answered.

And it has ability to prevent fraud by authenticating many ways. We have a whole session on that. We’ll talk in-depth. And then when it couldn’t answer a question, seamlessly handed the call to an agent.

Right? What you’re seeing on the screen is how AI is augmenting the agent. So Harry is acting an agent now. He has a call that listed there that has come in.

He can click open the call, and immediately, he has a screen called Nexus.

As an agent, Harry is able to see everything that happened with authentication of the call on the right pane. And in the middle, he can see the conversation history. On on the left, he has a AI that’s already looking and to find an answer to the question that a the AI couldn’t answer before transferring. So Harry, as an agent, now can see all of this is in one screen.

And if the member on the call ask a follow-up question, that is, you know, that has to do something with the behind the pin, you know, all as an agent, Ari has to do is type in a question and, AI has to confirm, if that’s what, as an agent, Arias wants to do. And if he clicks yes, the AI automatically connects with the x, the back end systems, the core systems to transact and get the transaction completed. So so agent doesn’t need to have fifteen different applications open across three monitors, often jump between them. So the AI is able to get things done.

Right? So, again, as a member on the call or a customer on the call, let’s say, if they ask more question to the agent, you know, hey, I’m looking to buy, you know, buy, you know, get a credit card or something. It is now coaching agent how to sell the credit card, giving tips and suggestions and valuable points that they could, you know, use to, you know, convince a member to get a new credit card. Now let’s say if the member is satisfied, they click, you know, agent has a way to say satisfied, and the agent screen automatically connects to the online application, enables the agent to start filling up an application form.

Right? Again, it’s all happening in one screen as member is taking the agent through multiple different questions. The the agent never have to, you know, go across multiple screens and different applications.

So, Harish, as an agent now filling up all the details on behalf of the member and, you know, it used him a a short confirmation to look at all the details to confirm.

And then at the end of it, it even suggests, hey. As an agent, you know, this is the best time for you to probably do an upsell of money market account. Right? So, you know, as you noticed, AI has ability to, handle in hundred percent incoming calls, automate, automate some of those calls, even upsell cross it during a call, or when, or or if it cannot, then the AI transfers the call to an agent where it has ability to, argument with the agent, assist the agent, transact and everything on behalf of the member, also enable and coach an agent or do upsell cross sell.

So that’s kind of kind of the end to end view there. So I appreciate Harish showing that demo. Thank you. So, you know, pretty much you you looked at, like, what we call a risk based authentication system.

Right? AI is continuously looking at every single question coming in the call, identifying whether it’s a low risk, high risk, or, you know, no risk questions, every question being asked. And then automatically determines whether to apply multiple different factors of authentication. Either it’s a modern authentication, which is device biometrics, voice authentication, which we’re gonna talk a lot about, you know, how do Pago leverage it, or it could be traditional authentication, which is in wallet, out of the wallet, or you could even use hybrid authentication mechanism to prevent fraud.

And this is kind of the warm handoff which, we saw this already. With

that, I would like to invite, Todd to jump in. You know, we’ll kind of begin talking about their journey with AI. And, to start off, I would love to hear from you, Todd. Could you please tell us a little bit about Dupaco, Community Credit Union?

Yes, Sri. And thank you, for the opportunity to, join this webinar today and welcome all of our, friends and guests on on this webinar.

I am the chief risk officer at Tobacco Community Credit Union, and we are about a three point four billion in assets. And we serve approximately a hundred and seventy thousand members across, three states, in the tristate area, Illinois, Wisconsin, and Iowa.

So, our footprint spans, multiple areas, and, we have, found, a great partnership and an opportunity to partner with, Interface AI to help serve those members in a in a diverse, way.

Awesome. Thank you, Todd. And and, pretty much, Dupaco today has launched the voice AI, which we saw saw in the demo. Right?

Like, you know, and of course, you can try talking to Dupaco’s Ray here in this number. What we show is is a is a demo study similar to that. But, Ray was launched in August twenty twenty three. Right, Todd?

And, you also followed that with, now enabling Ray to speak Spanish, recently. Right? So that’s October twenty twenty four.

And then you also launched, modern authentication, which is biometric and caller ID four and six in Jan twenty twenty four.

Right? So you Yeah. You know, specifically, you call that a three legs too.

Right? So, and this is your brainchild. And as interface, we partner with you to build this, and it was your idea. And can you take us through that journey a little bit?

Yeah. And and I think a lot of, folks on on this webinar can really resonate with this story.

I recall, prior to my introduction to, Shree, we had a a call going on in the contact center, and I was ushered down as the chief risk officer and security officer. Hey. Come down, and and you gotta you gotta listen to this overhear this conversation.

This person has been on the phone for the last fifteen minutes, and we cannot come up with a question that they can’t answer. However, they’re mispronouncing nearly every restaurant.

They’re mispronouncing the community I’m from, which is Dubuque, Iowa. So they were we could clearly tell that they were a fraudster, but they had absorbed the identity of our member, and there literally was nothing. I, in fact, jumped into the situation. I started asking, a variety of deep questions. I went back multiple statements, and and these people were really, really good.

That led me to the realization that in today’s future, fraud is twenty four seven. It never sleeps. How are we gonna combat it? It’s gonna take a multitude of things. It’s gonna take we gotta know who we’re talking to on the people on the phone that’s call coming in, and we can do that through ID forensics.

We need to ascertain the identity of the individual, and we can do that through biometrics.

And then we need to automate where we can.

So not necessarily just to cut staff, but to allow us to serve that low hanging fruit. What’s my balance? What are your hours? What’s your routing number? All of that stuff, we can move off our plates so we can serve members at a much, much deeper level for those deeper interactions that they’re seeking. So that’s where this three piece, legs or stool had originated from, and and I think it has much efficacy today as it did several years ago when we started putting it together.

Wonderful. That’s really helpful, Todd. You know, the you know, there is there is the biometric part of it and caller ID forensics kind of helps with combating fraud. Right? And and and then the call automation helps with efficiency.

We wanna start off talking about the fraud part first. Right? So and I would love to know the journey in a little bit more in detail.

You know, why don’t we get through some of these questions and understand the journey, Todd. So tell us, you already spoke about the problem. What led you here, in in terms of looking for a solution like this?

We’ll we’ll if there’s anything else you wanna share, I would appreciate you kind of, coming in more on the No.

So when we were you know, the problems, obviously, we’re we’re we’re trying to stop fraud. We’re we’re we’re looking for a a partner that that has the resources and tools and things like that. But then how do we solve it is the real key. And what we were looking for is a number of factors. Number one, we wanted somebody that was committed to the financial services industry.

If you go out and Google AI, you can literally come up with five hundred or a thousand different companies out there that are doing AI. However, they’re not specialized in financial AI. Big, big difference. The learning curve is exceptionally steep.

We wanted a partner that could come on day one with fifty, hundred, hundred and fifty built out, ready to deploy journeys that we could put members into, to help them right away.

We were looking for somebody that was committed to in reinvesting in the technology.

We were looking for individuals that had the resources to invest, and then, of course, the expertise with the AI and staying at the forefront of that of that AI.

And then, ultimately, the big two pieces were, yeah, a lot of folks do AI. A few folks do AI for FI channel, but nobody wraps around the forensic ID and the biometric around that entire transaction to really stop and really meaningfully impact the ability to control fraud losses through this channel. So that’s what really led us to the solution that we’re on today, Shree.

Yeah. No. Appreciate it, Star. Like, Doug, couple of things. Right? Like, you you did evaluate, like, other options like, you know, pin drop and who can do, you know, specific to caller ID forensics.

Right? So and, you know, and and then, for automation, other other solutions that could just do automation, it seems like the the kind of three legs to come together. And, you know, specifically, we even we didn’t offer that. But when we met you and where we saw the opportunity combining these in a seamless way, the the, you know, the reason we saw that is these solutions are not even though they’re available independently, by other providers, but they all need to work together for it to have a meaningful impact.

Like, you know, and because if the biometric authentication fails, you have caller ID forensics that, you know, that can work as a backup because there is a world of deepfakes. You can’t just go on biometric alone. You need to have kind of a seamless multiple factors tied in to create that, like, confidence. Right?

So, which is which is the caller deforensics. Right?

Absolutely, Sri. And and I think all of us on the webinar can relate that examiners today are looking for more and more layered approaches to fraud and security, layered approaches to fraud and security, creating that mosaic, if you will. What tools and techniques can you stack on top of one another that interact with one another? So the biometric, if that if that triggers a score, well, that that feeds over into into, the automation. If if we have the deep forensics and you know, if you think about running a hundred different checks on every single caller to the contact center, and it’s completed before before anybody even picks up the phone, that is powerful.

That is powerful.

To to make sure that these these phone calls aren’t originating in Nigeria and some of these types of things. And the a lot of this is proprietary that I’m I’m not gonna really get into all of the checks and balances they do, but it is it is really, really impressive stuff. And when you bring that all together and create that mosaic on top of one another, that’s where you begin to see that real value to your point, Sri.

Yeah. No. Appreciate sharing that. On on the partner selection, specific to this solution, right, like the member experience is, literally as soon as a member calls and within five seconds, they got multifactor authenticated.

Like, you know, they’re saying, my they’re we’re using voice. And then behind the scene, there is this hundreds of checks happening, based on the network forensics and the caller ID forensics. So that all happens within five seconds, you know, and imagine the, amount of time today it takes for verifying, a member and the kind of a painful process we all go through. Right?

Like, you know, several minutes it takes any financial institution to do that. And, you know, and you already addressed that, like, in terms of, you picked us as a partner given that it is all the three legs tool coming together. And and, you know, we were very fortunate to have Tod’s partnership.

You know, being dedicated to this industry, we only serve bank’s securities, that that’s all we do. And we wanna further invest all the time to improving, you know, value for the industry. And and this is kind of our commitment to work with Tod to create this, you know, solution. And now many of you on the call as customers and prospect can leverage that in a big way. And, Todd, would you like to share some, couple of, you know, thoughts about the implementation journey of launching this fraud solution?

Yeah. I mean, so so, you know, it it really just kinda came down to what were what were we after. We wanted to offer our members we wanted to offer our members better service twenty four seven because right now, we have twenty percent of our membership works second or third shift. Mhmm. So twenty four seven availability is really, really important because we’re not at the at the asset size yet where we can really, afford to staff a a contact center twenty four seven. So that was important. We recognize that twelve percent of our membership, would prefer to communicate in their native language, which was Spanish.

So we learned that. We learned that, you know, we have dozens of members, probably a hundred members, that call in over a dozen times a day to check their balances.

So if we can automate some of those calls and if we can gain efficiencies from some of those calls, it gives our team that opportunity when somebody does come in and they have a pressing need that present AI cannot solve for, but I’m pretty confident future will.

We will have the staff and the team and the time available to go deep with that member and really deepen that relationship and end with a much more satisfied member experience for that, interaction that really requires more in-depth, conversation.

Yeah. And and it’s a journey. Right? So, in a crawl, walk, and run. And initially, when, we launched for Tobacco, the biometric authentication, you know, it is a industry wide known fact.

You know, the the adoption, was low to start with. Right? Like, you know, it because the AI was self enrolling. Right?

So it was going through a sequence of questions after answering your question. It was offering an option within the call. Hey. You wanna enroll for biometrics?

It’s gonna be better next time. And lot not a lot of people, chose to do that. And, Todd, we we kind of launched, together with you, like, an agent led enrollment. Right?

Which is now enabling your agents who are handling their call to hand hold the members, to kind of, you know, enroll people into biometrics and things like that. And that actually helped us push that option further.

Right?

So and I want to show the impact and and, Todd, if you would like to kind of speak to I do.

You know, before we go into adoption, was is there any member feedback you’ve gotten you could talk about that?

And Well, you know yes.

So so we, right now, we serve every time zone in the United States, and we serve twenty eight different foreign countries.

So having that ability, to serve these individuals in a variety of ways has been well received.

Having the ability on the nights and weekends to serve individuals has been well received. And then when we have those conversations with members, sometimes, you know, they’ll ask, you know, hey. What’s up with this automation, or why are you moving to this? And when we take the time to explain that, we’re not so much trying to only drive efficiency as much as we’re trying to create more meaningful opportunities for engagement.

And did you know that when you call in, a hundred checks on your on your phone has been run before we even pick up the call to prevent an account takeover and to prevent you from being a victim of fraud.

When they start to understand that, then the light bulb moment starts to go off and they’re like, I see these individuals are working in my interest. I see that they’re trying to put the latest and greatest tools in place to help prevent fraud. One of the comments Sri made, to me early on in this is, you know and and it’s really resonated with me. I think we’re in in the golden age of of AI. Right? I mean, this is we are we are changing faster than at any point in in probably human history. AI is probably one of the greatest inventions ever done.

But Sri assured me that if you think about the sophistication of what the fraudsters are doing today, the only way that we’re gonna compete with them and catch them is to combat it with AI itself.

We have to use AI itself to go right after the fraudsters trying to do the exact same thing. Because right now, they are so good, there’s not a chance that we could defeat them without the use of this AI tool. So Yeah. I think it’s I think it really lends itself to the power of the product.

Yeah. And the combination of a voice biometrics, which is kind of AI driven, as well as the caller ID four n six, and and and the agent led enrollment helped you kind of achieve good adoption. Right? So and and Tobacco today is at about sixty eight percent, adoption.

Right? So, and and a huge kudos to your your team, some amazing team we get to work with, right, like Steve who have been able to continuously create APIs and integrate help with the integration is such a big team effort. And and Tony and team have been very judiciously looking at conversation, optimizing the AI, for better performance and and creating incentives. Even when you created incentives for agents to drive agent led biometric enrollment and things like that.

So the combination of that got us so far. So that’s, you know, great achievement there.

I I I wanna kind of talk about, you know, the best practices start. There are many as you saw on the poll, there are many in the journey of, you know, starting their AI projects or in the middle of it.

Are there any, you know, suggestions, in terms of best practices for implementing such a fraud solution?

Yeah. I think, Sri, through the process, I’m gonna throw out a handful here for folks that might be looking on the sideline to either deploy or they’ve had they have deployed and they’re and they’re looking to enhance.

You know, I think it’s critical that your organizational strategy and your organizational strategic plan matches what you’re doing with this AI.

The two have to come together, and it has to become part of what you are trying to accomplish as an organization.

The other thing is, you know, selling the benefits and promoting the benefits to the membership from the get go. So individuals don’t view the AI as a as necessarily only a cost savings or a a staff reduction type tool. In in our case, it had as much to do with being able to deliver twenty four seven service, you know, Spanish, prevent fraud, a lot of other things besides any efficiencies.

So it’s important to get that out there and then keep things simple for the the member. We’re continuing to add journeys, and I think that’s what’s critical in a partnership.

When you look at a partnership, are they committed only to financial institutions? If the answer is yes, okay. What kind of journeys do they offer, and what kind of journeys can they put into place on day one? Finally, look for a partner that is committed to and is lobbying in your best interest. In other words, their pricing is contingent upon your continued growth and use of the project or product.

And we found that through, our association with Interface where, you know, as our, usage continues to ramp up, there there’s greater incentive provided to Interface to help us get there. That, to me, is a true partnership where where we’re both rowing in the exact same direction. So those are some things that I’ll just throw out there for individuals to maybe take away and as as they start to make decisions, about AI. But I’ll leave it with this on this topic.

I can’t imagine not being in the AI space today. It is simply too dynamic. It is too critical to everything we do. And going forward, it’s even gonna be that much more critical.

Great. And then, Charlie, I know you have, you know, experience with, you know, at Allied as a AI adviser, you know, representing six six thousand plus credit unions and many different AI solutions, different kinds of them. You have significant experience, you know, guiding, you know, credit union community banking industry. So would you mind providing your perspective?

Yeah. Absolutely. I’d be happy to, you know, it’s amazing when you look out into the industry how fast things have been moving. And the pandemic, I think, really launched us even that much further from a speed perspective of how fast things were going.

Self-service was coming, at a very, very quick pace. And I would comment also about this, AI and all the things that you’ve been saying, Sri, and that Todd has been sharing. There’s two things out there, two generations that are probably not gonna do business with you if you’re not in the AI business, and that’s z generation and millennials. Three percent of the z generation work with credit unions on a regular basis.

Eight percent of millennials.

And bringing something to the table like what Dupaco has helps address that in a pretty big way. What you find in fraud out in the industry, it’s complex and fighting it is complex as well. So complexity drives it. And as Todd said, we need to partner with organizations that are willing to invest for the long term as well as the short term because it’s that fluid.

You have to be able to take a good hard look at domain specific optimization.

How are you gonna prevent? How are you gonna use forensics to stop these folks? Todd’s story at the beginning, was spot on. It’s like, wow. We couldn’t trip these folks up, but AI is what it’s gonna take to beat them.

You know, you got trust accounts, joint accounts, you have multiple voice profiles, and it gets into some deep stuff of how these fraudsters are trying to come at it. And they are good. They are very, very good. So the speed of the fraudsters versus AI, me personally, AI is winning without question. And, you know, AI to beat AI, really? I mean, that’s crazy. But if you think about it, we need to know exactly what it’s gonna take to stop some of the things that are happening.

The fact that Todd talked about anti spoofing, you know, on the call coming in well before well before and the member, by the way, and the consumer doesn’t even know this is happening. We’re digging deep and we know exactly what’s underneath the covers of who we’re talking to and the device as well. You know, we can go back to in wallet, out of wallet, but we’ve got the voice biometrics and what we have today is even device biometrics.

Device biometrics was not even here six months ago, yet interface is already working on that. So when everyone says that they’ve got an AI solution, I chuckle to a certain degree because if you’re gonna be adaptive and you’re gonna be innovative, one, you have to have the funds and the long term vision to do it. But the key is evolving natural language processing, which is what where you really started back in twenty fifteen. You evolved it, and now we’re into the generative AI side of things. And it’s amazing what has happened in the sense of how the advanced fraud in biometric, identification, you know, who are we in fact talking to?

But the thing that I love about it is Dupaco wanted to build something that would separate them from the marketplace and the other organizations that are, you know, they compete against across the across, you know, their area as well as the entire country.

There are so many AI providers out there. You said it best.

But there’s really only a couple that are financial institution focused. And I think what it really comes down to is, if your phone platform has got AI and you decide to go down that path, where are you gonna be in a year when you want to grow your AI? You’re not gonna grow it outside of your phone center because it’s proprietary.

You’re not gonna grow it out of mobile banking because, again, it’s proprietary.

You have to have one AI brain, you have to have one AI focus, so it can learn from one another and that the consumer, the member, the customer, however we wanna phrase that, they are getting the same experience no matter where they are. And that is so essential. I’ll finish with this.

What the the fact is data is king. Data is queen. It’s royalty. The more data you have in your wallet, the better you’re going to provide for a member experience.

And the fact that you said one point five million transactions a day, you know, that’s over five hundred million, transactions annually. Why? Twenty four seven, three sixty five. The next closest competitor trying to emulate and do what interface has been doing, and they got started, you know, two thousand eighteen ish, two thousand nineteen.

They do about ten million per year.

So you look at that, you know, I might not be the smartest guy in the room when it comes to technology, but I do understand that, data is currency when it comes to creating the member experience.

That’s what I’m seeing out there.

Well, thank you, Charlie. That’s, very impactful. And, you know, we put together this chart to kind of show how the Dupaco’s automation trend has been kind of going up. And it kind of shows it’s a journey.

Right? And especially with Dupaco starting, you know, we’re first customer launching this biometric quality forensic kind of three legs tool. Now we have many customers on it. But since they were the first one, like, you know, it it kind of this graph really tells you authentication and member verification is, you know, helps you fight fraud, but at the same time, it’s the bedrock of achieving higher automation.

Because if a member is unable to authenticate successfully, you’re never gonna get the value of AI. Right? So because, you know, they’ll not be able to complete any task. So we knew this very early on.

We were looking for a partner to brainstorm and build something, and we’re glad we found Dupaco as a partner. So as you saw when we launched, the automation was low, and then we had to continue to rework on, improving authentication slowly went up. And then we gen launched in AI, went up further. And you can see, at you know, on on on here, we launched a hybrid authentication, which is I talked about, which it starts with biometrics and caller ID forensics and uses the in wallet, out of wallet as a fallback.

Right? Even when it does a fallback, it still uses caller ready forensics. Right? So giving very strong, you know, as Todd calls it, like, mosaic approach.

And then as you can see, immediately, the automation shot up. And then finally, we launched, you know, Tobacco has a a lot of Spanish speaking members. We’ll we launched that too and kind of it now trending towards, you know, fifty percent. And this journey is unique to every credit union and community bank.

Right? This is a journey of Topaco, because it, it depends on their, security policies, risk appetite, their obsession to consumer experience, the kind of membership they’re, driving, and, trying to attract and the serving today. Is this journey is very unique. Right?

If we’re with over hundred financial institution, we draw this graph for every one of them, no two look alike. Right? So, you know, with that, Todd, I wanted to quickly, talk about the overall impact across fraud and automation. Right?

So, you know, this is something we, kind of pulled up. You’re watching about forty six percent automation as the graph showed, and and and net savings of close to three hundred fifty thousand dollars.

You know, I would love to see if you, if Tare, if you wanna add in.

Yeah.

Yeah. So I’ll just I’ll I’ll just jump in and and yes. We are we are stopping fraud.

There’s no question about it. Individuals are trying to impersonate, a member.

And soon as soon as they, soon as that call gets, it goes through its forensic review, it gets sent over to a live agent, has, you know, a stoplight essentially saying ask additional questions. And at that moment, they know that they’re not gonna get through, and then it’s a hang up.

So that has definitely worked. We’re seeing automation rise. And what really excites me, you know, we’re in the infancy of AI, folks. We are in such an infancy, but the next six months, twelve months, twenty four, thirty six months, this is gonna be amazing what’s what’s gonna be available.

And to have an adoption rate at at nearly almost fifty percent, and we’re ready after one full year seeing positive all expenses, less our savings. So it’s a true net savings, of three hundred and fifty thousand dollars.

I am really excited about the future of what we can do together as partners because we are in the infancy of this. And and when we see AI continue to grow and blossom and bloom, the sky’s the limit.

Yeah. Talking about the possibilities, I wanna, pull up this one, a couple one slide here. You know, some of the things, you you know, you and your team are working on. Right, Todd?

You know, soon you’re also launching device biometrics. Right? Now you’re using voice biometrics now and kind of device biometrics is another option, for your members and things like that. And you’re also launching in business banking in the coming days.

Right? So and then Spanish for business banking. And then you’re also now taking the AI journey to chat and digital. Right?

So, so having the same AI for the members on the website, online mobile banking to get answers, to questions twenty four seven, in a personalized way. That can also speak Spanish and whatnot. So future is exciting. We’re breaking the barriers.

Language shouldn’t be a barrier for any, any, you know, important tasks like financial services. So that’s, something, we’re proud about partnering with you doing that as well. I know we’re coming to an end. I wanna end with, you know, before q and a, for, the all the attendees, like, this is exclusive offer, you know, we’re offering to the webinar attendees.

The the first five registrants who go to our website you know, reach out to Laura.

You know, we’ll get, like, a device biometrics, which is what you saw in the demo, you know, at at no licensing cost. So, wanted to, you know, thank you for being part of the webinar and give you something back. So

appreciate joining. So we’ll jump into q and a, real quick here. So I will, take some of these questions, Todd and Charlie, we could all three of us together answer.

So, Todd, you spoke about it. If you have anything to add, what have been the member’s reaction to this automation?

That’s a that’s a fantastic, question, from from the audience. I appreciate that because it it it’s mixed.

We have members that are absolutely loving it, especially the second and third shifters, the individuals with Spanish, and the individuals that know the extension they wanna speak to or know the individual they wanna speak to, that all is handled so quickly and efficiently.

Where we’re seeing a little bit of, of resistance, and that’s what’s holding down our our numbers to that fifty percent is, you know, some of the the boomers and some of individuals that are a little bit even older than the boomers.

Those individuals are a little bit more reticent to the to the change, And there’s an opportunity and, you know, we certainly are taking that opportunity as much as possible to have a direct conversation with any type of concern expressed or if if we learn of somebody that’s not having a good experience, we wanna review that experience, which, by the way, is fully available within the system, and then call that member up and say, okay. Let me re recap that last experience you had, and here’s where things got astray.

And most of the time well, I can speak almost inevitably.

It’s not a technology failure. It’s it’s it’s a human it’s a human error. So, it gives us that opportunity to have that discussion. Thank you. Great question.

Thank you, Todd. And there’s another question.

If transferred to live agent at Tobacco, do you, do they the agents still see red flag, the member with risk rating before getting a live person?

Absolutely.

That that flows right over into the contact center. So when somebody when somebody either has has a a bad voice authentication or, their, forensics didn’t pass the muster, it comes over with a red stoplight saying, you gotta dig into this one. I we don’t feel comfortable. Now what’s I don’t wanna call those false positives because in most cases, they’re either using a borrowed phone or they’re traveling and they’re calling from Seattle, Washington when they normally live in Dubuque, Iowa.

So those flags actually did their job.

It didn’t mean deny the call. It just meant dig in a little bit more because because this could be suspicious.

So, we don’t see a lot of false positives. We we see a lot of opportunity. Most pass right through, but every the ones that that don’t, yeah, it it merits that additional questionnaire.

Got it. No. Thank you, Todd. Charlie, you wanted to add anything to that? Feel free to jump in.

No. I think Todd is, you know Okay. He’s so close to it. I think he’s Yes. On.

Yes. And and and there’s another question here. A couple of questions kind of related, Todd.

You know, as as a as a, Dupaco leadership team, how did you deal with, agents having the fear of losing jobs, you know, how to deal with the oral, you know, agent, call center agent, sentiment so that they’re more working, to make this AI successful than, being in fear?

Yes. And that’s a that’s an excellent question. And it’s it’s something that we all have to pay attention to because nowhere along the line did we start this out with saying, okay. We wanna reduce our staff by x or we wanna, no, we do we wanna get more efficient? You bet. But at the same token, we wanna be able to leverage the skills of our talented team to spend more time on complex member questions and service than simply regurgitating routing number balances and, you know, the last three items that cleared their account.

And, I believe those are the the types of things that that we can leverage to to deepen member relationships as opposed to, create fear with the member or fear with our staff.

Got it. Got it. No. Appreciate that. And, there’s a couple of questions here related to, you know, what percentage of cost, being serviced by AI only?

Have you reduced agents in a call center staff or time? What has been the cost savings? We touched upon this, Todd. If you wanna quickly, briefly touch, you know, reply to that, appreciate it.

Yeah. So so, we have been able to reduce our our our, our cost center, our, number of of staff members in the contact center, numerous agents, and, essentially, had that reduction has exceeded far exceeded by several hundred thousand dollars, what we would pay for interface, not counting the biometric and all that fraud tool protection that we wrap around the interaction.

So I will just end it with with one quick thought. If you’re not using AI today, you you really need to be thinking about it. And if you’re not wrapping fraud prevention tools around that AI with a partner that is exclusively focused on financials, you really need to be considering that.

Awesome. Thank you. Well, that’s a great, way to end the discussion about fraud and AI. Two big topics, this year, and this is, is the most important webinar.

I appreciate you all being part of it. Charlie, Todd, and everyone on the call, appreciate you all joining and, going through this. Appreciate thank you. Have a great, week ahead.