Listen: University Credit Union saved $2M in 1 year with AI-powered bot

University Credit Union saved nearly $2 million in expenses and revenue gains in the first 12 months after launching its artificial intelligence (AI)-powered virtual assistant Royce in August 2020. The savings easily amounted to more than the cost of the technology and allocations made to support the bot, says David Tuyo, president and CEO of the $1.1 billion Los Angeles-based credit union, in this episode of “The Buzz” podcast. Tuyo now encourages other banks and credit unions to embrace AI-powered bots. “When they go down the path, they're going to find out that it's cheaper up front than what they thought, and they're going to find out that they want to invest more than they originally thought,” Tuyo says. “It’s more inexpensive than you might think upfront, but then over time, you're going to want to scale pretty quickly and people are going to want to adopt this faster than you anticipate.” In fact, once the UCLA-affiliated credit union rolled out the customer-facing Royce, other departments such as human resources and IT soon wanted their own bots, he says. In this episode of “The Buzz,” Tuyo offers advice to banks and credit unions about how to get started in AI-powered bots , from how to choose a vendor partner to adding robotic process automation capabilities to chatbots, a move he says is helping the credit union weather the pandemic and the “Great Resignation.” "You need to think about AI as a marathon, not so much as a sprint or a one-time project,” Tuyo says. “You need to think about regulatory issues, you need to think about ethical issues up front, and be aware about what those could look like but, ultimately, recognize the value creation and the rewards that your organization is going to receive from diving in.” Subscribe to The Buzz Podcast on iTunes , Spotify , Google podcast , or download the episode .

The following is a transcript generated by AI technology that has been lightly edited but still contains errors. Loraine Lawson: Good day and welcome to the Buzz, a Bank Automation News podcast. In the first 12 months after rolling out its artificial intelligence-powered virtual assistant, Royce, the University Credit Union saved nearly $2 million in expenses and revenue gains. It’s made Dr. David Tuyo, president and CEO of the $1.1 billion credit union, something of an advocate for AI-driven virtual assistants. I sat down with Dr. Tuyo to discuss how banks and credit unions can growth with the help of chatbots, AI and robotic process automation.David Tuyo Yeah, I think we are looking at the most efficient ways to scale. And so when you talk about, you know, efficient ways to scale, you look to technology first, right? There are no more important things that we do in the crane than then serving our people. But at the same time, we need to find ways to scale from an economics perspective, the most efficient way possible to add value to our stakeholders, those that we serve. Also, there are team members as to making sure that they have robust compensation packages, as well. And so, um, you know, for those that are not in AI, they're really gonna be left behind. And I think that, you know, that could lead to some interesting things around mergers, itself, further consolidation for those that do or do not participate in AI, earlier later. And so, what cranes can do is obviously leveraging AI at all your touch points. So AI bots can replace many of the basic questions you most cranes and banks, when they look at their contact center, why people call in, they're going to say, Oh, well, that's because, you know, they check in their balance. So they're trying to transfer some money. Those are things that AI can handle very easily. And then as we're learning over time, of course, how to be more effective in serving the members as well, when you look at accuracy rates are our AI bots to interact with our members, which had over 35,000 interactions last month alone, they did so with a 98 point 98% accuracy rate. That is just tremendous. I mean, it's astonishing to think about near perfection throughout the entire month. And so I think the big piece that we're talking about going forward, though, is around AI, machine learning is also leveraging RPA robotic process automation, which many already do. But connecting those two technologies together to now you have that automated, low cost touch point in servicing your client base to then also bridge the gap and using rpa, to launch rpa, to then create automation for fulfillment. And that's where the magic happens is now you're using technology to really scale in a way that you couldn't otherwise, when you think about, especially right now, I mean, he AI and RPA has helped us weather the storm around the great resignation to help support our team as they went through whether it was directly or indirectly affects from COVID-19. And, and so it's been it's been a tremendous tool for us. And, you know, can't say more about I think I think, you know, as others look into this technology, we're getting calls every week from potential credit unions that are saying, Hey, tell us more about how you're doing this. And, you know, are people scared of the bots or do people really embrace them? And I think that's the biggest concern right now by Credit Union and banks. And we're finding out that, you know, across the country, we have members in every single state that our members are really enjoying interacting with our AI tools.Loraine Lawson Tell me about how you came to have an AI-powered virtual assistant? David Tuyo The talent we launched was in July of last year, so we started looking at it before that, of course, then we had the lockdown to March and May and everybody's trying to think, what what is the world going to, you know, is it going to even exist after this, if you think back to the culture that we had back at that moment in time, we weren't sure what this thing was or how bad it was going to be. And, and so we kind of put everything on pause. But then as soon as we came from that, and we unthawed from that lockdown, or that retreat, and rightfully so, we then jumped into back to work, right? Like we look, what does it look like. And so July, we launched our very first AI tool that in our tool, our AI bot, the first one we launched was called Royce. And we named it Royce, because we were founded at Royce Hall at UCLA. And that's the most iconic building you see on campus. And you see every picture, the one with the two columns, it's beautiful building, that's Royce Hall. And so we want to pay a little nod to our heritage, where it came from, and our founders at UCLA and that's what we named it Royce. But we launched Royce in July took about 10 days to get the first set of instances out and then from there, Royce has been learning and growing ever since, serving in loan applications, new membership applications. Now, when you call into our contact center, Royce answers the phone Royce is now handling all of our calls from the initial touch point and answering many of those, except the ones that that the bot's unable to which case then they will pass it off to one of our contact center specialists or coaches. Loraine Lawson You mentioned combining RPA with your Chatbot. Basically. Can you give us some examples of how you've done that to support automation or to drive growth? David Tuyo Yeah, I mean, so you think about just something simple, like we'll do like a credit card lost or stolen, right? situation, Royce now can take care of that complete solution for our members, without requiring any additional human interaction. We have a whole roadmap of things as we go through our list about why our members call, we start at the top and work our way down. And just automate each one one at a time to make sure that we can get that fulfillment, that speed of execution, the accuracy, most importantly, quality of service our members deserve. And then we're working through that piece by piece. With credit unions - in many that are our size, we're not quite big enough like a Bank of America where we have a skyscraper full of programmers. So typically, we have a variety of vendors and third-party partners that we have to use to make things happen. So it makes the topography of our ecosystem kind of complex. And the tools really do allow us to get and talk to each one, which is fantastic, but also many will find that the challenge is trying to Get and talk to each one. And so that is, you know, some of the things that smaller institutions are gonna have to face. But it's not it's not insurmountable, you easily can work through it and usually just with some education, and connecting, you know, partners together, you can find a way to make it happen. Loraine Lawson So Royce did, is that something you built in house, or was that sort of a partnership deal. David Tuyo So we have a partnership with a company called Interface.ai, fantastic company, they're out of the Bay Area, and up in Northern California, and they're the ones that have really supported us and helped us grow in this endeavor and support our initiatives around Royced, then we also have an internal bot that we've created, called Coach and Coach is not nearly as robust as Royce, but it's we wanted to, you know, learn and kind of, you know, get educated internally also build some skill sets internally to understand the technology a little bit better, a little bit more intimately. Because we do think it's the future. And so we want to make sure that we have that core competency on staff and in house as well as leveraging tremendous partners outside and an interface ai, I can't say enough about that company, but also, internally Coach has replaced our intranet. It's replaced our IT ticketing system, replace our marketing ticketing system. And so coaches really become a robust tool internally for our team members, as well as they say, hey, hey, Coach, I don't know how to do a particular function, then Coach will serve them off to a particular process or procedure they need to be to give that quick and accurate level of service to our member owners. Loraine Lawson Well, the benefit of having an in house too is you can add to it easily right. I wondered, though, how did you develop? Did you have, you know, an IT staff that then you developed the talent within staff, sort of training them? Or did you recruit? Or how did you staff that? David Tuyo We're heavily we will always probably heavily rely on Interface in the next few years. But with regards to our intern development capabilities, our team was really curious about technology, and they just want to learn more about it. The technology is not that complicated. And so, you know, we look at the tools are currently available to us, you think about even, you know, how we build websites today is still is significantly simpler than what we did 20 years ago, you know, and, you know, a novice programmer could easily go in and build, you know, a bot that can do basic functionality. And so our team was really curious about this and want to learn about it. So we did not add any resources or or hiring anybody that had specialization in this area at this time. Probably will change in the future. But we want to make sure, again, at least have some general list level of knowledge when it comes to some of these tools. Loraine Lawson And what a great way to keep staff, you know, by training them. That's what I hear from programmers the most is that they want investment in their career. They want training. The other question I wanted to ask you is do you have advice for credit unions? I know you like interface AI, but surely you went through a process where you evaluated the vendors and chose to work with them. Do you have any advice in that regard? There are a lot of vendors out there that do chat bots, and some in AI and machine learning. And I think it can get confusing. David Tuyo Yeah, I mean, even you know, not all bots are created equal. That's one thing we learned through our research, we started with about just over a dozen different companies that we were looking at and our due diligence process, before narrowed down to two, there were two that were really neck and neck and then we got down to one, obviously was the interface choice. And it really came down to what kind of experience we want to deliver, we want an Apple like experience, we want a partner that can deliver and that's that's how we got down to interface. But I think the biggest thing is also understand what you're getting. With interface, you're getting a truly intelligent virtual assistant, whereas with others, you might only get a chatbot. And there are a significant level of differences. When you - I'll give you an example. So um, you know, we serve everybody in the university community. So we have students, employees, faculty, administration, alumni, and sometimes students, they think a little more on Friday nights, and they want to talk to Royce. And so we've had some very interesting questions along the way. And, you know, you know, we had a record interaction 41 times back and forth. And ultimately, the one that Royce got stuck on was, you know, it was Royce available for a date on Friday night. And so, you know, Royce didn't know how to answer that. But there's questions that all the time, you know, where they noticed, hey, Royce, you know, where's the nearest pizza location near me, and you know, Royce doesn't necessarily know where the nearest piece of location is near them if they haven't enabled their GPS. But if they have, what Royce will do is the bot will come back and give you the nearest location of our credit union. So one of our either one of our short branch locations, our nearest ATM or free ATM that's nearby. And then you so again, allowing for a greater level of service. Royce learn how to do that we didn't teach Royce to do that and get answer response. Royce learn that over time to these interactions to ultimately serve the of that as a response Loraine Lawson has Royce come up with a question for that dating-- answer for that dating question. yet? David Tuyo not not not yet we should teach it how to respond to, to those kinds of questions. But not not at this point, I think, you know, when you think about some other bots that are out there, like a like Alexa, so speak from Amazon, they have some interesting responses, especially when you start asking, you know, are you better than Google assistant or Google Assistant better than you? You know, they have some really funny responses. We're not there yet. But but definitely something that we'll be looking at going forward. But it's a it's a great suggestion. Loraine Lawson Maybe you could suggest that they meet at the local credit union to find someone. Any lessons learned that you'd like to share with other credit unions before they get started on this process? David Tuyo Yeah, I think the big thing is, understand the size of this market. Most people in our space, I talk to seem to think that this is an emerging technology. And it's really not in 2020, there was just under $8 billion spent on AI invested in AI in the United States, that's just the United States. By 2026, there's gonna be $26.7 billion spent annually on AI. And so that's, that's a cumulative annual growth rate of 23.17%. It's astonishing to think about not only the size of it today, but the pace and size of it going forward. I think concerns that you need to think about as you go forward, I think you need to think about AI is a marathon, not so much as a sprint, or a one time project is can be a long term type of initiative for your organization, I think you need to think about regulatory issues, you need to think about ethical issues up front, and be aware about what those would those could look like and, but ultimately recognize the value creation, and the rewards that your organization is going to receive from diving in And leaning into this technology. It's been tremendous for our organization. And we're willing to hop on anybody that has questions, because we think it's impactful and is the future around certain levels of banking services in the next five to six years. Loraine Lawson Can I ask how much you had to increase or how much you recommend they increase their IT budgets, if they're planning on going down this path of like a true digital system, David Tuyo I think when they go down the path, they're going to find out that it's cheaper up front than what they want than what they thought, and they're going to find out that they want to invest more than they originally thought as well, because as soon as we launched Royce in July of 2020, guess what, you know, that was a member facing member touchpoint. So operations and Member Services, were very excited about that particular tool. But then all of a sudden, now HR wants their bot, and IT wants their bot, and you know, everyone's a bot - bots everywhere. And so, and it is tremendous, what can happen. And so that's that's that would be my, my caution is that, you know, hey, you're going to use probably in a more inexpensive than you might think upfront, but then over time, you're going to want to scale pretty quickly, and people are gonna want to adopt this faster than you anticipate. Loraine Lawson So it may pay for itself, I guess. David Tuyo Oh, yes. So in our first in our first 12 months, we saved in excess of the cost of the technology and allocations that we made. We've saved nearly $2 million in expenses and revenue gains because of our tools. Loraine Lawson You’ve been listening to the Buzz, a Bank Automation News podcast. be sure to visit us at Bank automation news.com for more automation news. You can also follow us on Twitter and LinkedIn. Please don’t hesitate to rate this podcast on your podcast platform of choice.

Listen: University Credit Union saved $2M in 1 year with AI-powered bot

CEO offers advice for virtual assistant startup

By the Numbers: BMO Harris and Chase among most-imitated brands in phishing attacks

Cross River moving core to Amazon Web Services by end of March

Related Posts

Why FICO built its own LLM

Employee feedback spurs AI innovation at Wintrust

Product roadmap, data governance key for AI vendor selection

Cross River moving core to Amazon Web Services by end of March

EMERGING FINTECH DIRECTORY

FinAi Podcast

SPONSORED

The Next Generation of Cyberthreats: Powered by AI

Build an Antifragile Strategy to Outperform the Market

How AI and Product Experts Turn Fuzzy Requirements Into Focused Dev-ready Roadmaps

Connect

Welcome Back!

Retrieve your password

Continue Reading with FinAi News Premium - Less than $2/Day

Listen: University Credit Union saved $2M in 1 year with AI-powered bot

CEO offers advice for virtual assistant startup

By the Numbers: BMO Harris and Chase among most-imitated brands in phishing attacks

Cross River moving core to Amazon Web Services by end of March

Related Posts

FinAi Podcast

Connect

Welcome Back!

Retrieve your password

Unlock This Article

Continue Reading with FinAi News Premium - Less than $2/Day