ACUMA ONpoint
ACUMA ONpoint
You Can Adopt AI Without Losing Control
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
AI in credit unions can feel like a risk until you see it as a workflow you can test, measure, and control. We’re joined by Eric Burgess, Director of Home Loans at First Commerce Credit Union, to talk about what it really takes to get comfortable being uncomfortable and start using artificial intelligence in mortgage lending without handing over the keys. Eric shares how he delved deeply into AI training and why he helps others learn through the AI Learning Lab community.
We get practical fast: prompt engineering as the starting point, why “human in the loop” is non-negotiable, and how leaders should think about data security when employees use public LLM tools like ChatGPT. Eric breaks down a crawl-walk-run approach for credit union AI adoption, including a simple but powerful pilot: running vendor contracts through an AI agent in seconds, then comparing the output to attorney redlines that can take weeks.
From there, we move into agents and automation using tools like Microsoft Copilot Studio, including a mortgage quality-control agent that flags risks and dramatically reduces QC time while leaving final decisions to people. We also explore what “advanced” can look like, from AI-assisted underwriting that boosts underwriter productivity to predictive analytics dashboards that improve planning and forecasting.
If you’re looking for real AI use cases in credit unions, mortgage operations, and lending productivity, this conversation is a strong place to start.
Subscribe, share the episode with a colleague, and leave a review to help more credit union leaders find it.
Sponsored by Xactus
Why Getting Uncomfortable Matters
SPEAKER_05This is Accomazon Point Podcast. On today's episode, we dive into why we need to start getting comfortable, being uncomfortable, more specifically, why it's so important for us, credit unions, to adopt AI and really how we could start using it.
Sponsor Message From Xactus
SPEAKER_05But before we get to our episode, just a quick word from our sponsor.
SPEAKER_01Exactist is a leading fintech committed to the continued transformation of the mortgage verification industry, pioneering a new class of technology, intelligent verification. Exactist is redefining how the industry originates and services mortgages. With Exactist 360, our industry-first intelligent verification platform SM, we put the full power of the market's leading verification partner into a user-centric technology that harnesses real-time insights to power automated actions, enabling clients to make faster, better decisions with the right data at the right time. Zactus is an Acuma services provider for credit reporting, verifications, and flood determinations. Exactly also provides Acuma members unique pricing benefits for flood services, valuation technology and appraisal orders, loan-level quality control solutions, and Xactus data solutions. By combining data at scale, advanced AI technology, and our deep domain expertise, Exactus is committed to partnering with mortgage lenders and servicers to lead the industry into the future as, together, we advance to modern verification. Our team of experts at Exactist is here for you, the members of Acuma to consult on your verification workflow and being provider agnostic. Here to offer custom solutions to accommodate your strategic goals.
SPEAKER_00The podcast discussion presented is conversational in nature and for general information only.
SPEAKER_05Ladies and gentlemen, boys and girls, hello, welcome to Academic Podcast, a series focused on sharing the stories of people who are making a positive impact in the credit union mortgage industry. I'm your host, Peter Benjamin. Today I am joined by a good friend and someone I highly respect, Eric Burgess, Director of Home Loans with First Commerce Credit Union. Eric, my friend, how are you doing today? I'm doing great, Peter. How are you? Hey man, just live in that dream. Live in that dream. You know, Eric, as I mentioned, we're going to talk about AI. You agreed to have this discussion with us on why it's so important for credit unions to start getting comfortable with the idea of AI. Because really, in my opinion, you are someone who has truly embraced that notion of AI. I mean, heck, I think you we've even started a Facebook page completely dedicated to AI utilization. But obviously, we'll discuss that a bit more in just a second. So looking forward to the conversation.
Acuma Updates And Annual Conference
SPEAKER_05But as always, I need to take a sidestep, bring Justin the Hawk into the conversation. Justin, my friend, how are you doing today? And uh please tell us what is the latest and greatest happening over at Acuma.
SPEAKER_04I'm good, Peter. How are you?
SPEAKER_05Hey man, like I said, living the dream.
SPEAKER_04Living the dream. I love it. Well, right now, we are what are we, six and a half weeks out from annual? Something like that. Yeah, something like that. Six and a half weeks out from our make your mark annual conference. So a couple things. If you don't know, we're gonna be in Las Vegas this year, September 20th to the 23rd. So if you don't know that, then I haven't done a good job. Or we haven't talked about it enough. I don't know. One of those two. I feel like we've talked about it enough. And if they don't know, they're just not paying attention. They're not paying attention. That might be what it is. Registration's open. So if you haven't registered for what is going to be the biggest credit union mortgage event of the year, then you need to go do that. You definitely don't want to miss this event. The agenda, the speakers, the lineup, the fun stuff that you don't know about that's not on the website. You don't want to miss these things. Secondly, if you have registered, make sure that you go book your hotel room. So you get a discounted room block with that. So make sure you book it before hotel deadline happens. Yeah, aside from all that, we have our webinars going on on a regular basis, so a couple times a month, and then our on-point podcasts are back on our regular schedule. So new episodes dropping about every two weeks. For all that information, plus so much more, head over to the Acima website. Awesome, man. I appreciate it.
SPEAKER_05Thank you very much.
SPEAKER_04Of course.
Meet Eric Burgess And His Story
SPEAKER_05All right, Eric. I know we're gonna talk about AI, and and you are someone who've I've been watching, and I mean, you really have gone down that path of adopting it, learning what you can about it, try and find ways to leverage it for your credit union. But you know, really before we get to that conversation, I have to always start our conversation with the same first and last question that I ask everybody. You know, and and again, Acima's on-point podcast is a people piece. It's focused on making, it's focused on the people who make a positive impact in the credit mortgage industry. It really started with that idea of what makes us us, right? And how do we keep driving forward? First question out of the gate is is more it goes back to that root. It goes back to our that fundamentals of who we are. I'm sure you can figure it out. That first question is is relatively simple. Who is Eric and what makes Eric Eric?
SPEAKER_02That's an interesting question. I've been trying to think what I would answer with this, but I get excited by innovation. I get excited by growth. So I've spent the last over a decade of my career really focusing on helping organizations grow on the lending side. Done it at a credit union before and then at banks, and then some independence. And for me, it's it's I get excited growing organizations and utilizing lending as one of the growth catalysts for the organizational growth, whether it's using mortgage to deepen relationships and cross-sell, et cetera. And when you can find ways to do that creatively, especially now with the access to technology, is getting simpler and simpler to really try and connect all the dots and uncomplicate your tech stack. That gets me really, really excited.
SPEAKER_05And I appreciate you walking us through that. You know, and I think you're a great example of someone who most credit unions could could take a page out of your book, right? You started off as one thing. You started off as a good old-fashioned mortgage lender. You saw great success throughout your career, but then you're like, hey, I have to innovate, I have to become better myself. You started really adopting AI and new technologies, and it's only aided in your success. And I think that's why hopefully our listeners, people that are listening to this conversation, take something away. Like you are someone who's fully embraced new technologies, AI, you've leveraged it for that success. And I think one of the things that most credits fear is just even the idea of AI, but also just overcomplicating it with things that they don't need. For background, like you said, you're a mortgage learner. You love AI and you you explore it a lot. You're also one of the founders of the Facebook page, AI Learning Lab. You really talk about where the really the group talks about techniques for using AI and new AI that's out
Building The AI Learning Lab
SPEAKER_05there. Just do me a favor, before we go much further, walk us through the journey to kind of start that Facebook page, that group.
SPEAKER_02I started using AI back in late 23 and really started playing with it. And then I really took a deep dive in 25. I took a step away from the mortgage and the banking side of the business, and I really spent 18, 24 months really diving deep into artificial intelligence. I wanted to understand everything, everything I could about it, everything from what makes it work on the LLM side of things all the way up to the agentic, the integrations, and everything else you can do, the vibe coding, all the buzzwords that are related to it. I have multiple certifications, engineering, generalists, the whole gamut. I decided to actually also join an MBA program. So I'm in a master's program for artificial intelligence as well, which has been interesting because that's more of a business use case as opposed to like an engineering use case, if you will. But I was invited to join that Facebook page by somebody in the industry who is struggling, kind of getting it off the ground as the AI expert. And it's kind of evolved into what has become today. In fairness, it my focus hasn't been completely on building that page lately as much as it should have been, because my role at First Commerce has taken a little bit of a priority over where that was. But I definitely found joy in trying to help people understand artificial intelligence and how to use it in their day-to-day. I mean, and there's use cases galore with it, whether it's organizing your schedule at home to developing quality control agents in a lending platform or processing agents or even underwriting agents that assist with calculations on income. So the use cases with it are relatively endless. And my passion was how do we do it, how do we do it well? How do we do it better than anyone else? And how do we make sure that we keep it on point with accuracy as opposed to the horror stories you hear out there with the hallucinations and everything else? And so there's a lot of controls that you can put in place to really harness and yield the optimal results that people are typically looking for in the artificial intelligence space, which is productivity, right? Artificial intelligence is a massive productivity boost. You know, if you asked me a year ago what my thoughts were with AI, it was a lot of people were talking about job replacements. And my my thought on that has changed or has evolved a little bit. But I enjoy helping people understand artificial intelligence, and that's kind of the crux behind where that group came from. As I was doing a lot of my own teaching on my own personal page on there, I was invited to help.
SPEAKER_05I love that. You invited me to be part of the page, and I I know I haven't contributed to that page. I most certainly read all the posts, and I I see the conversations that are happening, and it and it's fascinating how quickly for people all around the world are on this page, but also leveraging AI. I I say that because you know it seems like the people who are adopting AI are growing. It is growing, excuse me. But the thing is, is that when you look at our industry, and I think this is where it has to tie back to work a bit, and I've said this before. You have all the various different industries, you know, you have automotive, healthcare, banking, etc., right? If you look at healthcare as an easy example, they are light years when it comes to AI ahead of us, right? And then when you look at banking, you know, we're banking compared to healthcare, we're four years, four or five years behind them. But then when you look at the credit union space and you look at the rest of the banking world, you know, we're five years behind the rest of the banking world. So all in all, we're about 10 years behind AI when it comes to our utilization, our understanding, and and overcoming that fear, right? That's why I love so much about that Facebook page, is that you can see people are actually out there using it, seeing success. And so I guess that you know, one of the first questions I have is well, let's focus on something that goes back to why you started, you know, that Facebook page of how can people overcome that fear and that perception of AI and really start using it for to its full benefits.
The First Step Is Better Prompts
SPEAKER_05Like what is that first step to overcome that fear?
SPEAKER_02Just like everything has a process, right? So when you turn on your computer in the morning, you have essentially a workflow on how you turn on your computer. You press the button, it boots up, you log in, you open your apps, whatever. Artificial intelligence is no different. It has a workflow. And it all starts with essentially the prompt and the prompt engineering. And you utilize something called the crit framework. In fact, I have a book on my desk called the AI Savvy Leader, which I read last year. And he talks a lot about how you become a leader in the AI space and how you utilize it. And it really starts at the basic is understanding the prompting behind it and how that works to generate the results that you want. And once you're able to start prompt engineering properly, you can direct AI to help you in any aspect of your life, whether you want it to help you. I'll give you a great example. Everyone is struggling with costs right now, right? Inflation, grocery. I don't know about you, but my grocery bill is ridiculous. I have three kids.
SPEAKER_05I mean, you know how expensive fruit snacks and chicken tenders are.
SPEAKER_02I have a 15-year-old boy. Yeah, exactly. So, and I have I have a 13-year-old boy and an 11-year-old boy, and they ate me at a house at home. And so I was frustrated one day because I was tired of going to Publix and spending three, four hundred dollars on groceries. So I put my grocery list together, I uploaded it. I have an AI agent I use at home. I named him Enzo. I prompted it using the crit framework on what it is, what the goal is, what the outcome should be, and to do a full analytical deep dive on the grocery bill. And basically the result that I wanted was what's the best place to go shopping to save me the most amount of money. And the result came up and it said, if you go to Publix, it's this, if you go to here, it's this, here it's this, here it's this. I've seen Aldi, I've never gone to Aldi before, so I decided to take a chance. I went to Aldi. At Publix, this bill would have been 450 bucks. At Aldi, it said it was going to be between 220 and 250, and I paid $242 for that grocery shopping bill. Saved me over $200 simply by asking it, where can I go shopping to get everything I need on this list and save the most amount of money. That's a silly example, but it's a use case in how you can utilize something in your personal life that you may not realize. You can analyze a list for groceries, you can figure out who has best sales, you can organize your scheduling with your kids and their schooling or summer camps or everything else. It's really just about adopting it and starting to utilize it, but you have to understand the basic. And I would encourage everyone who doesn't use AI today or is hesitant to start understanding the framework when it comes to prompt engineering. And if you're a leader, you should definitely read the book AI-driven leader. I'm not the author, I don't get paid a commission for promoting it.
SPEAKER_05Listen, if this is a stupid question, I apologize. Actually, I don't apologize. I I want to know the answer. You talked about your AI agent, whatever whatever you called it. How did you create that? Like, did you just like go to like the chat or Claude or Perplexity or or Copilot and be like, yo, I'm gonna start calling you Jeeves.
Personal AI Agents And Data Security
SPEAKER_02So, like everybody, I started using AI on Chat GPT, right, back in late 23 sometime. And then from there, as I was going through all my trainings and everything, certification, everything I really spent time on, I basically locally host there's a lot of open software out there, if you will. So I went and I basically took an existing LLM and I hosted it on my server at home. And then I started training it and prompting it. It is like an open AI platform, but I host it at home. So all the information I put in there is secure. That's part of the piece of AI that I think is a lot of business leaders, and we can talk about this later, but I think a lot of business leaders need to ask themselves whether they are using AR or not, their employees are, and what information are the employees giving to public LLMs that should be hosted internally, right? If you have ChatGPT at your workplace, what are your employees using it for that they should be using a locally hosted LLM or maybe a nicely built co-pilot that you have on your network as opposed to going to some URL somewhere? But I built this agent on my own system and hosted at home. And so now it's got years of data on me. So now it calls me by my first name, it knows my wife and kids, knows what we like to eat apparently from the grocery store. You know, the whole gamut. It's actually it's quite interesting to utilize.
SPEAKER_05Now I I want to go have one of these. It's almost like because I mean we have I have an iPhone, so I have Siri on my paperweight, and we also have Google Home pretty much throughout the house. So we pretty much just ask Google for whatever we want. I want that agent that helps us do exactly what you're talking about. You know, organizes our calendars, helps us with the grocery list, you know, all that stuff is fun, right? And I think that that's that is a personal use case that is an excellent example.
SPEAKER_02You can build that on ChatGPT or OpenAI platform if you don't. Locally hosting's a little advanced, but as long as you're not putting personal confidential, I mean you can put whatever you want. You can upload your personal tax returns on there. That's that's completely up to you. But I can give you the prompting that we built out for ours, and you could basically update the settings in Chat GPT as an example to mirror the personality and prompt it the way you want to work with it. That's awesome.
Crawl Walk Run For Credit Unions
SPEAKER_05Let's kind of switch back to credit unions and AI. And I I think that is absolutely just it's it's fascinating, right? And so what we talked about, obviously, the overcoming of that hesitation, right? If you were gonna say this is the first place and the best place for a credit union to start leveraging AI, where would that be? Let's do crawl, walk, run type of mindset. This is where you crawl. Leverage this for crawling with AI. This is how you walk with AI. And if you really want to be advanced, this is how you run with it. What would those three things be in the credit union space? It can be related to mortgage, you name it, or use case for mortgage. But in your opinion, being so advanced compared to everyone else, every other credit union out there, where would you go?
SPEAKER_02So the first thing that I think the easiest place to start, I mean, the obvious is letting it proofread whatever you're writing or whatever you've written and offer suggestions from an editing standpoint. You know, that's pretty basic. A lot of people are doing that now. And if you don't prompt it right, you get all those M-dashes and you get the emojis and you get the loop backs, and that's not this. You know, they'll say some fact and then they'll say it's not because of this, and it's not because of this, it's because of this. You can obviously tell it's written by AI. So you have to you have to really learn how to prompt it so it
Faster Vendor Contract Reviews With AI
SPEAKER_02doesn't do that. But if you want to start somewhere, one place we started here was we started comparing when we do vendor due diligence and we would send the contracts to our attorneys to review for for red line on our contracts. We would, at the same time, we would send the contract to our AI agent to review it. And then we would compare the results, the attorney's review and the AI agent's review. And what we found is more often than not, the AI was producing better results that were more conservative than what the attorney was producing. And the AI was doing it in 35, 40 seconds versus three weeks waiting for a law firm to return a red line contract, right? So after we started testing and testing and testing and testing, we still utilize the attorneys, but we are getting more and more comfortable as we move into the AI space in regards to the contract reviews. Now, one of the controls we have in place here is that we can't let AI make decisions for us. So there always has to be human in the loop, right? But at my level, what I can do is I can look at the red line version, I can say, I'm gonna accept these, or I'm just gonna say I don't care what the attorney says and we're gonna sign it. So now that we've started using the AI agent for some of that, we can make some of those decisions much quicker in analyzing vendor contracts. That's a really good place to start, is you just have your traditional path and then create your AI path and then start comparing. This is where you keep human in the loop constantly. You always want to see the results and validate the results.
SPEAKER_05You never want to just take it on first run and before were you taking the attorney reviewed contract and then uploading it to AI and pretty much saying, okay, well, this is what how our attorney reviewed it. This is how we feel about these edits.
SPEAKER_02Start looking for things like this. What I would do is when the same day I'd send it off to the attorney, I'd immediately go to AI and I'd prompt the AI. And again, using the correct framework, you want, and this is very high level, you want to say you are a well-known attorney, 20-something years in the industry, you have a Harvard and Yale degree. You know, you are an expert in corporate law, specifically related to banking and credit union. You understand vendor contracts, the risk, yada, yada, yada, right? You build out this whole prompt. Then you upload the contract, right? I want you to review this contract from a risk perspective like any credit union or bank or financial institution would in terms of reviewing this vendor. You upload that, it's going to spit out a response. I'd then sit on the response and I'd wait three weeks for the attorney to respond. I'd then get the attorney response and I would compare them. You know, I have multiple monitors, I compare them side by side. And a lot of times my AI was significantly more conservative, but also calling out things the attorney potentially missed. So then, to your point, I would then take the contract with the red line and add it to the same chat thread, the same prompt, and say, these are the suggestions the attorney made. Please review them for accuracy. Do you agree with this? Should we make any changes? What are your overall suggestions? Again, this is very high-level basic prompting, something much more detailed than that. And then it does a comparison and it would tell us, yes, I agree with this, I agree with this, I agree with this, I don't agree with this, or say, This is you don't need this, this doesn't matter, it's not relevant, whatever it might be. And then it can do a comparison. And then from there, we can make a decision on do we follow the attorney's advice, do we follow the AI's advice, do we do a nice blend, or do we just go back to the next slide? The traditional way. And so that's how we've been doing it about cutting costs.
SPEAKER_05I guess Justin and I right now are really crawling when it comes to AI and our utilization. We pretty much do the same exact thing with papers that we get and we throw them into AI for accuracy or for grammar. I mean, we have chat, co-pilot, grammarly perplexity. I think you're looking in the Claude also, right, Justin? Yeah. We have we have copilot. I think this is absolutely fascinating. And so you kind of walked us through that crawling, right? But what I really want to do is I want to get to that next step where we're running. And I guess that's not that's walking, then running. Justin and I, and here's some inside baseball. We have all this data. Justin has it on all his marketing efforts. I have it on all of our events and registration. That's so does Justin. But internally, you know, I've been wanting to put together an event analytic tool that would help us with locations, budgets, projections. I mean, you name it, right? From our perspective, how do we take that next step? Or how does someone take that next step where they are really looking into creating that tool, right? Is that too advanced? Is that more like a you're sprinting and that's not a walk, that's that's not a run, that's a sprint. What is that for the AI person?
Turning Event Data Into AI Tools
SPEAKER_02Well, I think it's about defining the problem and defining what you want to get out of it, right? I keep my AI very laser focused on whatever problem it is. So I may be running multiple different segments of a similar version of AI that has different prompts. You know, let's use your data on location. So you have, I don't know, a thousand members in a Kuma. I don't know if that's accurate, but round numbers. And they're all over the country. But you want to create a heat map that determines where you have the highest concentration of members. You can potentially determine where your next event is. And then you want to create something that's easy for these people to get to, but also as easy for most everyone else to get to that has the highest concentration of population, right? You can utilize AI to help you determine what that ideal geographical location will be. Now, it might say it's a field in the middle of Kansas, and then you've got to figure out what the next major city is, or you can prompt it to say what's the next, the best city for people to go into utilizing this data. If you want to talk about the best time of year to hold events based on location, this and that, again, it's all about prompt engineering to get the data, and it's also based on what you feed it. So I tell my team this all the time: garbage in, garbage out. If you give it a bunch of incomplete data, you're not going to get complete data on the back end. But if you have like complete documents with locations, what companies they work for, so it can start analyzing like what conference schedules look like and what most people with certain titles might be attending, it can start pinpointing dates that don't necessarily conflict theoretically, right? And it's about prompt engineering, feeding it the data, and then asking what's the output you want. Do you want a spreadsheet? Do you want a heat map? Do you want a bar chart, pie chart, illustration? That's not too advanced at all. And then once you build out the prompting, you can pretty much keep utilizing that same agent by new new chat threads. And so the best place to build those are going to be in the sandboxes or the playgrounds. And like in Copilot Studio as an example is a great place to start since you guys, I assume Akuma operates off Microsoft. You have Copilot already within your Microsoft framework. You should have access to a copilot studio if you've paid the money for it. You can build your own AI agents in there, and you can say this is our AI agent related to member location, right? Density of member location. This is our AI agent for location planning and management or whatever output or outcome you want to get. It's not too complicated at all.
SPEAKER_05All right, last question on this topic. If I created an agent, can I share it with my team?
SPEAKER_02It'll depend on the permissions you give on the back end. And as with all these things, everything costs money. So if your team has the licensing, then your team can access
Mortgage QC Agents In Copilot Studio
SPEAKER_02the agent. So I built a mortgage quality control agent that we run through all of the mortgages before they get to funding, they run everything gets run through this agent. So we have a separate document in the e-folder and encompass. Everything for the QC, anytime we upload a document, if the document's identified as something for the QC agent, it goes into a separate bucket. That bucket then integrates into a cloud drive. The cloud drive integrates to the agent. So we can pick all the data we want, upload it. It runs a QC analysis, tells us if we're good, if we're not good, what our risks are, and then we can go back in and fix it, or we just move the file forward. We I built that in co-pilot. It took less than a week to get that built and deployed. What it doesn't do is it doesn't give us decisions, again, because we we have a mandate from the board that AI can't make decisions for us, but it gives us the output. We validate the output, and then we move the file forward. What it's done is it's cut our QC time down significantly. It could be a couple hours per loan down to 15, 20 minutes. AI takes two minutes to go through everything, and then the human verification is what takes the edit.
SPEAKER_05I actually love that because like that was actually going to be the next question I asked. I've really hammered you with Acumat could personally use it. And then obviously you told us about at a bare minimum, have it review contracts, right, for speed. The next step would be basically that AI agent, right? Build out that agent to help streamline that QA process. Now, if that's the next step, right? And you're like, that's the next step. What would be considered the an advanced thing that a credit union can can really leverage for their success to kind of help them be successful?
Underwriting And Predictive Analytics Next
SPEAKER_02I would say advanced is the whole idea is improving. In a credit union space, obviously we're not for profit, right? But the more money we make, the more money we can give back to our membership, the more we can grow and everything else. So with that said, I would say the goal for us in lending is to reduce our cost to manufacture alone. And that means to do more with less headcount is really the is one of the bigger differentiators. So I would say look at AI underwriting as an option. Not to eliminate your underwriters, but to empower your underwriters to do more. So if an average underwriter does, I don't know, 40, 50, 60, 70 loans a month, depending on your platform and systems, or 100 loans a month, you might be able to empower underwriters to double their productivity utilizing AI because now AI can do all the work an underwriter can do in a fraction of the time. And now the underwriter just has to verify some things. Just like we'd randomly select certain loans to audit, the underwriter can do a sample size out of every analysis AI does. We're going to pick income and assets on this loan, and we're going to pick verifications. The next one, we're going to pick income again. You should always verify income. And then we're going to pick, you know, liability assessment or whatever or purchase contract for viewing. So that's where I think they can see a relatively big lift to just again increase productivity.
SPEAKER_05I feel like we could have this conversation all day. I wish I really could continue on with this conversation because it is absolutely fascinating. I mean this as genuinely as I possibly can. I've learned more in this little conversation that we've had, this 30-minute conversation that we've had of things that I can personally do and how I can do it and approach it than I have attending AI sessions and conferences, multiple AI sessions and conferences. I think that's I think the moral of the story is that is that this is easier than what people think.
SPEAKER_02It's just a matter of just doing it, right? It's a matter of just getting involved in it and just getting your feet wet. You know, and we haven't even, you know, we're talking about pretty basic stuff, Peter. I mean, we haven't even gotten into how you can utilize AI to analyze data, to generate predictive analytic models to determine relationship deepening opportunities based on information on a 1003 and cross-referencing it against current promotions and offers you have in your organization, or analyzing borrowers' or members' purchase patterns based upon the transactions in their accounts going back X amount of years to predict what they're going to do. And again, that's use using AI in a predictive analytics model. I mean, we're talking about low-hanging fruit here, but really I mean, we generated a predictive analytics dashboard for home equity to predict what our application count will be and our runoff will be. And it's been accurate within $100,000 since we deployed it.
SPEAKER_05Wow. That's awesome.
SPEAKER_02That's awesome.
SPEAKER_05Well, well, uh I I hate to do this because I would love to continue this conversation.
What Keeps Eric Going
SPEAKER_05But we have to start transitioning to the second segment. But before we do, I have to ask you the second question I ask everybody. And that second question is relatively simple, just like the first one. What keeps you going? You're just like everyone else, right? You have good days and bad days, but something keeps you pushing forward.
SPEAKER_02What is it? 100% my family. If I wasn't married with kids, you know, I might be sitting on a beach drinking the margarita working as a bartender. Hey, that's the dream, man.
SPEAKER_05That's the dream.
SPEAKER_02Life motivate me. And and of course, you know, the ults the ultimate desire to do something that leaves a lasting legacy. But what's the point of leaving a legacy without a family? So it's my kids and my wife, quite frankly. Uh aside from all the exciting, cool stuff there is to do, but it only matters if there's something, something to leave behind for it.
SPEAKER_05I love that. Well, well, Eric, thank you so much for sitting down with us and and really just dipping your toes into how credit unions can and should be using AI, speaking from your experience. This almost stands to reason that we need to have you back to continue this conversation at some point. Before we even get to that point where we're bringing you back on, we have to transition to the second
Dad Jokes And Final Takeaways
SPEAKER_05segment. And this second segment is where we do our dad jokes, the most fan-requested segment of dad jokes. So here's what we're gonna do: we're gonna go around the horn, you'll do your two dad jokes, Justin will do his two dad jokes, I'll do my two dad jokes, and then we'll wrap up. Now I asked you to bring a third, and the third one is really just left just in case one of them bombs. We have the third one as a backup, almost like a redemption joke, but that's what we'll do. Easy enough. Eric, please.
SPEAKER_02You go first. All right. This one might not classify as a dad joke, but I think it does. I asked my mortgage lender for a joke. He said you'll appreciate it over the next 30 years. Okay, okay, that's good. What kind of music do chiropractors like? Hip hop? Hip hop.
SPEAKER_03Hip hop. Oh, good, good, good.
SPEAKER_04All right, awesome, go. When does red mean go and green mean stop? When? When you're eating a watermelon.
SPEAKER_05I ignored it.
SPEAKER_04What happens when you throw a blue hat into the ocean? It becomes my hat? No, it gets wet. It gets wet.
SPEAKER_05Thank you. What's the opposite of artificial intelligence? Natural stupidity. I'm rather used that why do birds sing every morning? They don't go to work. I like that. Well, Eric, thank you very much for spending time with us. Obviously, that wraps up dad jokes, but I think more importantly, truly enjoyed this conversation. We will have you back. We want to continue it. And I think there's most certainly uh a case for credit unions to just listen to everything that you just said, and of course, what you will say about AI. So thank you very much. Really do appreciate it. Thanks for having me, Peter.
SPEAKER_02It was a pleasure, and I do look forward to coming back, chatting some more.
SPEAKER_05Yes, sir. Yes, sir. And Justin, as always, thank you very much. Of course, it is my pleasure. And to close out, thank you again to Zakdas for sponsoring today's episode. And to all of you, we know your time is valuable. Thanks for tuning in to the latest episode of Acuma's On Point Podcast. We hope you enjoyed it. Until next time, be well, my friend.
SPEAKER_00Thanks for listening. We'll see you next time at the Acuma on Point Podcast. If not already, be sure to subscribe and give us a five-star rating. For more great episodes and information, visit us online at Acuma.org. And to get the latest updates, head over to our LinkedIn page.
SPEAKER_01This podcast was edited by Resonate Recordings.
