EP 143: AI Innovation for Financial Regulators

This is the Everyday AI show, the everyday podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business and everyday life. How is AI innovation going to change things for financial regulators? It's something if you're not in the industry, you probably don't think about it a lot, but it's actually going to impact us a lot more than we think. And I'm excited to talk about that today on Everyday AI. This is your daily Livestream podcast, free daily newsletter helping everyday people like you and me not just learn generative AI and what's going on, but how we can all leverage it and understand how it impacts our daily lives. So if you're joining us live, as always, thank you for that. Make sure to get your questions in. We have a very special guest today that I'm excited, excited to have on the show. If you're listening on the podcast, don't worry, you can always come in and join the conversation after the fact checked out. Check out the show notes that we always include in the podcast. Come back, ask questions, interact after the fact. So let's, let's get started today with the AI news as we do every single day. Some, some big ones today. So how competitive do you think talent is for top AI execs? Well, according to a recent report, we're talking pro athlete type money. So according to a recent report from Insider, OpenAI is offering compensation packages of 5 to 10 million dollars to entice top Google AI researchers. That was not a misprint. I did not miscommunicate that. That is true. Five to ten million dollar compensation packages. So the company reportedly has already, OpenAI has reportedly already hired nearly a hundred people away from Google and Meta and job listings for OpenAI are showing salary ranges of anywhere from about a quarter million dollars to a half a million dollars a year for research engineers with additional benefits and equity. When I read that I was like whoa, okay. That's like actually more than some pro athlete money. All right, second piece of news for the day. Microsoft ST stock hit an all time high late Friday as it prepares for its Ignite conference this week. So Microsoft stock reaching all time highs. The company prepares to announce some pretty exciting artificial intelligence announcements at a conference that starts actually tomorrow. The Ignite conference starts tomorrow in Seattle. So. So what are they supposed to announce? They've already started rolling out their Microsoft 365 copilot which kind of drops generative AI across the operating system. But it has been widely reported, but not confirmed that Microsoft will unveil its own AI chip codenamed Athena. And Microsoft's new, reportedly new AI chip, Athena, will reduce their reliance on Nvidia and give them a little more control over their AI infrastructure. So some pretty noteworthy things going on in the world of AI, but that's probably not why you're here. You probably tuned in live or are listening on the podcast because you care about AI innovation for financial regulators. So that's what we're going to talk about. And I'm extremely excited to have on the show. And please help me welcome, we have Kyle Houptman, the vice chairman for the National Credit Union Administration. That's a government agency, y'. All. So, Kyle, thank you for joining us on the show.

Thank you, Jordan. I like listening to the news updates too.

Thank you. Yeah, big, big stuff happening. You know, I apparently I should have as a kid given up on my MBA dreams and just gone into generative AI earlier.

At the time I had NBA dreams too, but I didn't have to give them up. The world just gave them up for me. You know, I just didn't have the ability. But I use that expression too about NBA NFL money sometimes when talking about the more complex and harsh a regulatory scheme is, the higher the market value for former regulators. So if regulators are doing their job well, there shouldn't be a massive market to pay to navigate the system. That means the system's not that good, right? Anyway, you mentioned the pro athlete money and it was reminding me of that.

Oh yeah, same, same with me, Kyle. Something about I stopped growing in eighth grade. Didn't help. Yeah, but maybe, yeah, maybe tell us a little bit about, you know, your work at National Credit Union Administration. Just, just so people know. Because, yeah, it is confusing, but it's a government agency. So just give everyone a high level overview of that.

Short answer is if you know what the FDIC is for banks, that is essentially what NCUA is for America's 4600 credit unions. We are an insurer. If you have a bank account, if you have an account, a credit union, it's insured up to $250,000 per account, exactly the same as FDIC insurance. And we're also the regulator for about two thirds of the credit unions. For about a third of the credit unions in America, we are just their insurer, like FDIC is for banks. And then for two thirds, we are also their regulator.

So, you know, and I'm, I'm not even sure of this. I'm learning along live with everyone else but you know, how is the National Credit Union association or are you all using generative AI internally? Is that not, you know, are there too many kind of, you know, pieces of red tape to go through? I mean, how does that work in government agencies and using AI? Because it's obviously important, but there's a lot to regulate.

By the way, it's administration.

Yes. Oh, sorry.

Thank you, thank you. But that's the thing because there's a trade group that has similar letters and people confuse it with that. I am a government employee. And so anyway, I mean we're still in, you know, top of the first inning. If you don't follow baseball, that means the very beginning. But so I think the most obvious use for us as a regulator will be anti fraud fraud detention. I know we're talking about it, I don't know as of yet today we've actually launched it, but either way we're not going to talk that much about it. The same way that entities in general don't go into detail on their cyber defense and antivirus software. You know, you don't tell people exactly what you're doing. But, but that'll be thing one. But quicker than us using it as an, as a government agency are the entities that we regulate and ensure America's credit unions, they're already using it for a whole bunch of cool ways including their own fraud detection because you know, that's a major issue for them.

Yeah. And obviously, you know, in this, in the financial sector, you know, people, some people know, some people don't. But AI is not exactly new. It's, it's been used for, for many decades in the financial sector. But how the generative AI aspect and you know, when you start talking about large language models and things of that nature, does that change the game for financial regulators? You know, just broadly speaking, I think.

The main thing is credit unions and banks for that matter just have to do what they've always done, which is adapt. No credit union or bank in America can operate the way they did 30 years ago, which is no website, no online bill pay, no mobile app. Right. Had you been frozen in time or had your regulator made you frozen time or didn't say you couldn't use this thing called the Internet, but was really negative every time it came up. Right. Well, those people would all be gone and that regular insurer would be bad at their job. So I think you agree that chat GPT last November 30th. Right. We're coming up on 12 months now. That was a game changer because like you said, AI itself has existed for a while. I know the US government uses it in a bunch of levels and probably tons of ways I don't know of NSA uses it to monitor Al Qaeda cells and see if there's any patterns predicting terrorist attacks, that kind of thing, AI. But normal people sitting at home, middle class people and regular small businesses weren't using it because there wasn't a consumer useful front end this. So Chat GPT all of a sudden for free was something to play around with that regular normal Americans can use on their phone or on their desktop. And it wasn't just, you know, for the military and big business. This reminds me of the Internet itself was around for a while and you didn't have a front end, a browser or even aol. Right. So that could tap into the power of it. You know, like the Pentagon had said, other people were using the, the Internet, the actual piping. But then at some point, the 90s, it wasn't as a, it wasn't one day like Chat GPT, but this thing happened where we had front end browsers, you know, we had our Netscape browser or you had something like aol. And all of a sudden this technology that had been around for a while was now usable by everyday Americans and everybody around the world. And you know, if you're old enough, you couldn't go a week without getting like two of those AOL discs in your.

Oh, I remember those.

Yeah. But anyway, there's a front end and now it's explosive. We're talking about it. Right? Literally one year ago today, November 13, 2022. I don't think we would have had this conversation the same way because that was a, a bombshell. Chat GPT and I played around it, you know, as a regular. We write rags and we write guidance. That's all we do. Right. And I just to start, I, I just think anybody that has to write things, I just feel like it's eliminated writer's block. It, you can just say, all right, just spit something out and then hit refresh and try again. And you don't sit there with a blank word doc. You know, it's hard, you know how people procrastinate because they can't get started for that matter, important emails like, okay, how do I start? What do I do here? You know, I, I think it's super cool just internally, but I just threw a lot at you. Sorry, man.

No, no, I love it. You know, Kyle, one thing you mentioned there, that, that, that I want to dive a Little bit deeper in is you kind of talked about the Internet and, and kind of what that meant for the economy at large. So how do you think, you know, even if we're talking about it from a US perspective, right. How, how do you think that generative AI will impact the economy? You know, and kind of drawing on that parallel of what the, you know, what the Internet did for the US Economy, is it going to be similar where generative AI is really driving the economy forward?

Yeah, I think. Listen, all technology that's widespread, that is disruptive, it wouldn't be disruptive if it didn't have some negatives. And, but my view, you know, policymaker, and before I was at ncua, I worked at just in the US Senate, the bank committee and we had a broader purview there. Right. So if I just put on my American hat here, we have the potential to dominate it. We have the potential to get all the negatives, I'm sorry, all the positives. We're going to get the negatives no matter what. Okay. Every new technology has negatives. And people say, oh, I'm worried. Yep, yep, you probably should be worried. But what matters is how much of the upside we get. You know, I always say, you know, do you know there were zero plane crashes before we had plans?

Good point.

Yeah, there were none. You can read the entire Old Testament, New Testament. There's not one auto crash, there's not one Internet virus. And these are all negatives, right? We're going to get the negatives anyway and there's not a whole lot you can do about it. The question is, how much upside do you get? Like, and, and even the negatives can produce good jobs. For example, antivirus software everywhere in the world has to deal with Internet viruses. Everyone in the world has to deal with cyber attacks. Hey, it'd be great if that didn't happen, but it does. But only some places in the world get paid for a high paying cybersecurity industry and a high paying antivirus software industry. So, you know, with the Internet we capture the most upside. And I credit 1997, Bill Clinton, Al Gore, I think it was an important moment. They put five principles for this new thing called the Internet on WhiteHouse.gov a website that had only existed for two years and it was from, you know, from upon high, number one, the private sector shall lead, etc. And listen, we got all negatives, right? We lost all of our video rental jobs and, and travel agents and all that, right. But guess what? Canadians and Mexicans also lost Their video rental jobs, but they don't have Netflix headquarters, Hulu headquarters. The employees that got in early and got stock options and changed their life. The early investors that got huge returns, the endowment funds and pension funds that invested in those venture capital funds. The worst thing with disruption is to get the negative and not the positive. So I would like us to replicate our success with the Internet, not try to think we know it all and recognize that yes, there's going to be negatives. And the question is how much upside. And Americans certainly are positioned to capture more upside than any place else.

You know, I love, I love what you said there, Kyle, about the, you know. Yes, there were, there were zero plane crashes. You know, when you go, you know, hundreds of years, right.

Thousands of years without a single one.

Yeah, but I think, I think even when you talk about kind of the, the general perception of, of AI, it does seem that there's this kind of overwhelmingly negative perception. But you know, you just kind of likened it to the, the Internet. Right? So I'm, I'm, I'm curious from a financial regulation perspective, can you talk a little bit about the public perception of, of specifically generative AI? Are like, are people excited about it? Do they not understand or do, are they saying exactly what you're saying? This is our Internet moment to as a country, you know, leverage this technology and push our economy forward.

I think it's going to be like if you try to talk to, you know, a 12 year old or even somebody in high school right now, even the terms we use don't make any sense to them. Like how do you return a movie to Netflix, right? Like, even the words, it's not like we got a cheaper deal and you know, rewind it before you. How do you, what do you mean? How do you rewind it? What are you talking about? Like, the terms don't even make any sense anymore. I think we agree that a lot of things are going to be like that, like just the way we speak about these things. And so there's already been AI, credit unions, for example, online lending. Right. Doing more advanced underwriting than just using your FICO school for some, some really, really neat stuff. And that's already been happening. But the one advantage that I think AI has over another newish industry, blockchain, digital assets, crypto, is that regular people are going to use things like chat, GBT and make their life easier. So you can't just say it's all negative. You can say there's negative aspects. That was an advantage the Internet had Your first White House staffers, etc that had webpage and oh my gosh, email. I can hit a button and you get it across the country. You can read the news without having to go find a physical newspaper. Like Internet had problems. There's all kinds of crime on the Internet. There's illegal pornography, there's all kinds of stuff. But you couldn't just say this thing is bad because you knew all these neat things, right? You were able to say, yes, cars get in accidents, but cars also get you to the hospital faster. People used to die before because they couldn't get there, right? So normal people saying, whoa, you know, it just did some draft paper for me and it created some art and it can be aware of state law in all 50 states seven days a week, which comes up in my business a lot. Right? Super freaking useful. So the worry about it. Yes. And just say we want to have responsible development. Okay, yeah, but like, you know, the, the deep fakes and it's again like the Internet, same problems, but more aggressive and more hardcore. Like false advertising was illegal before the Internet. Fraud was illegal before the Internet. Impersonation, stealing identities was illegal for the Internet. The Internet created whole new ways to do those things. And I think you agree, Jordan, AI is going to have lots of ways for criminals going to create a whole bunch of new problems. But what do we know about problems that are really harsh and really expensive? We know you can get paid really well if you come to the table with solutions.

Are you still running in circles trying to figure out how to actually grow your business with AI? Maybe your company has been tinkering with large language models for a year or more, but can't really get traction to find ROI on gen AI. Hey, this is Jordan Wilson, host of this very podcast. Companies like Adobe, Microsoft and Nvidia have partnered with us because they trust our expertise in educating the masses around generative AI to get ahead. And some of the most innovative companies in the country hire us to help with their AI strategy and to train hundreds of their employees on how to use Gen AI. So whether you're looking for ChatGPT training for thousands or just need help building your front end AI strategy, you can partner with us too. Just like some of the biggest companies in the world do. Go to your everyday AI.com partner to get in contact with our team or you can just click on the partner section of our website. We'll help you stop running in those AI circles and help get your team ahead and build a straight path to ROI on gen.

Yeah, yeah.

Geez. As, as we talked about at the top of the show, you know, we got, we have these jobs now at, you know, OpenAI, you know, quarter million, half million dollars. Yeah.

When somebody calls. Yeah, sorry, you know, like deep fakes when somebody calls up and it is your voice, it really is, with your accent and your intonation and it can pass those voice detection, it can literally pass it, you know, you can speak to a relative and they think it's you. That's a major issue. Yeah. And people who can prevent present solutions to that. Those are going to be good jobs, let me tell you that.

Yeah, it sounds like, you know, there's always been these, these positions that, that we've talked about, you know, in anti fraud, cyber security. But Kyle, like, like even specifically as we talk about financial regulatory institutions, you know, credit unions, everything, are those jobs just going to like explode in, in popularity and in, you know, more, more and more, you know, credit unions needing these.

Well, I think we agree that we're going to look back like we do, like talking to that 12 year old kid. Right. They don't understand how much harder things used to be that getting lost and taking the wrong turn and trying to figure out people missed appointments. Right. You didn't have this thing called GPS to try to communicate with people. You know, like when I call my wife from the grocery store and I say, see this thing in my hand right here? Is this what you want me to get? Right. And we don't have to have these screw ups and trying to book travel like just monumentally harder. So many things in a young person's life just were not possible. You know when you say, oh yeah, I talked to Jordan this morning. Oh, you did? Okay. Really? When I was a kid, you had to physically be near a certain phone. Not any phone, one certain phone. And you had to be on one end for us to, you know, to talk to each other. I think it's going to be like that. And you're going to wonder, oh man. Remember the old days? That was a nightmare. Right. I remember as a little kid going through the yellow pages trying to look up airlines and then calling you me and waiting on, you know, do you fly to Denver? Yes. And writing it down. Fraking nightmare. So there's no way, I think we're going to look back at pre AI and say, wow, that was better. The same way, you know, you don't want to just not have smartphones anymore and get rid of the Internet. We wouldn't even know how to function.

Yeah, yeah, Please, please Don't take any of that away from me right now.

Yeah, anybody that complains about innovation and gets worried, I'm like, okay, lose your smartphone for a day, see what happens. Right? It's a nightmare. We can't even function right. And it's to me the same. So people are going to wish that there were aspects of AI that were gone. Yeah, that's true. We wish that there weren't Internet viruses, that we could have Internet without cyber attacks, without hacks. We, we wish we could. But from a policy make perspective, we're going to have those things and we can't stop it. We can't. Everywhere in the world has car accidents. Only a few countries get a high paying auto industry and we're one of them. Okay? And every country in the world has computer viruses, cyber attacks. Only some places have a good cyber defense, cyber security industry. So everywhere in the world is going to have negatives with AI. Everywhere in the world's financial institutions are going to have negatives with AI. Only some places are going to get all the benefit of it. And I think America certainly can be if we're not too heavy handed on it. But credit unions use it for customer service. Do you know what's already solving language issues? Like you, you have a new immigrant community. You know, like I live in D.C. right. There's a large Ethiopian community. If I ran a credit union, and I don't necessarily, and I want to have stuff in the language and speak to them, it just got so much easier. As you can imagine. Last year I spent a fair bit of time talking to the Ukrainian American credit unions. There are 11 Ukrainian American credit unions. They told me 40 of their phone calls are in the Ukrainian language. Jordan, I'm sure you can understand it's hard enough to find good customer service people. It's even harder to find ones that are fluent Ukrainian and English. Right. So a lot of that just got easier, didn't it? Where you can have, actually speak and write documents and have chat bots that aren't terrible. You know, the opportunity to be served in the financial system, the opportunity to serve somebody, serve them well in the financial system just got easier because all of our entities are now fluent in Ukrainian and Mandarin Chinese and everything. And I think it's great for financial inclusion.

Yeah, that's, that's a great point because we've kind of been focusing on how generative AI may help the industry or help credit unions. But I mean when we talk down to the, to the individuals, it sounds like it's, it's it's going to make things a lot better for the individual. Correct?

Yeah. You know, in a free enterprise economy in which we still, you know, thank God largely are, technology usually lifts up the have nots and makes theirs. If you just add technology to a status sort of, you know, low free enterprise society, then it just enriches the elites. But in a society like ours, normally it helps to have nuts. For example, I've had two in my life. Deaf ride share drivers, Uber drivers, it's not a big deal. You get in, you put us on. And if you know how ride share works, you put in your destination, whatever. And for that matter, I've taken other countries where I don't speak the language, they don't speak English. I mean, it's the same thing. I've never had a deaf cab driver and you can't even get a license. So somebody with a disability or now has a way to earn their keep. And people are happier when they earn their success. Most people don't want to be, you know, very few people I think are actually moochers. The system may make you one, but people in wheelchairs, ebay. Meg Whitman used to always say this. How many people made their primary living buying and selling on ebay? Technology has always helped less fortunate. If you were deaf, there were a lot of jobs closed off to you and you now have it. A taxi drive. You could never get a taxi license. Viewer death, right? It's made it so, so much easier and it's made it easier for us to get around. The wealthy never had this problem. They had chauffeurs. The financial industry, right? How easy it is to access good investments. Now you can get low cost index funds sitting on the sofa, just with your thumb on your phone. You got, you got your paycheck, you want to invest, boom, you can do it. The wealthy always had concierge financial services. So this is going to allow people to do things. You might not have tons of very high priced lawyers or be able to find one. And even if you do, they don't work 24 hours a day like AI does. And, and it's aware of the latest developments. You know, I think it's going to be the same way that we now have. We're going to have services that only the elite had before and it's going to benefit it on the service side. But if those of us in policy making rules don't screw this up, will also get the other two benefits, the jobs and the investment. Everybody in the world gets to watch streaming Media. Only some countries got the jobs and the investment that came with it, you know, so that you could tell. It's a theme I keep coming back to, but this country obviously has an opportunity to just win and win big with the AI industry.

I love that. Yeah, that's, that's getting me pumped up. Right. Because we, we always talk about, hey, everyone can use it. It's, it's, it's creating so many new sectors. But yeah, those sectors, sectors and jobs for the most part are, are here largely. Which, Kyle, is, is a great point to bring up. We could go on for a very long time. Kyle, I want to be respectful of your time, but let's, let's maybe focus on what is the one key takeaway as we wrap up today's show that you want people to understand about how maybe, you know, credit unions, financial regulators are, are using generative AI and what that means for the rest of us. Right. So what does it mean for us all? And what do you kind of see as the future of generative AI in this space, you know, hopefully pushing this whole sector forward?

Well, again, to go back to the Internet analogy, the core of traditional finance has not. Didn't change how they did. Those things change. The core of banking has not changed since it started in the Middle east thousands of years ago. People make some money, they deposit it, they withdraw some, that money is lent out, most of the time is paid back. That's the core of it right there. Okay. Money moving around, being lent out, being paid back, etc. That hasn't changed in a long time. How we do it has changed. So the, for a regular person out there, the two things that jump to mind, right, put aside inside a credit union, but from a customer of a bank or member of a credit union, you're lending decisions are going to get a little more complex, but probably quicker. Meaning you don't just have to look at your credit score. People complain about fico and I understand it, but I've been in countries that don't have something like that and believe me, FICO is better. But what you post on social media, depending on what they use, you're going to find lending risks that are better than people thought. Like finally being able to get credit for the fact that I've paid like 320 consecutive monthly cell phone bills since 1998 and I don't get any credit in my FICO score because it's not technically credit. Right. The fact that I paid rent for 20 years. Right. That kind of stuff too. But obviously you know, there can be negatives as well. You might look like you have good credit, but dig under it. So the lending decisions are going to get more complex but better. Remember, we don't want to be in a country that misallocates credit. We had a version of that 2008, the housing crisis. So more accurate, faster but different who gets lent what money at what rate. Okay, so I think that would be a key one. The other thing will be customer service. You're going to be able to get forms in your language. You're going to get information that's up to date. So again, we regulate credit unions, you know, in all 50 states and many of them are state chartered. So they're regulators of state. You know, I don't know what happened in Oklahoma this weekend. There was some enforcement case, right. But good AI Regtech, you know, should know. So customer service and lending decisions, what I would say for a regular person, that's where you're going to see the difference. You're going to get money at a certain rate and for that matter, your money in the bank or credit union is going to be lent out in a certain way in a different way, more complex, but probably faster and fairer than it used to be.

So many, so many great insights. We covered so much in a very short period of time. Thank you so much, Kyle for coming on the Everyday AI show and and give giving us some great insights on AI innovation for financial regulators. Thank you so much for joining the show.

Thanks Jordan.

Hey and as a reminder everyone, we did cover a lot. If you miss any, don't worry, go to your everyday AI.com Sign up for the free daily newsletter. We're going to be recapping today's show, sharing out a lot more information as we do always do. So thank you for joining us and we hope to see you back for another episode of Everyday AI. Thanks y'.

All.

And that's a wrap for today's edition of Everyday AI. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. It helps keep us going for a little more AI magic. Visit your everydayai.com and sign up to our daily newsletter so you don't get left behind. Go break some barriers and we'll see you next time.