Episode #23 with John Sanders

Hello, today I'd like to welcome John Sanders to the Violet Group report. John and I have known each other probably for a couple of decades. Really one of my go-to guys when I think about financial services. John has had a lot of success in the industry. He is a co-founder, a CEO and a managing partner of Bridge Force. It's a financial services consulting firm working here, not only in the US, Europe, a global reach. And I was very excited to have John join us to get his view on financial services, a specifically FinTech, and some of the technologies and trends that are shaping the industry, including AI. John, welcome today. Glad to have you. I'd like to just have you tell us a bit more about the journey you've taken. Founding Bridge Force. I think back in 2000 and where things stand today. So John, welcome. Glad to have you. Well, thanks Rich. Real pleasure to be here. I appreciate the invitation. So that was a wonderful introduction regarding our business. As you mentioned, I've spent 25 plus years at really the intersection of data, analytics and consumer lending. My firm Bridge Force helps financial institutions of all shapes and sizes. So clients include global financial institutions, banks, credit unions, FinTechs, non-banks, and specialty finance firms. And we really work to solve a number of problems around credit decisioning operations, compliance and performance improvement. I am one of the co-founders of the business. And I guess my journey that led me to starting the company was that in my travels prior to Bridge Force, I had a number of large advisory firms and consultancies work for me. And can you really I saw some really good talented folks that I liked and then I also saw some things that I didn't like. And one of the things that I wanted to do and was motivated to start Bridge Force is to bring people that really come from industry and have that depth that can help our clients turn data into better business outcomes. So that was the journey that got me here and I'm sure as you can appreciate, there's a number of organizations back then where there's the business and strategy component and then there's the technology and automation component. And trying to connect those two or Bridge that gap was a bit of a tricky proposition. And we always thought that we did quite well in that space. So we tried to bridge or connect that gap and set the metaphor for the name Bridge Force. As I fast forward to today, it's interesting, I think the metaphor could not be more relevant to the event that was back then. That's really great. And a lot of our followers are entrepreneurs, some at a very early stage of the companies, others that are going through a growth stage. And with the experience you had, when did you realize or see that you really had something that and felt you had, this thing's going to work. What was there an event or a time? Because it might be helpful for some of our followers to recognize the path you've gone down. And maybe it was something that you already passed that milestone didn't realize it till a few months or years passed. But what was that moment if you could think back to that? Well, I could spend a lot of time talking about the early days as an entrepreneur. I think for us, since you mentioned that AI is a topic of interest, I worked within the walls of MBNA, the large credit card monoline back in the 90s. And actually my first exposure to AI or machine learning was through that organization, a company called HNC, Alan San Diego had a technology that they wanted to find another use case floor and he brought it into banking. So MBNA, at the time, was really a beta client. And I was involved with that initiative that eventually became an industry standard solution called Falcon. So fraud, falcon, and that eventually was acquired by Fair Isaac. So having that experience, the person with whom I worked, she also, Dawn Willey, co-found the company. Actually she was the founding member. And at that point, once we had hung a single, we didn't look back. In terms of a defining moment, I think, for entrepreneurs out there this thing, I always encourage you to just hang in there because the early years are usually just survive. So for us, by the time we got traction, year three, four, we built our brand and the phone started to ring our direction. That was a, that's what I knew we were going to be good. That's a great feeling. And I think you really hit on something very simple, almost, I'd say profound, this bridging of the business, the technology. And what you were doing back then, you know, call it machine learning, it's, I think, following the same fundamentals. And what you bring, I think, in your team brings is an understanding of the business and the technology. You just don't plug in an AI or some of these other models. You need to understand what you want out of it. And how does it kind of land in your business? How does it become if you're operating processes? So it's more than, and I think as you work with your clients, more than just working with the IT guys, okay, you know, plug this, this application and it really has to be understood and incorporated by the whole business. But you know, some simple thing that bridging, that connecting and helping, you're a bit of, I guess if you were, you know, kind of a Sherpa guiding your clients through that. What's the business look like today? The range of services, the key issues you're working with and what kind of clients do you work with today? Sure. So we are, we've been in a juxtaposition because we are laser focused and all things related to consumer lending. We advise across all asset classes, auto, credit card, mortgage, home equity, you name it, we do that across the entire credit life cycle. So if you think about a credit life cycle from marketing end to end to underwriting, once a council and book portfolio risk management, all the back end operations, collections and recovery, we play in all those spaces. So while laser focused and consumer lending, we actually once there, cast an app, pre-defined wide. And, and that ebbs and flows, we do a lot of things in between as I mentioned fraud that sort of never goes away. A lot of compliance and regulatory work. And we do a lot of work over the past several years and all things, fair credit reporting. So FCRA, this force is very entrepreneurial. We have incubated a number of handfuls, I'm saying number, but a handful of organizations that we've spun off outside of Bridge Force and one is a sister company called Bridge Force Data Solutions. So in terms of what's hot that tends to end of flow with truly macro trends, as you know, I've done a lot of work broadly speaking in all things risk management. And that's really our home turf. In fact, that was a tagline of our company. We used to, we changed it since, but the focus on our experience. But we had a tagline that said turning risk into opportunity. And I think today a lot of the, the fundamentals are still very relevant, but say you can look at risk management in a couple buckets. There's macro challenges. There's operational challenges. We mentioned AI, right? So those challenges are now new and relevant and then some organizational challenges as well. I'm happy to go into any one of those to whatever that would you like. But I think, you know, if I was to put all that together, risk management isn't about avoiding losses. It's about choosing the right losses and pricing for that risk. Right. You could eliminate, have zero losses, but you'd have zero business. So what is that trade off? And an iterative process as well, because it's not static. You just don't have a, an environment that's, it's influenced by the actions you take and by the macro economies you work with. And looking at, at your business, John, is it, are you dealing primarily with the US, European clients? I'm just interested in if you've done, if there's been much work in some of the emerging economies, a lot of challenges just to get to the data. So maybe I let you answer a question about the client said and then talk a little bit about how you, how you help clients that are maybe entering a new market where there's not an established or rating agency or credit bureaus. So to speak, how do you help them, you know, build out their product and develop that market? Sure. So our clients, again, some mention are all shapes and sizes. Right now we have sort of a nice mix of actually less than combant banks and a nice mix of smaller institutions, a lot of credit unions, which I find to be quite an interesting phenomenon and is mentioned Fintechs. In terms of the geography, mostly the work is, you know, stateside, but certainly we do a lot in the UK. We've been present in the UK market for 24, the 25 years that we've been in existence. And that becomes a springboard for other markets. It's a little bit more opportunistic to be honest because we have so much to get this busy between the UK and the US by how we help organizations think about entering a new market or introduce a new product. We've done a lot of that work over the years and this probably will sound like a very underwhelming answer, but it's really taken a step back and think about what you want that business to look like in one, three, five years out. So we take a mindset of starting with the end of mind, which the target operating model to look like and again crosses people, process technology. And you know, even in thinking about that, we've seen organizations just jump into the deep end of the pool and they really don't have a good blueprint or a roadmap. So we like to sort of step back and say, okay, what design principles must be true? And then I built out the plan appropriately. I think we've certainly been working with FinTechs before, long before everyone consider FinTechs, sorry, they were so in error where they were sort of defining their identity or their marketplace lenders or their alternative lenders. And now that's obviously evolved. Their motivation, 12 years ago was largely growth and now they've evolved to a level where you have to have the right compliance controls and other frameworks in place to manage risk. So I would argue that probably every organization now, especially with AI is becoming a FinTech. Yeah, let's jump into the AI and just how you're reading it, your view and like I said, I really like the thought, you know, back with H&C in the felt. I mean, machine learning's been going on for a while and we've had to some degree, you know, that's what our programs have been and our apps have had, you know, some kind of AI, but it's now evolving to it. But what are you seeing with your clients is you think about AI, are there certain caveats you have in mind? A client comes to you, you know, things that say, this is watch out or misconceptions, you need to correct. I kind of, your approach is you as you work with your clients with AI. What's the landscape you try to lay out for them? It's a great question. So we could go really far and wide on this. I would start out with how we conceptualize AI and my philosophy and you sort of hate on it in your ordinary marks. I don't view AI as a product to buy. I don't view it as a single platform. I actually look at AI as a new way of thinking about work as we mentioned. AI has been around in a different form through machine learning and neural network capabilities for three decades, plus. So to us, this is just the natural evolution of those capabilities. And I also have a philosophy that it's not about augmenting, it's not about replacing humans, I should say, it's about augmenting human judgment. And our objective, Rich Force, is I'm not trying to be the expert in all things AI. My objective is to help our clients use AI effectively. And right now, Canada leader is a lot of AI to smoke, being blown around the industry. So our job is to help our clients see through it. I've seen or I've heard some interesting remarks of how others view AI and one of the best analogies I've heard is that kind of like fire. So without a doubt, this is a watershed moment. And one person without, I'm hopefully over-demanded to dramatizing it, said it's a right up there with humans discovering fire. Well, fire can eat your home, it can cook your food, but it can also burn their house down. So I think our view is AI has to be respected. It's interesting to see how our clients are starting to embark on that journey of imply AI in various use cases. But that's, I think for us AI may feel new to many people, but in financial services, as you know, data driven decisioning has been around for decades. Right, absolutely. I think of it. I like that analogy and fire and how you channel it or put it to work. And you have to put it towards a goal. I think at a high level, I've been thinking about using AI to advance some business goals. How do you compete? One example may be you compete on providing access. It may be an innovative service that, you know, it's serving the unbanked, haven't had access to banking services. And thought like how AI might help you better manage that risk. Because it's the unknowns you're going to lend to a population that you don't know much about or you're trying to separate the very risky from the less risky and price accordingly. And just to cover it out, I thought in terms of AI to help you with this access call it or maybe it helps you provide convenience. It's onboarding somebody with three clicks rather than ten. And then it's maybe a matter of efficiency that you compete on price and will AI help you take costs out, you know, simplifying some processes so you have an advantage. And I don't know if that helps, but I've been playing around with that construct and just trying again to find ways to put the dodge together. Because right now it's kind of like that's on a piece of paper and what do you see? And I think you've got a great vantage point again with the breadth of clients you have but also being able to go back for 25 years and plot out that trend line. Yeah, there's a couple of themes there that you hit on. I mean, certainly there's a lot of strengths of AI. There's certainly some weaknesses, but you know, it's just speed is definitely a strength. Pattern recognition has always been a strength, scalability, consistency, handling unstructured data. Those things we're seeing AI, just some really good things and neat things. But the weak part about it and financial services thus far is the explainability, right? Potentially bias risk, data drift, hallucinations. So our clients are sort of a, it's funny. Being on the other side of the desk, I never cared for consultants that would use consultant jargon, you know, best practices, right? I mean, there might be best practices, but it may not be best fit for someone. I'm hearing it on the other side of the desk everyone's talking about reimagining their world, which is the buzzword. But I think that, again, financial services, AI, anything that touches an agent or an end customer is something that is, it's powerful, but there's a lot of risk that comes with it. And I think right now for us, I'm having two different sting conversations around AI with clients. As AI is science fiction and one is AI is workflow improvement. I live in the second world. I also think you use an example about really credit, underwriting and decisioning them. And we are seeing that AI is being used to evaluate particularly alternative data, cash road, bank transaction data, even device data. I think it's certainly been used in fraud for it on time, but it's going to, and that's probably one of the clearest RLI's today, because fraud moves at machine speed. And so the fences now have to move at machine speed and continue to move at machine speed. I am seeing AI in marketing and personalization. We traditionally didn't go a lot of work in community banks because community banks are really the wrong animal where I live. If I drive 40 minutes in one direction, I'm in the middle of the land. So the Eastern Shore Maryland has one set of distinct characteristics. If I drive 40 minutes the other direction, I'm in Lancaster, Pennsylvania, and that's a lot of farming. So when you think about AI and being able to offer very targeted, you know, credit line management solutions or retention models, there's some relevance there. And there's a lot of back office automation that's being cared for, leveraging AI. And certainly customer service and operations. So it's here. And I think that organizations that are trying to lift these solutions, another great analogy I heard is think of it as sort of like a dog park. So if you go to a dog park, you can let your dog run around, try things, experiment, you're still inside a fence that you can see. Right. And maybe we mentioned one thing I'd like to delve a little bit deeper is the biases and data. And how do you manage that? Because in some ways it's only good as the data you put in and you have to be careful about, you know, is it biased data? Is it just wrong data? And which is an issue, it doesn't matter whether it's AI or not, but what do you say to all you know garbage in garbage out if you don't have a good data, so as you're just making bad decisions more quickly? So it's a great line. So it can, it's a big cautionary line, right? It can be very good at getting you to the wrong answers fast. I don't purport to be, I know it's to be dangerous about modeling. I'm not a, the deepest of modeling jobs that we have colleagues who have deep modeling jobs. But I think at a high level, we're seeing this transition from rules based deterministic models. You know, basically if you ask a model, the same question, broadly speaking, you should get the same outcome or you know, close to it, right? Same, same input, same output. But with AI, it changes more to an outcome based scenario. So it's probabilistic. It's contextual. It, it, it, it gave you different outputs if you ask, get the same question and likely should. So that's the challenge. And I don't know if that challenge is fully being solved for you yet. But we just came through a decade of, you know, our industry, heavy compliance and regulatory world. And I'm, we're, we're always going to be regulated. But the operating word is evidence. And how do you evidence consistent, compliant use of AI when those answers can vary? So how, how we approach that is some of the old fundamental constructs hold true. It's its model governance, it's auto trails, it's, it's explainability, it's controls. So I think as though folks begin to think about AI freedoms, if you will, it still needs to live inside of regulatory discipline. And I don't know if you've started, there was a, a, a, a caught, um, you can trigger another thought, Rich. Um, I think it was Stanford had a, an interesting article related to, or study related to prompting. And one of the most interesting things that, um, that, that I see is if you think about a AI, and you can even do this with your own sort of chat GBT or co-pilot, it's, it's trained to be helpful, harmless and safe. That kind of leads to very polite, very conversational and sometimes very boring answers. So the bias is already sort of there. It's not just an AI, but it's in humans because we as humans, we tend to want familiar answers or conventional answers or what have you. And so, uh, Stanford's paper and, and I'm going off a memory here, but, uh, it suggested a different prompting. So instead of saying, tell me a joke about coffee, try, try it differently, generate five jokes about coffee with their proud, proud abilities. So when you do that, you're breaking the AI out of the most typical answer mode. And uh, I've been using that myself just to sort of, you know, see, see how I response. I like that. That's really, there's so much to this that is there's a, there's a, there's a higher, I'm calling it a hard science to it and, and, and, and a social science aspect to it. And I think about is you look at your clients, you know, as a first, uh, jump into AI, is there a, is there a certain kind of, of application that is a problem that might be a good solution? Maybe more of a workflow, uh, efficiency type, uh, what, maybe the, the dog park fence is a little tighter, you know, not as much, uh, but you need to have practice with this. And I think it's, it's an iterative process. It's, um, and it's not just one variable. It's not just, and here's a question put, put data into it. But you're, you're really opening my mind up to the, you know, the number of variables and how you frame the question and you bring your own bias in. I'm sorry, I'm rambling a bit here, but I'm thinking maybe the, the better question, it's just, you know, where, where, where would you recommend a client starting, uh, to, to, to get comfortable and familiar with, uh, AI as part of their, uh, capability for their business? Sure. Um, well, there's so specific use cases that I know our clients are, are working one real time. Uh, but I, I have, literally, I'm still in the, uh, the, the, the throws of a number of C suite conversations and conversations with our clients with all different, uh, and just take over throughout and broadly speaking, uh, we just take it up a level first. There's a sort of, uh, three, I'm calling him phased approach, uh, but some of our clients have actually put timetables to this. And, uh, the, the overarching theme that I'm hearing is that this is the year of urgency. So 2026 is all about readiness and thinking through what sort of the, the game plan is, uh, at the enterprise level, uh, many organizations feel like if they don't start now, the train will leave the station. And I, I actually, you know, subscribe to that, that feeling. I'm the second phase, which might be a next year, a thing is to get into, again, more experimentation. So in after, after the sort of the, the strategy is set, uh, there's going to be a period of evaluating the tools, testing use cases and then building out basic infrastructure. And then again, these are, we're playing what, what I'm hearing from our clients and then sort of that, that third step is going to be all around scale and optimization. So just like every other technology wave, there's going to be a phase of, you know, real, real integration, operational integration. And they continue to optimization. And what we haven't really talked about, you hit on it, but what I don't see our clients to always talk about is what's the, what's the RLI? And we put everything through an RLI lens. So, uh, before you decide what thing you want to start with, um, it's, it's really just typical sort of, like any other investment you judge it by cost, benefit, risk, sustainability. And what I'm encouraging our clients to think about is just because AI can do something, doesn't mean it should. I mean, if it can, if it's a three-hour sustain something up and it saves you a one hour, that might not be the answer. So I actually think the winning organizations won't be the ones with the most AI. This might be a blasphemous statement three years from now. But I think it's going to be the ones that have the best combination of AI and human judgment and really put a good plan together. I would tell you, the lowest hanging fruit to answer your question very pointedly is that we're seeing a lot of AI being leveraged in, uh, sort of, uh, uh, uh, uh, text-to-chat communications. That's, that's, uh, there's a lot of lift there and it's, it's fairly low risk. It's a broad risk, so there's, I thought, as you go through all this, I think about the factor we need to consider is, is permissions that with the various stakeholders in the company, as you look to implement AI, uh, kind of permissions that'll need to be granted or maybe it's, requirements you'll have to meet, you know, investors are going to want that ROI. But I think about your, your clients, your customers and, uh, having their permission to take this data about them and in some ways maybe probe more deeply into them. I think you need to show, uh, to both, uh, all of those stakeholder groups that, uh, you're managing the risk. Um, I think about, you know, privacy data, uh, concern among consumers that, you know, this AI is going to, you know, get into my, get into my life, get into my phone, get into my, my laptop and just know everything about me, um, and, and there's, uh, having to manage the perception versus reality. And while we're still really figuring out what the reality is, but, uh, I think, I think that's maybe a softer side of it, you know, how, uh, you know, making the people feel, and I think it's a winning competitive that, that even feel confident that boy, this company, this Bridge Force client really understands how to use it. I like it. I'm going to work with them more because, uh, it's making my interaction with the company more, uh, more rewarding, you know, more satisfying. I would like to think so. Hopefully, uh, I mean, for us, and we, we, we're working on it, AI internally, we leverage AI ourselves to be more, more efficient. It's not just about, you know, organizations putting 2000 co-pilot licenses on everyone's desk. It's, it's really taken a step back and thinking about, you know, what's, what's the problem or trial I'm trying to solve for? And, and we've always approached these types of efforts as operators, not as theorists. So, start with the process, look at the data, you look at the outcomes, and then you figure out where AI fits in. I also contrast that against some of the other, uh, advisory firms, uh, there's a lot of consultancies that have jumped straight into, you know, the, you know, then all things AI, building branded AI, platforms and tools. But, uh, and is the fundamental question, you know, skit, which is, you know, I would clearly define the problem we're actually trying to solve. And I think without that clarity, AI, he's at risk of just becoming an expensive science project. So, it's not my conjecture. Actually, there's a, there's been an MIT research paper of some sort of position paper that's been swirling around in the financial sector for the past month or so. And, uh, it speaks to exactly that that at the enterprise level, the success of lifting AI is very low. I'm a high single digit, maybe 10 percent, if I remember correctly. But if you do the things that I mentioned, which is, you know, understand, you know, start with the end of mind, understand the requisition points, understand the data, then that success level goes up significantly. And I think the final component of that is just, you know, understanding, you know, that, you know, this AI is transformative, it's powerful, but yeah, it's, it's, uh, it's not completely magic. That's right. Right. And really, I think that discipline you've had, the risk management in the lending space really lends it well in terms of a methodology because it has a discipline of, of, uh, you know, an iterative process, you test, you measure, adjust, uh, think about the champion challenger types of models and, you know, result in, you know, scoring a population who's performing better and, and providing a feedback loop. And you actually have to land this technology. It's, I see a lot of times some grades for geeks, your consulting firm has come up with a, you know, a strategy. It could be AI, it could be other things, but it, it, it, it, it, it, it, it, it, it, it, okay, here it is. Just go do it. But as soon as you, you enter, you change things and it's, uh, I think about the Mike Tyson quote, you know, everybody has a strategy and they go into the ring until you get the first, until you get for, you know, for the first punch in your head. And then you've got to adjust, I don't mean to be so violent, but I think that what, what you bring, John and Bridgeforce this, uh, and, and maybe other firms that have worked with us, but I think that's the kind of approach. I think AI is looking for a practical methodology and it's not just one, but how do you deploy it and find a way to get past the superficial of it's going to write, you know, I think right in my paper or write this email, uh, to something that's, you know, can it, I think you have to put it through, I like that ROI model, put it through, and that's a discipline where you have to quantify it enough, uh, and then recognize what's the gap you're going to have to bridge and, and then come back to it and take, uh, take some small steps, but do it in a way that you make sure you're learning, uh, and then you can, you can build on that. Um, yes, so maybe it's, it's as much, you know, that methodology as it is the technology. Uh, I don't think you're here enough about methodology. Well, and to corroborate your point, I mean, we're seeing right now, organizations that are trying to put a sort of a, most of the centralized enterprise level AI sort of program in place, uh, up till now, I think like anything, you might find pockets of organizations that champion parts of AI and that's where it gets tricky. I think the Holy Grail would be to have some kind of agnostic AI platform and as other AI solutions that haven't even been invented yet. Come on nine, uh, those make their way into this agnostic sort of framework, but, you know, it'd be interesting to see where this goes because, you know, historically, technology, you might have point solutions or vendor solutions. And we are seeing some signs of folks getting it right. I mean, there's regional banks, they're modernizing their legacy processes and using AI in certain, certain use cases, as I mentioned, uh, FinTechs are using AI for more narrow, measurable problems from what I see. And then, you know, where you can integrate AI into existing workflows, that's, that's not a bad thing. We, we actually, before it was AI, we were using, uh, RPA, right? Um, you know, robotic, uh, you know, automation, um, back to fraud, for example, uh, use that very, very specifically. There was a, uh, it takes just the same amount of time to work up a fraud case, you know, over, you know, over, say, it's a $2,000 claim versus low within $50. And we've had some, some of our clients just draw a line and say, you know, arbitrarily, anything under a certain threshold is not worth our time to, to, to check. But using RPA, uh, the, uh, the ability to, to look at those claims, we were able to get in year ROI of seven figures. So I, again, go back to thinking of, uh, AI is, uh, as a capability, not a product. I think those that are missing the point, if I can be a little bold, are those firms that are chasing AI press releases, you know, we're seeing a lot of vendors slap AI on old rules engines and it's not having a AI. Um, but there's a company that are maybe it's just market marketing and branding right now and we'll get there, but they're putting, they don't have a data foundation. Um, and I think I've talked to a couple executives that, uh, yeah, are expecting, you know, the magic of AI without being meaningful process change and, you know, who, who, who will be in that organization? Who, who's going to be the one to shake those trees? And so that's why, if for us, uh, in other everyone heard that will maybe, you know, external consultancies will, will, will be extinguished because of AI, I, I, the demand is still there. Maybe the delivery mount will change, but back to your point, AI on top of bad data, just gives you bad answers faster. And, and the thinking through you, you're talking about, you know, who, who in the organization quote, well, I'm not saying owns it, but I think about the initial introduction of, uh, the, the thing called the internet, I was with visa, I think when I was saying 94 and we were still using, uh, uh, FTT protocols, I hadn't been browsers yet and at visa, is it, is this a, is it good or bad? And, uh, and quickly, this is not quickly, but fairly swiftly to say, well, we better be involved with this, but there became a big issue about, well, who quote, someone who owns this in the organization and a big company like visa, well, it's got to be IT. Oh no, it's marketing. No, it's, it's a product. And, uh, it, it, it, it, it, I think it's some people that all had to come in language and start to use it and we think about how that's, it's evolved today and, um, you know, and, and even how that's morphed, what was, you know, now a website maybe is less important than, you know, a company's presence on other, uh, other channels, uh, but I think we're, we're going to that education piece about, uh, you know, where this sits and get beyond the superficial that, uh, and just lap, you know, label it, uh, you know, it's, it's, it's, it's a buzz word and how you measure it, I think it is back to initial internet. Well, it's, we had so many, uh, visits or hits to the web, but what did you sell? What was your conversion rate? And I think that comes back to the discussion we had about, uh, about ROI. It is, as you look a lot of our, our clients at biopsych group, our Fintechs, we, we have a, have a, have a global focus, uh, you know, us as the big market, uh, but we'll cover other parts of the world as well. Uh, where I saw overlap with, with us in Bridge For, certainly with AI and, and I think, uh, some, some very innovative and, uh, leadership work you're providing there. And, and, and also on, well, we, we help companies grow in the funding is a big one, but the other one was, it really last couple of years has been around regulation, licensing, and it's, that's kind of old pay to a lot of companies. It's like, well, great. I've got a great use case. I've got my funding, the platforms up, let's go and then stop. You don't have this, this license. And, um, if you run into that, you're in trouble because the licenses aren't earned in a week or a month. Sometimes it takes multiple months. Um, and that was a trend I saw. So my, my stepping back is kind of looking at 2026 and, uh, maybe first, I should get two questions. One in terms of AI, what do you see as some of the, you know, the key issues for this year? And I made, we're talking about Fintech, how you see that industry. But in 2026, uh, you know, what do you think, uh, you know, the big, uh, the big question, or about the headline news, we get to the end of the year, they say, here's, here's, I've put you on the spot, Johnny, here's, here's what we learned learned the most about, uh, about AI. Maybe we didn't learn anything. We're still searching for it. Uh, kind of your thoughts on a, you write the headline, it's, uh, it's a, a, a January 1st, 2027, uh, what, what was the AI news for our industry this year? Wow, yeah, I'm not only in a puff, but I think you should, um, and I, I think that, uh, it'll be a year of, you know, what are the things we've learned and of what are maybe some of the biggest mistakes we've seen. And I, I, I think, yeah, you know, similar to risk management lessons over the years, right, using AI that, it might, it might be an over reliance on AI in certain areas, might be something that is a, that's a big headline on, you know, certainly, you know, depending on how things cycle, and we, we saw organizations, you know, take, take a beating, you know, chasing volume at the top of a cycle, um, actually under investing in other areas. I think also, uh, ignoring early warning indicators, uh, could be a headline and, uh, treating compliance is an afterthought. You know, right now, there's sort of a, it's been tempered, uh, a little bit, but, uh, uh, and we, we are in a regulated world and in, in, in another year, two, three, uh, when we just came out of an era, um, which I think many have a little PTSD, but, you know, there's, uh, lookbacks can extend further than, you know, two, three years, right? So, let's say where, you know, you know, the year is 2030 and there's some things that are unearthed and, you know, there's a, so some, uh, you know, regulatory, uh, motivation to look back and see what were your A-models doing in 2026 to get us here, right? So, uh, yeah, in terms of, you were talking a little bit about, uh, you know, where it sits in an organization. I'm, one of the things that I wanted to comment on that I do on this site is how we're looking a little bit, agente AI, anything, it better to bit more as an word structure. And, and this actually comes from a, a colleague of mine and, and some have stricter definitions of, of, of this because if you're giving something agency, it's meaning it's operating independently, but, uh, we're all thinking, you're thinking of it as an agenteic sort of structure, like an word chart. No, you know, one agent does, does the work and other agent checks the work and other agent provides the quality assurance and other agent handles the documents. You can have a AI models, uh, you know, sort of covering those, those various bases. And that's probably a positive theme that we'll see that, you know, a best practice would be to, you know, have, uh, you know, various AI functions, checking each other. And, uh, I think that's where you'll get your optimal, uh, opportunities. That's, that's a great takeaway. And I look at it, um, and it kind of wrap up here and, and this, uh, fine February morning, uh, I think what you're doing in this, this, the methodology you've laid out, the perspective is something I'd like to recommend to, to our clients. We, we have a reach that, originated with Africa. A lot of my work at the past 10 years has been out of Africa. And there's, there's a lively, vibrant FinTech community there. But I've also talked about the need to, whether it's Africa or not, really to make sure you've got a lifeline into the US, uh, whether it's for funding, whether it's so much of financial regulation is dominated by the US, by our treasury, by FinSEN, uh, that I think you really need to be engaged, whether you're in London or, uh, Lawanda or Singapore. And I'd like to recommend, you know, the folks that they, they'd look to you if they have questions in that regard. And, uh, John, just how, how can people reach you at, at Bridge Force? What's the best way if, if folks are interested in, uh, working with you, um, how do they contact you? Yeah, I know. I appreciate that, uh, that endorsement, Rich. There's a lot, especially coming from you, and, uh, our website, bridgeforce.com. Uh, we are all very accessible. Uh, there's a, uh, a way to contact us through their website and, uh, we, we will respond. Uh, we will, uh, and, and we always say a bridge force, you're, you're one call away from a managing partner at any time. Uh, and so, uh, yeah, if anyone wants, just another brain to pick, uh, we take those calls all day long. Uh, and, and, and, and likewise, to return the courtesy, uh, you know, it's, I talked a lot about what we, we do do and what I like to think we do well. Um, but where, where we're not as versed or go as deep, um, is where I think your, your, uh, capabilities, Rich are a nice compliment. Um, and, and you're, you asked about FinTech, um, as I mentioned, we do a lot of work, uh, with FinTechs. And I think, you know, the trends I'm seeing for this year, there's, there's a handful of them, but obviously instant payments, um, real time, everything, uh, faster settlements, real time risk decisions. Um, so, you know, how we can compliment that is, you know, risk and fraud controls, obviously have to operate in, you know, real seconds. We're seeing embedded finance everywhere. I commented earlier that I think everyone's becoming a bit of a FinTech, if you really think about it, but, uh, you know, you've got the BMP all at the point of sale. You've got cards there being issued by non banks. And so you sort of, you, the banking is a service. I think we'll continue to evolve. Um, the data explosion is unbelievable. This is where the UK, um, in my view, sometimes trailblazes what we do, stateside, but the open banking, the cash row underwriting, the, the payroll integrations, uh, you know, all that can get to better decisions, but it also, as you alluded to, um, can, um, make privacy and compliance a little bit more complex. I think, uh, the last, like, sort of closing remarks for FinTech, uh, yeah, there's still a lot of modernization taking place with API layers and, and cloud migration. FinTech's had a little bit of an edge over in coming backs because they didn't have to unwind decades of spaghetti code. Right. But again, I think I like to say a lot of the innovation in, in financial services is, is really renovation. So, you know, we do a lot of work with helping our clients modernize their tech stacks, even those that just came online in the past five, five to 10 years are, you know, are, are, are, are, do for upgrades. I think the regulatory landscape, I'll be remiss if I didn't speak to that. The CFPB focus has been uh, neutered considerably, but that's being offset by state level, um, our regulation and, uh, attorney generals. And I think we'll continue to see more convergence of banks and, and FinTech's more, more partnerships, fewer disrupt and replace stories. And then, uh, I do, I think that you mentioned growth. I, I feel like, you know, there'll be a continued, uh, focus on growth as always. But with a, with AI, uh, an overlay of, uh, of, sort of, you know, unit economics really, really matter. That's great. That's great. John, I really, really appreciate it. I think, uh, there's enough topics there for another podcast. Maybe later on this year, reach out to you again. And, uh, really want to thank you for joining us and, uh, uh, introducing us to introducing you and, and bridge force to our followers. John, uh, John, you take care and we'll make sure our paths cross again very soon. Yeah, I appreciate the opportunity, Rich. Thank you. And, uh, I guess, uh, yeah, I liked what you structured this was between FinTech AI and risk management. It, it, it, it all threads together. So, you know, if FinTech's changing our world and how our financial services are delivered, AI is obviously changing how decisions are made and, uh, risk management, make sure, you know, all that's done responsibly. Very good. Well, you'll be well. Take care. We'll have you back again soon. Take care, John. Thank you. Thank you.