Latest / Real Investor Radio Podcast / Ep. 112 | AI, Fraud, and the Future of Mortgage Lending with Alex Goldovsky

Transcript

Ep. 112 | AI, Fraud, and the Future of Mortgage Lending with Alex Goldovsky · Real Investor Radio Podcast

• Craig (00:00) Hey, welcome back to real investor radio. We don't get jack. We do not get repeat guests often. But when we do, we think it's important to have them back. So back with us today is Alex Goldovsky. He's the CEO of ProTidal USA Docs Solution USA and one due diligence. He's the holder of nine US patents, the two time author we were talking about a book recent just before we hit record today. And he's the guy that Wall Street Fanny, Freddie and the FDIC all call when there is a problem with title or doc problems and they need to disappear. Alex, honestly man, loved having you on the show the last time it was episode 99. I'm sorry that we couldn't have you for the, for the greatest episode of all, which was the hundredth episode. We had to have Fred Lewis on for that one, but Alex, you were 99 and that's, that's the pen ultimate to 100. It's great to have you back. Alex Goldovsky (00:52) Almost there. I know. Craig (00:54) Honestly, love the conversation the last time can't wait to for all of the directions that we want to go in today. So welcome back to the show and thank you for taking the time. Alex Goldovsky (01:03) Thanks a lot. I always have fun with you guys. when they call me, I call Jack. said, Jack, what do think? So ⁓ we know each other well. And I think there's a lot to talk about. I agree. ⁓ And I can tell you, things are happening in the markets. ⁓ Yeah. Craig (01:21) Alex can, but first, can you just give people like, I gave a very haphazard intro. Can you give people like the, the 30,000 foot view of your career, what you're really focused on and what you're really focused on now. Alex Goldovsky (01:34) Sure. Let's see. So let's start with I'm a great guy. Instead of all this nonsense with the CEO here and CEO there, ⁓ let's see. ⁓ Started the business in 2008. ⁓ Really focused on capital markets for all my businesses. We made the ING5000, I think, seven years in a row. So we're growing like crazy. ⁓ We conquered. Craig (01:40) I'm Yes, yes. Alex Goldovsky (02:01) the analytics on the Wall Street where some of the most complex title issues are analyzed through our analytical dashboard for ⁓ pretty much everybody on the Wall Street. Some of the bigger names, you mentioned, know, GSEs or HUD or, you know, any significant buyer on the Wall Street of assets, they know us, they know me, ⁓ and they can read our analytical text title, diligence. materials without knowing anything about the title. Right. So that's where I made the name and my background is engineering. And I used to work for Intel, Luce and AT &T Bell Labs, you name it. ⁓ Thankfully, my wife said that I got to start your own business. So that's, that's the path I chose. And, you know, my wife, I got a, I got a credits here. So if you're listening. Craig (02:50) The better half. Alex Goldovsky (02:56) ⁓ Then let's see, so that put me in a unique position because I see all the flows in the capital markets, who is buying, who is selling. Typically they use my firms ⁓ and I can sense where the problems are. ⁓ In fact, we have a seven year contract with FDIC and we just received a few orders. I'm not telling you from which bank or for which bank. It's not public yet. So things are a little shaky on the market. ⁓ know, if you look at the overall picture, you know, I sense that things are getting sour. Craig (03:37) Well, let's not give let's not put the cart before the horse. But but so all of the, you know, average listeners and myself know so you, you were basically able to use your engineering mind to look at the title workflow and match that with your knowledge of capital markets. Alex Goldovsky (03:38) Yeah. Craig (03:59) And that is essentially the companies that you have built around all of that, correct? Alex Goldovsky (04:03) Yeah, almost there. So I use my engineering mind to design and architect the initial analytics that everybody's copying on the capital markets. ⁓ And I didn't know the capital markets before I had a great connections with a few partners of the, I'd say tier one REITs or funds that helped me to design what what is now a standard in the capital market industry. folks that are trying to sort of jump as high as I did, they're trying to mimic what we're ⁓ Luckily for one of the things that I recently launched, which is full AI underwriting, we have a patent that I authored three years ago. That's number 10, baby. Yeah, so that's the... Craig (04:51) Is that number 10 pattern or is that still number nine? ⁓ look at you. Jack, do you have any patents? Slacker, the slacker. Alex Goldovsky (04:58) I highly recommend that it's a weapon in a fight against a competition. ⁓ Plus gives you the authority. Jack BeVier (05:07) I'd have to have original idea first, so I'm working on that. Alex Goldovsky (05:09) You have plenty of those. ⁓ But yeah, so the patent was issued. took 2 and 1 years to approve. But this is the, in my mind, one of the initial AI patents in the mortgage and finance industry that's related to cross-document verification of common fields. So whenever you have a note, mortgage, appraisal, you have the same borrower name or address. Jack BeVier (05:11) Working up Alex Goldovsky (05:35) The AI automatically cross verifies the same fields across documents and gives you the accuracy score, which is 100 % accurate rather than per field per document accuracy score, which is now irrelevant. Right? So I'm sure Jack knows exactly what I'm talking about where you don't care about everybody's asking what's the accuracy of data extraction. And my answer is, I don't care. If the fields match across multiple documents as is, it's a hundred percent. Craig (06:03) Of By the way, Craig knows that too, not just Jack. So thanks, Alex. Appreciate that. Appreciate the call out. Jack BeVier (06:04) huh. Alex Goldovsky (06:07) ⁓ there you go. Now I'm just looking at Jack and see, you know, he's building his AI and ⁓ I'm trying to figure out if he's building the same thing. Craig (06:17) There's 200 people there's so Jack jump in. What are you excited about? Jack BeVier (06:17) Yeah well, yes, so I'm really interested, ⁓ Alex to talk to you about. Yeah. So what's, what you're seeing. So why the, why the souring tone? What, what are you feeling out there? Craig (06:30) So we're talking market here. Alex Goldovsky (06:32) yeah. So, I see a lot of things, right? So everybody reads about layoffs, tons of layoffs, blue collar work. ⁓ Folks that have households, they have mortgages and homes. That's part one. And it's keep increasing. And I think it's related to AI. don't know. People are still doubting that AI will somehow not impact Jack BeVier (06:32) Mm-hmm. Alex Goldovsky (06:58) unemployment, it will drastically. I used to be in the camp, AI will not impact unemployment. The number of people will be brought into the sort of train AI and do all kinds of rewriting AI. Now that's BS. You will see a high unemployment because of AI and ⁓ we're going to start talking next year, probably universal income and things like that. Craig (07:20) Alex, do you think there will be like, ⁓ my, my, my theory is that there will be a major disruption. But then I have to subscribe to the Jack Bevere, David Moses theory that like, humans figure it out, right? Like I just got laid off from my job. Do I languish in, ⁓ you know, my unemployment? Or do I go and learn the skill that appears that all the world is learning right now? And It's good to be the guy on the forefront of all of that that's been playing with AI for the last year or so. But those people who get displaced, I do think that there's a place for them somewhere ultimately. And whether that's an entrepreneurial path or whatever, ⁓ there's going to be disruption and then, then, then sort of the comeback. Give us your take on. Alex Goldovsky (08:09) I think there will be some of that. Some of the high paying jobs will transition into a, you know, do everything by hand jobs. You know, the handymans of the world and the roofers and, you know, with the lack of immigration. I think there will be some of that. ⁓ College kids are still calling me. They can find jobs. Folks from, you know, Harvard or Yale or Penn State. And I present in Penn State campus all the time. Craig (08:18) Yeah. Alex Goldovsky (08:37) in their entrepreneur club and I teach them what to do and what skills to pick up. And I tell them, guys, you got to find a job before you graduate. So it's at least you're going to have some experience. I don't see that, Craig. I still see that some of the stuff will happen that you're saying there will be, say, 30 % of ⁓ changing your careers. But a lot of folks are obsolete. A lot of folks are obsolete. Craig (09:02) So why your why your Dow or theory on the market? Alex Goldovsky (09:05) Well, ⁓ once people will not get laid off and not find a job, ⁓ so the mortgages will go sour. We'll see a lot of NPLs and foreclosures. ⁓ In fact, I see already a lot of fraud going on. People are maxing out on their equity. Craig (09:22) Is it regional or is it or is it just sort of across the board? Alex Goldovsky (09:26) I mean, I see it across the board based on where the big companies are. I can tell you from my own experience, right? So we try to, and Jack knows this too, and I'm talking to Jack because we stumbled on each other at the conferences and chatted about some of the bigger names in the market. You talk about Bank of America, JP Morgan, Goldman. They're all looking into this year, not two years ago, three years, only this year they started rolling out some of the AI initiatives. And I can tell you with the new models that were rolled out by both ⁓ Claude and OpenAI, they're very powerful. Entry level jobs are not needed. You need the subject matter expertise and you need the QC people offshore and you don't need anybody else. I do see that you still need call centers. You still need people talking, you know, real people talking on the phone. I don't believe that AI is there. Funny story, one of the companies launched AI bot that talks to people, seniors, and they misprogrammed the ⁓ delay between words to 30 seconds. So you have hello, and then 30 second pause. Believe it or not, people responded. I couldn't believe it. Yeah, that was funny to me. One error in the delay in word interval ⁓ was not just great. from the satisfaction perspective. anyway, ⁓ yeah, it sounded like an AI voice. So everybody sort of recognizes when AI calls. ⁓ But I did the ⁓ LinkedIn video analyzing how many jobs will be lost in our industry, mortgage and finance, next year. I come up with about 120,000. Craig (11:08) as a percentage of. Alex Goldovsky (11:10) That's a percentage of rough assumptions on how many jobs will be cut from brokers, salespeople, LOs, underwriters, ⁓ trending to the vendors as well. Jack BeVier (11:11) This is behind the scenes. Craig (11:25) And as a percentage of the total employment, you're saying 100,000 plus, is there a million people that work in these positions? Or is there like, is that 10 %? Alex Goldovsky (11:26) They're not replaceable. Yeah. I'm not even touching the pure banks and funds, et cetera, because they're pretty skinny. I'm only looking at the mortgage and finance industry, right? So folks in the servicers, folks in the originators, and diligent firms, and all the vendors around. Craig (11:57) Jack, do you concur? Jack BeVier (11:58) I don't, but I'm not sure if you know, I, you know, we'll, we'll, we'll see. We'll all see. I can't believe I'm the optimist. It's really funny for me to, if I feel, I feel funny being the optimist, but um, Craig (12:05) I can't believe you are either. I've always looked Jack, I've always looked at you as ⁓ initially when you know, we've known each other for 20 years and it was very and you and I've said this, I'm not saying anything that folks that like you wouldn't want me to say, you know, look, when you're first starting out in your career, and you're building something, you tend to be very close to the vest and you tend to be very like, hey, I've got this thing and I don't want to like tell anyone about it. But I feel as the company has grown over the years, Jack, you you and Fred have very have become very like, Hey, we've got some real killers that work for this place. And, you know, we've got a mission, and these people follow the mission. And we're just going to scale this thing to as big as we can. That said, what surprises me about your take on displacement jack is that the entrepreneurial kind of like the guy who really knows the numbers doesn't go, that's a position that could be easily replaced. Like, like that that surprises me. Jack BeVier (13:01) I know. Well, I just don't think it's a I think it but I just I guess I'm I'm at the pie gets bigger guy like I do think that our everyone's job description is going to change. I 100 % agree. Everyone's job description is going to change. Are we going to have we going to need less people though? I don't think so. I think it's just going to be that we're Yeah, I think we can go the pie. Now the counterpoint to that is that Craig (13:19) because you think we can grow the pie. Jack BeVier (13:25) There's only so many how you can't grow the mortgage pie without growing the number of houses. So that market is fixed. And I think that that is a very super fair counterpoint. And so I'm probably a little bit more pessimistic, but I'm pessimistic on the mortgage market in the where it is plain vanilla, like the conventional mortgage market, the 5 % down to the 5 % to 20 % down conventional market. where you don't, where like you have a sophisticated counterparty and the product is very simple and you put in your W-2 and whoever gets the best rate wins. Yeah, those people, if that's what you do, if you underwrite 20 % down conventional loans, yeah, you probably, you aren't gonna be doing that next year. Craig (14:12) It's the small mechanized box jack, right? Like, like we've got a small box, we've got to check some boxes, you know, we collect docs, those docs are conditioned by AI. And the box is so small that like we click the boxes and we're ready to go to close. Here's my here's my here's my ask. And we've been talking about this almost since the second or third podcast jack, my theory of DSCR. My theory is that hedge funds and Wall Street ruin everything. Jack BeVier (14:26) Yeah. Craig (14:36) that you take a Main Street business and you turn it into a Wall Street business and they suck the life out of it until like they're making all the money. But the the next guy down the chain is making less and the next guy down the chain is making less. So if Wall Street is funding all of DSCR because they're buying all the notes and the box tightens and tightens and tightens and DSCR becomes a very commoditized business. Why wouldn't you want to suck all of the payroll out of that business as possible to get better margin? Jack BeVier (15:05) I don't think that you can. don't think that it is. I don't think DSCR is as easy as those other mortgage products. I think that you have to catch you're doing its business to business loans. You there are hustlers and thieves and entrepreneurs and people who like whose job it is to find angle in their business. And that's who your counterparty is. And those need watched by people who understand the game that they're playing and Craig (15:28) there's risk. Jack BeVier (15:32) everybody and if you you try to if you create us we did we created a DSCR box in 21 and 22 and by 2023 they'd figured out how to hustle it and they and they extracted you know 75 million dollars in Baltimore alone by 2025 so like because we were checking boxes for a couple years there and now everyone's now everyone's coming back and like the end Craig (15:53) But it's still low margin business, Jack. It's still a low margin business. And the entrepreneurial side has to say, how do I how do I get my margin as they are compressing because it's becoming a more and more of a commoditized business. And I think the the first line item and all of that is payroll. Jack BeVier (16:10) Yeah, I just I think that you but you need subject matter experts. I think there'll be a consolidation within that business where the brokers who are doing DSCR business, but then in the next moment, they're doing a bank statement loan. And the next breath after that, they're doing a second mortgage. He lock and they're not they're not specialists in anything. They're just box checkers. Those guys are going to get hustled out of the business and adversely selected and eventually cut off because the. note per the DSCR loan purchasers are going to realize the quality of loans that are coming through that broker channel that that general the generalist broker channel is paper because the hustlers were going to go take advantage of those guys and they'll eventually get cut off and we'll have this pendulum swing where back into you know back in 2020 and 2021 it was a business purpose loan product distributed by business purpose lenders by 2023 and 2024 it had become everybody can do it and so and all the hustlers and thieves started taking advantage and picking on the brokers and stole from them. And now what's happening is that the default rates in that category are going up and the DSCR purchasers are realizing, ⁓ crap, this isn't the same product. And now that box is getting tightened and it's going to swing back to this is a business purpose loan product that needs to be distributed by business purpose lenders. I'll be and I think that there will be enough business purpose lenders to service the entire market with the help of AI. So my the portion of like my the portion of the pie that I get a shot at is going to grow. And so I'm not going to let anybody go. I'm going to add AI. And I do agree that like there are losers in this story, but they never added any. Those humans never added any value to the equation in the first place. They were just box checking like they weren't valuable humans. to this supply chain anyway. So yeah, they get displaced out of it, you know, this is capitalism. gotta... Craig (18:07) I won't press you on those job descriptions, Jack. Alex, jump in. We're having an argument here. Alex Goldovsky (18:11) Craig, I have, I'm writing down my counter arguments right here. So you ready? All right, let's go one by one. ⁓ With the current Claude infrastructure rollout and AI ⁓ bots that you can create to do various functions, I now have ⁓ somebody, I will not promote that party, title agent, that can do closing with those bots with no human. Jack BeVier (18:14) You Let's go. Alex Goldovsky (18:39) No human. Usually it's 10 to 12 people. Now it's no human. So that's if that will. Jack BeVier (18:39) Mm-hmm. Craig (18:43) And it's in minutes, it's in minutes rather than days, right? Alex Goldovsky (18:46) minutes exactly unless you have a cooling period where by law you have to wait three or seven days. So that's dangerous. He calls me and says, Alex, I don't know what to do with my people. If this is successful, he has a over 100 person staff that he needs to pay the salaries for while he deployed the bots and he was successful. So that's first argument. Let's go down. ⁓ What's the fastest growing? Yeah. Jack BeVier (19:11) Yeah, but hold on, but nothing closes smoothly. Nothing, like nothing, like. Alex Goldovsky (19:16) Right. So it's Jack, it's 80-20 rule. 80 % will be clear to close. 20 % will have hair. Craig (19:18) Pachack, pachack, pachack. Couldn't agree more, Alex. I think... Jack BeVier (19:24) It's the opposite. 80 % have hair, 20 % will go through clean. That's going to be the crux of our debate. That's going to be the crux of our debate. Alex Goldovsky (19:27) I don't know. DSCR maybe. DSCR conventional. Craig (19:28) Now. But because Jack, because Jack, you set up, if you set up a system to catch every edge object, you're going to get every edge object. But those, but like, yeah, if you want to, if you want to hire, if you want to have payroll for those people to take care of all that morass, great. But why not just set up a box that like, you know, you can close in 10 days. Alex Goldovsky (19:32) Commercial is easy. Jack BeVier (19:50) Yeah, by the way, I probably I probably will concede title because title is also one of those like title insurance has never been worth the money. You know, the paper it's printed like, know, like, excuse me, if someone's willing to underwrite you a title insurance policy, you don't need to buy it like the default rate. The claim rate on title insurance is absurdly low. 80 % of the premium goes to the guy who's sitting at the kitchen, sitting at the settlement table preparing the documents. Alex Goldovsky (20:10) Jack is dictated by weight. While I agree with you, Jack, it's dictated by rating agencies and securitization. There's guidelines and limits that you should get the title policy when you're above the limit. That's it. To change rating agencies, it'll take decades. We're still operating in this old schema of the dates when we report to the rating agencies. it's not the whether you need it or not. It's dictated by... Jack BeVier (20:22) Yeah, yeah. Yeah. Alex Goldovsky (20:43) capital markets and rating. Jack BeVier (20:44) Yeah, so like title producers, like title shops, I agree. Like that is probably something that can be highly commoditized. You'll keep the attorneys on staff because the attorneys were only ever dealing with the exceptions anyway. so you need, but like the people who were, I don't know, the people who were reading the abstracts and saying, yep, looks good, check. Yeah, we can have that automated 100%. And then the closings themselves. Everybody 35 years of age and younger will say no, I'll just close online I'm not interested in showing up but there will be a long tail of old people who like to show up and sign and sit at the Settlement table and have the guy yuck it up and explain to him what this escrow agreement means because they're old You know me and above will like might still want to sit down So there'll be a long tail though. There'll be a long tail like it'll be a dying industry Alex Goldovsky (21:30) See you. Okay, so let's agree to disagree, right? I have five items, so you jump me on the first one, right? Which is okay, Jack, all right? Craig (21:40) ⁓ Alex, you Jack BeVier (21:40) Okay, okay, okay. We Craig (21:42) I Jack BeVier (21:42) got lots of time. Craig (21:42) think you and I agree. I want to I want to know ⁓ when we when we were speaking about the market. I didn't know that we were rolling up all of employment into sort of what you're seeing in the overall housing market right now. And I thought that you wanted to speak to that. Jack BeVier (21:58) You're seeing an uptick in the weight. Hold on. He has five. He has four more things that he wants to bust my balls on. Yeah, yeah, let's go. Let's go. Keep going. Keep going. Hit me. Craig (22:02) let's go. Let's bust. Alex Goldovsky (22:02) Exactly. Yeah. All right. What's the fastest growing market in mortgages today? Helix and closed in seconds. Right? Okay. So, Jack BeVier (22:09) Probably Helox, yeah. That's not a good thing. Yeah, I agree. That's not a good thing. Alex Goldovsky (22:14) Listen, so. Craig (22:15) Use that house as an ATM, Jack, why not? Jack BeVier (22:17) Mm-hmm, yep. Alex Goldovsky (22:17) What's the new publicly traded company that just went on the market? It's a HELOC company figure, right? So it's fully digital HELOC company. No title. Everything is automated. I actually spoke with somebody who applied for their loan. I'm not promoting them, but what I can tell you, everybody is in the market, took a notice, and they're designing the similar infrastructure. No human HELOC. OK. Jack BeVier (22:22) I know. I'll figure you. Craig (22:42) So how's it go? Alex Goldovsky (22:45) have an app on phone, you say, I want a mortgage for $100,000. OK, and you apply online. And then you put your data and click a button, apply. You get a text message. Are you ready to sign the documents? You click on that text message link. In five seconds, you have a notary looking at you on the phone. You do a consent that you can assign everything electronically. It takes whatever time it takes to show your IDs. You're done. That's it. Only one document. Craig (23:11) Jack, can we build that into private lending like tomorrow or? Jack BeVier (23:14) If we did, there'd be a tremendous amount of fraud. just think that only works that only works like they and there's going to be a big tale of fraud to it. But like it, it works on the house that's worth, you know, 800 grand. And then she's got a $250,000 mortgage and you want to give her $150,000 line of credit. Sure. You know, like sure. When they're when the loan to value goes above 75 % when there's anything related to business purpose, when there's anything outside of W2 income. Alex Goldovsky (23:22) All right. Craig (23:30) When does it not work? Alex Goldovsky (23:41) Agreed, agreed. So this works when you tap into equity. different companies trying different things. Craig (23:42) Yeah. Jack BeVier (23:46) It's a credit card, by the way, it's not a mortgage, it's a credit card. It's a credit card that happens to be secured. Yeah, yeah. Alex Goldovsky (23:50) Bingo. It's a virtual credit card. Yes, exactly. Craig (23:52) Yeah, Alex Goldovsky (23:53) Yeah, Craig (23:53) of course. Alex Goldovsky (23:53) yeah, yeah. So it goes to your Apple Wallet. That's what Avon Financial has it, right? Avon sends you a credit card, right? Avon is my clients. I guess they don't mind if I disclose that. So they send you the credit card, and this is HELOC. You just charge against your home, which is great. ⁓ It's a great concept. I might disagree that some people will go crazy and max out. Jack BeVier (24:09) Mm-hmm. Alex Goldovsky (24:16) And Wall Street will find another way to make money. those type of HELOCs, fastest growing figure, think was it top three or the highest ⁓ originator in the United States? So people are applying for the mortgages on their phone. They used to just e-sign everything. Now, granted, it was not tested in the court of law. If somebody defaults and I want to foreclose. not HELOC? Do I have the note, the mortgage? Do I have all the legal documents to go on foreclose? I don't know yet. It's not there yet. Craig (24:51) Jack from a non from a non QM standpoint. ⁓ You know, you and I have talked about it and we've made no secrets about it that like a guy called a real estate investor calls us and he's got a deal on the table but he first has to get pre approved. have to make sure that he's got experience and sort of like credit and and all of those things. Where in this process Can you say, ⁓ I see where the market is going. The market is going to like instant approval, instant sort of like we've got you underwritten. You know, where can you see that applying to what we're thinking about here at Dominion? Jack BeVier (25:30) Yeah, like instant pre-approval of your fix and flip ability as your resume as a landlord. think that's here. We're doing that now. If you can... Craig (25:39) easy, right, right, right. And then what about the PFS PFS sort of like the you know, your financials and credit and all of those things? How fast do you think we could turn that around? Jack BeVier (25:49) Yeah, that's that is I think I think we can do instant things that will be just as good if that we better than the current system. It will be slightly different than the current system, but it will get us a credit answer that is just as good. That is better than what we are currently getting an instance. So I'm with that. I think. Yeah, and I think that the close, so I think that I do think that the closing time frames, even for business purpose loans, are coming down very quickly right now. ⁓ Craig (26:05) and how instant. Jack BeVier (26:16) So I and with AI, the confidence that what I'm really excited about with AI also in mortgage specifically is I think default rates are going to come down like tremendously because you can do two things. You can make it much easier, Alex, like you're talking about, and you can do the same amount of work and make it much, much easier. And or and or you can do a ton more work from a due diligence perspective and find all of these edge cases and weed them out because the cost of due diligence has gone down so significantly. There are like the level of diligence that you can do because before, you put a human on computer and you do you could spend you could you could put a human on the computer and spend 100 hours researching a counterparty that you're thinking about making a loan to but it's not cost and you'd find you dig up some You know, and, it's not cost effective to do that, but now you can put a team of agents on that same task and spend $7 in tokens and come back with an intelligence report that found every freaking edge case. So if you link in all the databases, right, you can pull in all the freaking APIs. So put up, pull a hundred APIs together and, and, and make that part of the diligence closing process, add 30 minutes to the, to the, know, to the You know, the the the you know, we talked about you can close five minutes later, close 35 minutes later, and your default rate will come down by 10x. I think that the biggest change in mortgage that's about to happen is that the is that the default rates are about to come down tremendously. I think that we're going to get we're going to be able to be able to make much better business decisions because the cost of diligence just came down to 100x. I think we just started. Craig (28:06) Alex. Alex Goldovsky (28:07) I disagree, Jack. Yeah. Well, let me jump on that. Well, yeah, yeah. Okay. Let me get my boxing gloves. ⁓ Let's see. So I do agree that the quality of paper will be much better. Underwriting techniques will be much better, but we're getting, I mean, I want to say we're going to recession probably this on next year. So the level of defaults related to the unemployment and Jack BeVier (28:09) I think we just started implementing. I think we just started putting that in. Craig (28:14) Ding ding, next round Alex, go. Jack BeVier (28:32) Yeah, so tell. Alex Goldovsky (28:36) AI displacement will be much higher than the benefit of ⁓ the crisp underwriting that you're talking about. ⁓ Jack BeVier (28:44) So you think that there are AI displacement already happening or you're worried that AI displacement is gonna happen? Alex Goldovsky (28:50) already happening. Already. I already see where we deployed AI, folks that resisted AI are out of the door. And the way it happens is this. We deployed with one lender, full digital AI QAQC on their origination workflow. And the manager for the department that's supposed to run that QAQC tool that we did, ⁓ before AI spent, let's say, an hour per load. After AI, the CEO of the company said, how long does it take? says an hour. I said, that does not make any sense. So what was happening is he was using AI and then scrolling the PDF files as usual. And I go, well, you know, what's the point? I mean, you got to trust AI. At some point you have to let go. He was not able to. He's an elderly guy and he got let go. Craig (29:29) They were reviewing, Alex Goldovsky (29:42) So AI displays that level of workforce. So people need to adapt. Greg, as you said, they don't, they're out. As simple as that. Now, ⁓ just one more thing on my list. We got through two items. ⁓ Did you guys know that the law smith departments and some of the servicers have over 100 people, 100 employees? And you know what they do? Paper pushing. Craig (30:06) yeah. Yeah. Alex Goldovsky (30:08) So for example, for FHA claims, they have to match every corporate advance by the servicer with the invoice by hand to get paid. How is that not being replaced by AI? I don't know. It's just, I visited servicers and when I look at what they do, I feel bad for those people. First of all, that's a pretty boring job. And second is, if they see what AI can do, Craig (30:29) Yeah. Alex Goldovsky (30:33) Those jobs will not be available if the management decides to replace those people. Jack BeVier (30:37) How long do you think that tail is? I think that we might be debating a timeframe here. Cause I agree that it's going to take, I agree that those positions are not going to be the same. just think I'm having, I, are like really very, I think very forward leaning into trying to implement AI and dude, is the, difference between vibe coding an app and making something that is enterprise grade is very significant. is a huge, there is a chasm between those two. can make a, I can prove concept over the weekend. Getting that into production is 90 days later, which is better than the fricking nine month timeframe that it would have been if I had to like hire a developer and go back and forth and back and forth and back and forth. But it ain't, but it ain't I'm displacing jobs in a month or in a couple of months. It's like, Craig (31:05) Huge. Huge. Alex Goldovsky (31:22) Yeah. I agree. It's a slow rollout, right? I always said, I think we mentioned that on the prior podcast, the most difficult task for a company in AI business is to go from the very cool demo to full production. That's the most difficult part of, you know, to know all the corner cases, apply the subject matter expertise. That's why we said subject matter experts will be employed and they will probably get the bump in the salary because of Jack BeVier (31:54) Mm-hmm. Alex Goldovsky (31:55) the efficiencies that AI will bring them. ⁓ Jack BeVier (31:56) That'd be some money. Yeah. Yeah. Alex Goldovsky (31:59) So you better know your right? And then you'll be safe. ⁓ But in my mind, ⁓ like you said, there will be jobs that will require to look for fraud. That doesn't go away. With new AI tools, you'll have a new AI fraud. Right? By the way, yeah. Good. Jack BeVier (32:13) You know what I think? That's true. Yeah, that's very true. The battle that is about to happen over the course of the next five years between public sector union employees and taxpayers, think is like brutal. like, like government and like paper, literally like the definite, like the stereotype of paper pushers, right? Like if you're a state employee pushing paper all day, like Craig (32:20) Mm. my god. Alex Goldovsky (32:29) Thanks Jack BeVier (32:41) That is like that should just be that we should have as taxpayers. We should be having service levels that are extraordinary, predictable, like and but those public sector unions are going to go are not going to go quietly into the night. And that is about to be a fight. I think. Craig (32:59) Yeah, agreed. Alex Goldovsky (33:00) I agree with you. don't know what to do with it. I love it or hate it. I wanted Elon Musk to be all over that. ⁓ He's a bit out there, but what he wanted to do with the efficiency gains in the government, two hands up. Yeah, let's go do it. ⁓ I agree with you. It's going to be a bloodbath. In fact, my God, when Trump... Craig (33:16) Yes. Alex Goldovsky (33:24) says that, we're going to have the IPOs for GSEs, and then everybody from FHFA is gone, who knows what they're doing, and nobody knows how to do that anymore. What is that? So it's such a mess, such a turnover. it's things that Trump says and things that happen in the government, two opposite things. Craig (33:45) Alex, what indicators are you seeing that there is a recession coming in the market? Jack BeVier (33:46) So. was my question. Alex Goldovsky (33:53) ⁓ inflation and unemployment, two indicators that Craig (33:56) We, so we just got a, we just got a very rosy employment report last week. It was, I think over close to a hundred thousand more than most analysts were expecting. You know, I feel good about where things are going. I love the Iran war. Just kidding. Just kidding. I want $2 gasoline or less period. But, but, most, and as do most people who are listening to this show and all other Americans, However, that doesn't appear to be slowing down the economy. think the GDP right now, like we're growing at like 3.8%. We've got a pretty rosy ⁓ employment report. What's your problem, man? Why are you so dour? Alex Goldovsky (34:36) the growth of the debt. We're already at 30 plus trillion and that's not slowing down. ⁓ Warsh being nominated today, he was approved today or he's starting what? In a couple of weeks. ⁓ He has a different measure of the inflation. ⁓ He does not account for, don't quote me, Craig (34:50) That's right. Alex Goldovsky (35:05) housing or something else, right? So he will revise the measure of ⁓ inflation and you're going to see much better numbers. And then it will basically give him an ability to drop the rates even more. And that will translate to more inflation on top of the Iran war and the gas prices. ⁓ Jack BeVier (35:12) Thank you. Craig (35:26) Mr. Bavair. Jack BeVier (35:27) No, no. Hey, congruent with the no, no, yeah, I can't fight over everything. Craig (35:30) You want to fight over this one? Alex Goldovsky (35:34) But, know, it's again, every down cycle in the economy causes a lot of fraud and all the, you know, guys that try to cheat the economy and try to figure out how to take out the mortgages. ⁓ I don't know. mean, is a celebrity in my mind. I saw an article that Jack posted on LinkedIn. And it's like, big news, new fraud discovered in DC. And you guys better watch out and do things in diligence. So that got me excited, Jack. I'm a geek like you whenever it comes to fraud schemes, because I need to verify, I check for the fraud scheme in my tools? How can I deliver the results to the client? How can I be unique? How can I differentiate myself? Craig (36:28) Before we jump to the fraud scheme, the fraud schemes. So give me your take on ⁓ if you have one and I apologize if I'm putting you on the spot, but I feel like the bond market is going to getting back to your theory of the of the debt. I feel like the bond market is going to have a say before the next presidential election. We I think we're seeing it have a say right now in Great Britain. ⁓ Alex Goldovsky (36:30) Yeah, yeah, yeah, tell me. Craig (36:55) And I think there's just a lot less people lining up right now to buy US debt, especially given the tremendous debt already of most developed European Western countries. So what does that leave that leaves the Middle East and and ⁓ you know, a few other countries to buy our debt. So give me your take on the debt market and the bond market over the next ⁓ couple years and how that affects where we're sitting 24 months from Alex Goldovsky (37:21) Well, I'll give you an answer from sort of my view of the market itself. I do agree with the assessment that not a lot of foreign investors are now investing in sovereign debt of the United States. In fact, Fannie Mae and Freddie Mac put the risk transfer securitizations and RPL sales on ICE ⁓ just because there's not a lot of interest as before. So a lot of folks are on the sidelines, wanted to see what happens in the midterms. So I think there's a lot of, you know, scrape deals ⁓ are off the table this year and not as much participation in the treasury ⁓ sales. ⁓ Yeah, so from my perspective, it's not great for US economy. ⁓ Craig (38:13) Jack, does it concern you at all? Bond market over the next couple years? It feels like the Treasury is trying to push all sales to the short end of the market right now, because there's just not a lot of buyers in the long end of the market. And so I'm just wondering if there's any take on that. Jack BeVier (38:17) Dude. Yeah, I mean, especially with as fast as things are moving. It doesn't, you know, that's supposed to be, you know, the Treasury curve is supposed to be a function of like, you know, the how much getting paid higher risk premium for a longer period of time. Well, dude, the world is accelerating and accelerating and accelerating. And like the concept of buying a 30 year bond today or even a 10 year bond today is insanity to me. Right. Like what what you could you possibly Craig (38:50) Yes. It's got that crystal ball. Jack BeVier (38:57) Conceive the world looks like in 10 years from now. That is an insane idea. Anyone who says they have any clue as to what the world looks like 10 years from now isn't is just it's a ludicrous idea. So like and yet we still have, you you know, other than maybe Americans are going to be in their houses. So like I mean, maybe maybe the US mortgage is the is the last place where you can put money in for the long term. ⁓ But I don't feel great about it. So it makes a ton of sense to me that we're seeing the the yield curve steepen and as it should be. ⁓ since the Iran war, you look at the two year in the five year over the course of the past year, it was like down, down, down, down, down, down, down. Iran war. And we gave it all back in the past 60 days. And do the five years like for one right now, which was like, which sucks. Like that's not good. Like that's not good for mortgage pricing whatsoever. Alex Goldovsky (39:52) Yeah, from the investment perspective, you know, I don't know anyone who would be interested in investing at 4 4.5 % Treasury market. can get knowing that the inflation will go up and just putting structured notes on things like gold and silver with a return of 12 to 15%. I mean, that's where the investment is. It's hedging against the inflation of US. That's, think, where the trade is. And that's where, you know, I'm personally investing. Craig (40:02) Right. Alex Goldovsky (40:20) ⁓ I'm actually an investor, I should disclose, in fracking wells. I'm benefiting from the high oil a little bit, which makes me a bit unique. ⁓ I'm betting on the inflation. I'm betting on the oil price. Treasury is not a good investment. Craig (40:39) seal this one up with a bow. prediction is the bond market has a say in the next election for sure. ⁓ Jack BeVier (40:45) What do you mean? What do you mean has a a say? What does that mean? Craig (40:48) James Carville once said that he wanted to come back as the bond market because it has the ultimate say in every election. If you look over the course of like European elections over the last 30 years, the bond market goes to in a regime that they want to kick out and they're gone. I don't have examples to give you right this second because I didn't know we would be talking about it, but it's happened time after time after time. Jack BeVier (41:07) ⁓ That's interesting. Craig (41:14) And what administration in the world right now is going against all of globalist policy over the last 25 years but the Trump regime? And so trying to restack the geopolitical checkerboard, yeah, I think the bond market's gonna have to say at some point. Jack BeVier (41:17) Pimp. So PIMCO is gonna decide the midterms. That's... Craig (41:35) they are Jack and I'm not saying midterms I think I think the 28 election will be will be an interesting one for sure yeah all right what's next on the what's next on the list of things that Jack BeVier (41:41) Cool. Alex Goldovsky (41:45) No, the next is wanted to chat about fraud in whatever time we have left here in the podcast. ⁓ So I don't know. Jack BeVier (41:55) Yeah, dude, I have been I feel like I have been under assault since I was talking to another guy at another hard money lender a couple hours ago. I feel like I had this conversation. I feel like I've been under attack since the second third quarter of last year with ⁓ you know, soon as soon as the DSCR fraud kind of like snug its head out. It was last call, you know, all of sudden the credit side of the shop was, you know, of all of Wall Street, like woke up and came back to work and, you know, drank their coffee and they've been tightening the screws since then. And there have been more overlays and more people added to blacklists. And just just that we've been hacking down at the guidelines since then. And word has been out on the street in that period of time. And so everybody who was. either a straight up fraudster and knew it or just not a very good business person and had some concern for their liquidity have been running for the doors to try to get liquidity and you know, and just like, you know, robbing Peter to pay Paul and then, you know, leaving, leaving the business. And I've been paying people to kill loans for the past nine months. That's all I've been doing is paying people to kill loans. Like my conversion rate has of like loan sales to close loans has gone down tremendously in that period of time. And frankly, it's made it a kind of business as an originator because I'm just paying because I'm you know, I'm closing 30 % of the pipeline like Craig (43:30) Jack, is that mostly DSCR or is it RTL as well? Jack BeVier (43:33) I'd say it's the worst in DSCR, the bad behavior is also happening on the RTL side. If they happen to have an RTL loan, they're trying to get too much in draws. ⁓ They're doing more and more straw party transactions or ⁓ non-arms length transactions. You put a property under contract for 200 grand or let's make it worth it. You put a property under contract for 800 grand. you sell it to your buddy, you assign it to your buddy for a million. That's not really an arm's length transaction, but you get a 90 % LTC loan on the million bucks because you get past a week underwriter and you actually, you know, the lender thinks that you're putting, you know, hundred grand down, you actually cashed out a hundred grand on the thing. And if it goes well, great. And if it doesn't it and you know, you just, you know, walk away on the thing. And we're just seeing, I'm just seeing tons of non-arms length activity, tons of marked up prices. And with people who are having some cashflow trouble selling to other people, it's literally a, it is a conspiracy, right? When you get three or four people together and we decide to just move all of our debt, one turn to the right and like, you go over here, you go over here, you go over here, you go over here. It's really freaking hard to catch. yet everybody then just, you know, and there's so everyone cashed out or has the opportunity to get a liquidity event and or pass the buck and make it harder. ⁓ And as an origin, and there's so much liquidity still in the system that if, you know, you know, if somebody in my little circle of financial friends for fraudulent financial friends. can't find a lender, just go to another lender and go to another lender and by the fourth one, you found somebody to fill out the circle. And so I feel like there's all kinds of circles of fraudulent financial friends going on right now as they're running for the exits for liquidity. And I've been just spending all of our energy trying to catch these sons of bitches. Craig (45:38) How would you want Alex to reply to that? Like, that's a lot. Alex Goldovsky (45:39) So, yeah, yeah, hold on. yeah. So, first of all, I'm not paying the ⁓ revenue share or commission to Jack, but after Jack's podcast of the Baltimore City Scheme, I developed a product that now three top DSCR lenders or buyers are using today to ⁓ figure out the straw deed purchases. and going through LLC to LLC to figure out if this is a related entities and friends, as Jack pointed out. So thank you, Jack. It's an extra revenue for my business. More work for you, though. But it's probably a similar technique of figuring out how LLCs are related, if they are related, if they've been related in other transactions, and figuring out the time intervals and thresholds. on the money exchanges between parties before the mortgage and whether or not there's, think we talked about last podcast, right? We talked about, there any history in RTL or DSCR deployments and a real business for the LLC? So all of that is now being captured by some of our diligence tools. But I did want to come back to Jack and ask him, is DC article that you posted? Is that related to the same scheme as Baltimore City or that's a unique in a way where you found something that was done a different way? Jack BeVier (47:11) Yeah, no, was no, that was the more of what I just described just now. The DC thing wasn't related to Baltimore at all. It was that it was an RTL loan and it was that ⁓ several related parties, different names you couldn't find that they related until after the fact. And then you look back and you're like, ⁓ like we've been connecting dots after the fact. And though I do have there is an optimistic note to this because Alex Goldovsky (47:15) Okay. Jack BeVier (47:38) in the past 60 days, I think we've figured out how to actually catch this on the front end. And I'm really excited about that. I think it is going AI. It's of course AI. Yeah. Well, AI driven. Yeah. AI driven. Yeah. So, I mean, the public records are available, right? And now they are completely searchable and the horsepower of AI researching your entire history, right? Like I can... Alex Goldovsky (47:46) AI or not AI? ⁓ Of course AI. Jack BeVier (48:05) I can create a report and I can tell you every house you ever bought and who you bought it from and who you sold it to and paint a picture of your real estate network going back however long I want to go back. so if there are, you know, as soon as you can be clean, but as soon as you do your first misstep, that will be the last time. If everyone uses the tools, you can screw up one time. Craig (48:12) How much? Jack BeVier (48:33) and then you're going to get blacklisted forever is what I think is going to happen. So I actually think that it's going to bring default rates on RTL loans down significantly, which will be good for all of the people just doing good work out there because they'll benefit from lower rates because it'll be a lower risk profile loan product because right now all these fraudsters are priced in. Right. Like that's why RTL rates are still 10 and 11 percent because all these knuckleheads default from time to time and take us for 10 points of principal loss. But If we could just never lend to those guys, then this product should come down by 100, 150 basis points. So I think that that's where we end up in a couple of years ⁓ as the world gets a little bit safer as a result of this better diligence. But yeah, no, that LinkedIn post that Alex is referring to was a financial friends circle scheme that we only caught on the back end. Alex Goldovsky (49:23) I have to ask this question, Jack. I want to get as technical as I can on this podcast to see if you're doing what I'm thinking about. So let me set it up. For DSCR and RTL loans, there is an appraisal that is pretty standard. There is a 1025, which is the sort of a single family appraisal or the 1007, which has the comps on now. grants, right? Apparently, the diligence firms and TPR, third party due diligence firms, do not look at the comps at all because they only look at the subject property and whether or not it was properly appraised. But it looks like if you want to catch the fraud, and correct me if I'm wrong, maybe you're doing this already, you need to scrape the comps from Jack BeVier (50:02) Yeah, no. Alex Goldovsky (50:17) The other addresses, the rental or the lease data on those. So the rental, monthly rental and the end and the beginning of the lease, whether it's recent or not. And then take the addresses that used for the comps and look up if those are part of the fraudulent schemes. Is that something you're doing or that's something that I'm going to deal with? Jack BeVier (50:39) That's something. No, I'm with you. That's something that we are doing manually in high density geographies. So I'll tell you, I'll just tell you what we're doing because I told a ton of people what we're on the comp selection side. The so there's this thing called a housing market typology map that is a product of the reinvestment fund out of Philadelphia. And it's basically a color coded map of a bunch of dense urban American, know, dense American cities. And so All of the Baltimore frauds were accomplished because the comps were in a they rank them a to J in the most recent version. And so, yeah, the typology map is the most competitive. J is the most distressed. All of the I'm generalizing a painting with a broad brush here, but all of the pro the houses that were the subject of the fraud were in I and J very distressed typologies and all the comps. Craig (51:17) in the topology map. Jack BeVier (51:35) for those were pulled in E and F, you know, middle of the road transitional locations that happen to be point six miles away. And so if you if you just identify the housing market typology of the comps and it doesn't match the subject property flag, that's it. Like you can't you can't you can't find a two hundred thousand dollar appraisal. You know, you can't find a two hundred thousand dollar house in an I or a J location. You have so. Alex Goldovsky (51:53) ⁓ that's hard. Jack BeVier (52:03) You couldn't, you know, if you force the I've been telling all the DSCR loan purchasers make the housing market typology required like part of the instructions that like, hey, make it part of the guidelines that when you tell you an order and appraisal from an AMC that the appraiser must choose comps that are within the same housing market typology as the subject property. And here's the link so that you can so that they can, you know, pull up the PDF and pull up their comps and be like, yep, in the right. Craig (52:32) If that subject is in a dense topology area where there is a topology map, yeah, because for those for those listening point, God, Jack, I'm sorry. Jack BeVier (52:32) Topology Cool. Yeah, we're one of these maps exist. But by the way, Baltimore. But by the way, Baltimore, Philadelphia, Atlanta, Newark, Kansas City, like they are like that's where these maps exist. So. Alex Goldovsky (52:47) New Orleans. Craig (52:51) any post industrial city in America. We got that. The so for those listening point six miles away if you don't know, because you don't invest in one of those areas is a universal way. You could be two blocks away, which is a stone throw. And the guy and the appraiser is pulling comps in a completely different universe, which is only you know, might be five, five blocks away. Jack BeVier (53:14) But I do agree, I do agree Alex that like a ⁓ technique of fraud is to use your own comps, right? Like set your own comps with a friend and then use them to refinance off of, right? Like do three arms length transactions with your buddies marked up and you know, so manufacture comps and then have your other friends refi off of those. Now we have, you know, we have three comps and now we can, you you can refi off of those sales. Alex Goldovsky (53:38) Jack, that's on sale. What about on rentals? I see fraud on rentals. So in other words, you jack up the monthly rental to get the DSCR value much higher and you get a bigger loan. So that's where I've seen the fraud is. So basically, I don't know who fills out 10.07, who provides the rental comps. They appraise me, right? Jack BeVier (54:02) the appraiser does. And I think Craig (54:02) PRAISER Jack BeVier (54:04) it's the weakest. It's the weakest part of the system. it's there's there's such limited there's even if you put many most, you know, most rental activity does not show up on the MLS because the dollar amount doesn't isn't doesn't justify it. Most private most of those are private transactions and you and you don't get and there's no standardization of information with respect to tenant screening, months of free rent, anything along those lines. So like Craig (54:06) really is. Alex Goldovsky (54:17) Mm-hmm, yeah. Jack BeVier (54:30) To me, the 10.07 is like the weakest part of the underwriting. And it's also the hardest part to catch because I think what it like... ⁓ God, what's the one we use? ⁓ Craig (54:43) and a meter. Jack BeVier (54:44) Is run on? Yeah. I'm going to come back to me. You know, like there's the standard for the industry of like looking up rental comps is it's not very good. Like it's not great. Like. ⁓ Alex Goldovsky (54:56) There's no reference point, right? Whatever is there, then that's what you assume. And then it's very easy to put the fraud in the rental agreements and jack it up from a thousand monthly rent to a 10,000. And all of a sudden your DSCR looks great. Jack BeVier (55:11) Yeah. Yeah. Craig (55:12) So Alex, what's your conjecture then on the technical side of what you feel that you're Alex Goldovsky (55:15) I don't know. I'm thinking about it. I read the article and then right off the bat, I'm like, how do you catch the rentals? So I was thinking what Jack said about taking the rental addresses that are used for comps and checking for fraud. That's number one. ⁓ And two is ⁓ just getting more samples because you're limited by the form itself, only three to six comps you can put in there. And that's such a small sample for the rental. I think we need to modify the form to have maybe 10, 20 comps. So you have enough samples there. Sale you can track. The leases you can't. That's my problem. Jack BeVier (55:55) Mm Yeah, I think it's a very weak part of the of the underwriting there. Yeah. Rent range rent range. Sorry. Yeah, we use rent range. That's what it's called. And I haven't found anything better, but it's not good enough. Alex Goldovsky (56:03) All right. Greg, when I come up with a product, first of all, I'll start making money and thank Jack again. ⁓ A new patent is coming. That's right. And then I'll be on the podcast again talking about this new product that we got. But I have a few ideas. I don't have it formulated in my mind yet how to catch it. But certainly what I read the news about, ⁓ loan officers are putting fraudulent leases. Craig (56:12) The 11th patent is coming. We're working on the patent as we speak, Jack. Alex Goldovsky (56:36) and putting the amounts to qualify the borrowers for the rent, not the brokers, right? The regular loan officers that want to get paid the commission from the origination of that loan. So that's, know, how do I find that? how can I, you know, Jack BeVier (56:42) Beep. Yeah. Hey, you know, know, the other gaping hole, the other gaping hole is operating agreements, man. Like we had I had a situation last week where a deal got a partner to two partners doing business together for a long time. And one of them decided to ⁓ be the sacrificial lamb and rack up a ton of credit card debt. So racks up, you know, 100 grand of credit card debt so that the company has more operating capital and they're using it for the business. But then he comes to us and he says, and his credit score goes down because his utilization goes sky high. And so all of sudden his credit score is a flag. So he says, Hey, you know what? I'm not part of the LLC anymore. And he submits a new operating agreement and, and, know, where the operating agreement formally transfers the LLC to his partner. And so it's like, so we can disregard him because now this other guy owns the LLC and it's just his credit and he's fine. And the company also doesn't have all this, all this credit card debt anymore. And so it's like all the all the sudden all the boxes are checked. Everything looks good. And there's and the operating agreement like what about you know the day the loan closes they just handshake and like put him back on the operating agreement. Dude who's operating agree by the way whose operating agreement is up to date like no one's operating agreement is up. No. Yeah. Like yes checkbox. And it's the easiest thing to manipulate. And I think it's probably the document that gets frauded the most because it's not recorded anywhere. Alex Goldovsky (57:52) This is genius. Who checks? Nobody checks it. If it's there, it's checkbox. Jack BeVier (58:20) You can change it at any time. Any all you need is all the parties to consent. So it's completely unauditable, right? Like an operating agreement is an unauditable trail of like a financial transaction and we have no answer, right? We have no answer for that So. Alex Goldovsky (58:37) By the way, I suggested before the podcast for Jack to write a book about all the cases that he sees in the market. Jack BeVier (58:43) It'd be a long buck. Craig (58:43) I've always said that as I've always said that a landlord over a thousand units, should have a coffee table book of all the pictures of like when they walked into a property like, you know, just to say so that's a second project Jack that I think you should move on all the war stories. So, so Alex, you know, from a technical standpoint, when you get, when you get an idea that you want to implement with AI, Jack BeVier (58:55) All the war, all the war stories. Yeah. It'll be called war stories. Yeah. Craig (59:08) Do you use your engineering mind to develop that or do you hand that, do you hand your idea off to better staff that can go do that for you? I'm just really curious. you know, you get this big idea. Alex Goldovsky (59:21) Yeah, the way my mind works is I get geeked out just like Jack. I get obsessed with the idea how to solve this. I draw up the architecture. see the elements and the puzzle pieces, and then I hand it off to the CTO. And he takes it and says, yeah, we can do it or Alex, you're crazy. One of those two results. So I hope he can do it. And, you know, I don't do wipe coding. I'm not there yet. So Jack is here and I'm still below. Craig (59:41) Yeah. I'm shocked. Shocked actually. Alex Goldovsky (59:49) Yeah, no, no, no, not yet. ⁓ No time, but it's on my plate to go and learn. Jack BeVier (59:54) It's not that hard. You'll love it. Alex Goldovsky (59:56) Yeah. Craig (59:57) For a smart guy like you, I can't imagine that your prompts wouldn't be very structured. Alex Goldovsky (1:00:01) The prompts are structured. So we use 14 AI technologies in our AI business, and they're all multiplexed together. This is a freaking, we're three years ahead of the competition. That's what rating agencies told us, three years ahead of competition. Everybody else is stuck on the old platforms. ⁓ Craig (1:00:18) Mr. Viver, Jack BeVier (1:00:19) Nice. Craig (1:00:19) anything else for you? Jack BeVier (1:00:21) No, man, I'm good. You know what? Hey, Alex, you'll enjoy this. I vibe coded a little app that pulls together a bunch of compliance things that like aren't being paid attention to the Baltimore fraud being like the you know, the impetus for it. So like, in Baltimore City, you you have to have a rental license to legally rent a property that that is a Baltimore City requirement. That's part of zero DSCR guidelines. No one checks. whether this rental, whether this DSCR loan is secured by a licensed rental property that's not part of anyone's guidelines. You know what the non-compliance rate of a loser gold was? 57%. 57 % of those mortgages were originated, to, you know, secured by properties that did not have a rental license. No one checked it, right? Like that wasn't part of anyone's guidelines. Alex Goldovsky (1:01:11) Does Baltimore City has a violation at the township or municipality office to say that, you know, this property has like a daily fine because the license is not there? Craig (1:01:22) It's a flag in a website that no one checks. Yes, they have it. Jack BeVier (1:01:25) Yeah, it's yeah, it's funny. And I was I literally testified at a subcommittee meeting last night in Baltimore City, ⁓ literally on this subject, because I'm like, it's been seven years since you implemented rental licensing and you haven't sent out one citation yet. No one. And the fact of the matter is that 40 percent of the rental properties in Baltimore City are rented without a rental license. So they have a the city itself doesn't enforce its own laws at the moment. the moment. There's some there's some well intentioned city council people that are trying to change that. But ⁓ but at the moment, it's a gaping hole in compliance. But it is highly predictive of whether you're just a good guy or not as to whether you get a rental license. So whether you get a rental license and whether you get a lead certificate, because that's something that the state of Maryland requires, and whether and whether you pay your water bill, because water bills in Maryland are a super priority lien that prime the prime first mortgages. So those three things, if you're an asset manager for DSCR loans and you have any assets that are in Baltimore city, you should be checking is the rental license current is the lead cert current and is the water bill paid? Cause those are your three, highly, I bet you a hundred percent of their default. Like the correlation between violation of those three things and anything that goes into default, like is like the correlation is 0.95, right? Like that's there. That's it. So I vibe coded a little app that checks that and I just sent it to all the DSCR loan purchasers earlier today. I was just like, Hey, here, just get, just give this to your servicing department. Like somebody do something. Craig (1:02:57) Alex, just so you know, he's been talking about this for a few weeks now and it's a very exciting project that he was able, honestly, Jack, you were able to build that in like a weekend and then obviously revision it out over time. But when we talked about it initially, I was like, Jack, what are you gonna do with this? Cause you're feeling a little snarky right now. Like I feel like you're just, you're gonna put, and so little. Jack BeVier (1:03:15) That was, yeah. Alex Goldovsky (1:03:17) I smell monetization somewhere, you know, how do you make money with it? I just didn't figure it out yet. Craig (1:03:21) First of all, it starts it starts out with like, do I report all of these sons of bitches who are like, they're not playing by the game to the Baltimore city, which I think may have happened last night at a meeting that I wasn't invited to. And then second, do I call all the DSC guys, DSCR buyers and tell them about this? That's the monetization play obviously. Jack BeVier (1:03:42) Yeah, Goodwill. Alex Goldovsky (1:03:42) This is similar story. City of Buffalo, New York has this little fee for rental properties called sewage license. It's like 50 bucks a year and nobody knows about it. And it's a super lien. It can foreclose out the mortgage. And I've seen one mortgage got wiped off because somebody didn't buy that sewage license for the rental. So that was so weird to me. I bet that every jurisdiction Jack BeVier (1:03:46) Mm. Alex Goldovsky (1:04:10) Has a similar thing that you're referring to. Yeah. Jack BeVier (1:04:11) Yeah, it's all little things. Yeah. Craig (1:04:14) attacks. Yeah. Jack BeVier (1:04:15) Anyway, yeah, just a nerdy nerdy nerdy things you do when you don't have kids and you get a bug up your ass about ⁓ somebody stealing your money. Craig (1:04:24) Alex, I don't know if you know that we've started a podcast just for AI called forked.ai and we do it with a friend of ours, David Moses, who is ⁓ really building some amazing stuff for his business. Jack. Alex Goldovsky (1:04:38) How do I know that name? What is he doing? Jack BeVier (1:04:40) He's in Detroit. runs an HOA management company does some real estate investing as well. ⁓ He's been on RAR before as a guest and now he's a co-host on Fork. Alex Goldovsky (1:04:45) I know that name. Craig (1:04:51) We'd love to have you on that because you're just a great mind in the whole AI space, Alex. So thank you for your time today. Jack, anything else, Jack BeVier (1:04:53) Yeah, that'd be fun. Alex Goldovsky (1:04:58) I'll be happy to, of course. Jack BeVier (1:05:00) No, thanks for your time. A fun conversation as always. Really enjoy it. Alex Goldovsky (1:05:04) As always. Craig (1:05:04) Jack, honestly, I'm just going to say when Jack and I dreamed up the podcast of like, hey, let's just get on the podcast and just fight with guys like that today was today was like one of those great days. So mission accomplished, Jack. Jack BeVier (1:05:11) Yeah Alex Goldovsky (1:05:14) All right, next time I'll be harder on Jack Jack BeVier (1:05:17) You Craig (1:05:17) Alex, honestly, thank you so much. Is there anything that you wanna tell folks to go and find you or the books that you're writing or the 12th patent that you're going to get across the? Alex Goldovsky (1:05:28) Forget the patents, the books, you know, if you don't know the can read it. ⁓ You can probably find me on LinkedIn and ask questions. I have a podcast as well. typically I invite the ⁓ mortgage community guys. It's really geared towards AI and mortgage community. Just talking specific hardcore mortgage news and the economics around it. ⁓ So, and I don't do it to make money. I just do it for the heck of it, just to... like you said, to battle or to find out more interesting information from folks. But I'm very passionate about developing new products. So whenever I get to the podcast with you guys, I'm super excited. You know, I already wrote down two products that I'm going to think about. Monetize it, Jack, not give it out to all the people in DSCR. Monetize it. No. Jack BeVier (1:06:12) yeah. Hell yeah. Craig (1:06:18) And we get no bips, no bips off of it none whatsoever. Alex, thank you. Seriously, it's it's it's such a great pleasure to have you back on the show. Love talking with you always. Honestly, I went back and I don't listen to a whole lot of podcasts admittedly, but I did listen to the last one that we had with you and I loved it. And just thank you for your great thoughts and ⁓ really appreciate your time, sir. Thank you. Jack BeVier (1:06:21) You Alex Goldovsky (1:06:42) Thank you so much. Craig (1:06:42) Folks, that is Real Investor Radio for today. Hope you enjoyed it. We'll welcome your comments. Please send them along to craig at the dominiongroup.com or jack at the dominiongroup.com. You can find us online everywhere where podcasts are found. We'll see you on the next one. Thanks.