Episode Transcript
[00:00:00] Speaker A: All right, we live. Excellent. Welcome back everybody. I'm Gregory and this is my buddy Paul.
Say hi, Paul.
[00:00:07] Speaker B: Every single week we do something new.
[00:00:11] Speaker A: Yeah, it's a Gregor and Paul show where we break down plays, SaaS, startups, AI. Sometimes we have guests. This week we have Anisu and he is from a company called careshield. AI and I believe the two of you are actually neighbors or were neighbors.
[00:00:26] Speaker B: Yeah, that's correct.
[00:00:26] Speaker A: In Toronto, right?
[00:00:28] Speaker B: Yeah. I'm really excited to have you on the show, Anisu.
Let me. Maybe I'll do a little bit of intro and then we can get into it. So I, you know, Doge dumped a large Medicaid data set a few weeks ago or a couple months ago and you actually have a company that basically catches fraud specifically in the medical industry.
Welcome onto the show. I'm super excited to talk about this with you today.
[00:00:55] Speaker A: Hey. Hey. Before we dive in, I would like to hear the origin story about how you guys met.
[00:01:01] Speaker B: Yeah. Okay.
[00:01:02] Speaker A: You want to start there and then we can talk about. Yeah, I'm curious, yo, like, did you go together? A cup of sugar?
[00:01:07] Speaker C: Sorry, because. Hi everybody, I'm Anesu from Keshu. AI But I met. We live on the same corner. That's right on the 62nd floor. So one day a package came to my doorstep and it happened to be. It was supposed to go to a post girlfriend and then I was like, oh, let me, let me, let me go and just hand it because this is not mine, etc. And we became friends and you know, is with these condos you are literally neighbors and friends and you're always the elevators. We started meeting a lot in elevators and talking to the point where, you know, I love, I consider myself, you know, I love cooking and other stuff. So routinely, maybe once every month find some amazing dish, French dish, Portuguese, whatever we are trying out there and then say hey, nice. Paul and Kelly, can you come over for dinner and wine? Yeah. Great friends and a pleasure to be, to be on the show today.
[00:02:05] Speaker B: Yeah, it's awesome. So we live literally right across from each other on the same floor of the condo. We. You actually met my then girlfriend Kelly before you met me and we just kind of hung out and we saw each other in the elevator for such a long time. And then I think eventually we decided to be friends, which is awesome.
[00:02:28] Speaker A: Yeah, I mean I came to visit and we all even went out to dinner together.
[00:02:32] Speaker B: To dinner together.
[00:02:34] Speaker C: We're going to do it again. That was, that was a special place as well. So it was nice to get to hang out with you, but yeah, great stuff.
[00:02:41] Speaker A: Okay. I wanted us all to be friends and kick off this episode with that perspective in mind.
[00:02:47] Speaker B: Perfect.
So, okay, how about, how about this? What's the. Tell us a bit about your origin story. What made you decide to start Kirshio? AI.
[00:02:57] Speaker A: Yeah, yeah.
[00:02:58] Speaker C: So. So I'll start with the background, I guess. I've been always around insurance companies for some, for a while. My undergraduate was in actuarial mathematics from the University of Manitoba. It was very interesting because I had a passion for just working with numbers and analytics and just understanding how that world manages risk.
And then was very fortunate enough to go into my MBA and my, my master's in financial engineering at the Schulich School of Business. And I'll tell you that story because that's where I met our Cato and my co founder. So we managed to build a very good friendship in graduate school and you know, we traveled to the. All over the.
The US together. And so the origin story was I had read an article in 2020, this was an article by the Global Global News and the Toronto Star. So they did sort of like an expose and it mentioned how the Ontario Drug Benefit Program was losing over half a billion dollars a year to crooked pharmacists, you know, in the province. So we decided to create a blockchain prototype that involved the chain of custody and using self sovereign digital identity. So this was a process in which you would be able to validate yourself in real time. So we built that infrastructure and we managed to go to San Francisco as part of an incubation and it was pretty clear that blockchain wasn't ready for adoption at the time.
So we immediately started thinking around pivoting and lo and behold, Covid came along.
Covid came along. We started to see a lot of services in the home really become to proliferate and people just starting to, you know, people didn't want to be in skilled nursing facilities in hospitals. They preferred getting care in the home. So we decided to start a company called Keshe Udai in and around the COVID times of 2023. And Kshudi is a SaaS company that is focused exclusively on home care fraud, waste and abuse years. So in 2025 we went and successfully completed a nationwide proof of concept with a data set from Komodo Health and we identified over $80 million out of the $2 billion of healthcare spend. So this was around 4% of the overall healthcare spend that we found in this pilot Data set. Then upon completing that pilot, we also managed to work with a company called Paradigm Senior Services.
So this is a large personal care provider vendor for Veterans Administration. So this is now the government. And working with the government, we did a pilot with them and we also identified over $8 million in overpayment. So now we are working with ParadigmC. We bid together in some of these contracts, you know, in RFPs, to try and help the government understand and shape this market. And on top of that, to show how our solution has evolved, we completed a successful pilot with one of the largest healthcare insurance companies in the United States where they gave us their data set around home and we were able to deliver tangible savings such that they went forward and they are looking forward to implementing this solution and going forward with it. So we have a very good understanding of the healthcare ecosystem, how our solution interacts with other data set, and the depth of the problem of fraud, waste and abuse in healthcare.
[00:06:46] Speaker B: Okay, so, so let's, let's kind of dive into that just a little bit. Right. For, for people that don't really understand healthcare, what does, what do you really mean by fraud? What does that actually mean?
Can you kind of describe the end to end process where this happens?
[00:07:04] Speaker C: So there are definitions and I think this is a very good question. And this is the difference between what you can normally hear, because when people hear fraud worse and abuse, they normally assume, right, talking about the same thing, fraud, Western fraud has intent. So this is something that can be prosecuted. This is something that goes for special investigation. So an example of a fraud worsen abuse is a kickback. Right. So, you know, there's a funny story about fraud worsen abuse, which is like, hey, the pharmacist owns the building.
Here's the pharmacy down, you know, in the, in the first floor and the doctors are upstairs sending all the prescriptions downstairs. Right. So essentially you can have sort of like a kickback system. I see that happens right where now you generate a lot of that is fraud with intent, kickbacks, you know, falsifying of claims. And then there is west, west, which is something that can be recoverable. So this could be something like duplicate billing, right. It's not necessarily flawed. I wouldn't persecute it for you, but I'll recoup my money, you know, in the process. And as we talk about solutions, how to solve this, it becomes much more complex because you no longer need just, you know, a hammer to, to get to the problem. You really need to be surgical because each of these things can cause A different set of abrasion in the revenue cycle management in the way care can be administered in the hospital, which can be a big problem in the long run. And then there's abuse, which is, hey, I'm prescribing you more injections or more things than you actually should be getting. Well, I am billing a certain service within a certain regional code. So one of the things that we look at is, hey, why? Where are you billing relative to your peers within the certain zip code or area code for certain services that are coded the same. So if you are billing a level 4 and 95% of all the people offering that service in your area are billing at a level two, that could be seen as wasteful or abusive.
Are you offering more services to the client than sometimes other options would be there and available? So the definition of fraud Western abuse is different. Fraud intent. West is a little bit of a gray area where you. You can. People are making errors and then abuse has to do more with the patient.
[00:09:31] Speaker A: Oh, my. Oh, my God. I have to say something. Like, this morning on Twitter, saw someone I know pretty well. They posted a video of some girl talking about going to the dentist. And the dentist claimed that she needed all this work and had like eight cavities and refused. Went to a different dentist who said she had no cavities. And it was funny. I replied to that saying, like, over 20 years ago was the first time I had some dentists tell me I had a bunch of cavities. And I was like, I've never had a cavity. Like, there's no way. Like, I'm just not gonna let you do it. They wanna drill all my teeth and everything. And I did do it, and then I didn't have cavities. I went to another dentist and it was fine, right? And I was like, basically every new dentist I've been to has told me, like, you need all this work and I never get it. And now, you know, I still have my teeth, right?
[00:10:13] Speaker C: So.
[00:10:13] Speaker A: So it's like. It's funny because I was listening to you talk about, like, abuse and waste, and I was like, oh, my God, like. Because he replied to my response saying, like, how common is this, right? That, like, treatments are recommended or prescribed, that perhaps.
[00:10:28] Speaker C: So there are ways you can get.
[00:10:29] Speaker A: People don't need this, right? Yeah.
[00:10:30] Speaker C: The insurance company sometimes will have what are called prior authorizations, right?
[00:10:35] Speaker A: Correct.
[00:10:36] Speaker C: Is that, you know, if you need a certain services, you, you know, there has to be either a medical history that can justify that particular service or sometimes, you know, they can look at your plan of care which can determine what type of service you, you have or you're authorized to have under your plan.
So it's really interesting when you start thinking about, you know, you know, fraud, waste and abuse. Because when you are thinking about those cases where you go to the dentist and then the dentist says to you, hey, I know you came here for a cleaning but I can see that you need seven of your teeth done.
[00:11:10] Speaker A: I can tell my teeth pulled out.
[00:11:12] Speaker C: I'm like, that's right, for the next three, four weeks.
Right. And some, and sometimes, you know, they might be, you know, they might be billing something, let's say a filling, but they're billing it as a crown. Right?
[00:11:24] Speaker A: Yeah.
[00:11:25] Speaker C: Which is a more expensive billing. So those are some of the things that you want to see where they are kept. It is very important as well because I know we're going to get into this very shortly. But it's important to understand that there are different techniques to how you capture all three. Fraud, waste and abuse. Payers are working on trying to have solutions with less abrasion but still be able to capture them. And I'm happy to talk about, you know, as we get into doors to be like, what is the actual dollars here that you can get to and what type of work do you need to do to get super interesting.
[00:11:58] Speaker B: Okay, maybe walk us through how are, how did these companies, providers catch waste, fraud and abuse and how have they been doing it or are they doing it in it?
[00:12:11] Speaker C: So traditionally you have had what is called a pay and chess model, right, which is, you know, you pay for the claim and then you do the investigations and then now you've heard a bunch of lawyers and other people chasing together the money.
The problem with that situation a lot is that you know, by the time you're chasing for the money, the guy has already gone to his vacation, he's probably has his third wife and on top of that, you know, there's no money to go and recover. Right. So you are now the insurance company just paying a lot of money for lawyers operational costs and you're never going to get 100 cents on the dollar.
So what we have been seeing is that a lot of companies are now going into the prepay module. So this is the period between when they get the claim coming in and when the claim is processed so that it's ready to be paid. So that's the time. So let's say I submit, I am a home care provider, I submit my claim on Monday and I usually get paid on Friday. There are some algorithms and edits that you can apply between Monday to Friday that can help you capture certain aspects of fraud. And those edits are usually, usually ones that can be referenceable to a CMS guideline.
[00:13:28] Speaker B: So.
[00:13:28] Speaker C: Or to a regulatory guideline, because you don't want any abrasion in that prepaid. Because if you are dealing with millions of clients and you are only looking at them for a millisecond, so when someone sends an API, I want to apply your edits and algorithms in a prepaid solution. And they want you to have 100% certainty that once you trigger that process and stop those dollars, it's clean. So when you work with an insurance company during a pilot, that's why pilots are important. Before you work with a client, you are getting their data set, you're getting to understand their provider base, you're getting to understand their policies and how you can fit your edits and algorithms very early in the revenue cycle, which is.
[00:14:17] Speaker B: Yeah, got it.
[00:14:19] Speaker C: Then there's postpay. And postpay is an interesting language now because you can now still have 100% certainty with Postpay. Yeah, but your look back period is now longer than Friday. Hey, somebody was billing for services not in the home. But I can only verify that three weeks later. But I'm 100% certain when I find you now, I can still recover that money with 100% certainty. And then the last one are what are called SIU referrals. So in the news recently, you have had Dr. Oz sending letters to New York, to Minnesota, to California. And one of these things is, he's asking, I want to see your referral list. Are you doing the compliance? Are you tracking these things? In a way that I can feel confident that every dollar that is being sent out there is, is actually going to be efficiently used. And there's a lot of gaps in the system.
[00:15:16] Speaker B: Got it. Let's dive into the DOGE data set. So as somebody catches fraud for a living, is. Is Elon right? That fraud is super easy to identify.
[00:15:27] Speaker A: Wait, were you part of the DOGE thing? Explain to me what the context is for doge.
[00:15:31] Speaker C: So for. So Doge is, you know, in our world, what everyone understands, Doge. It was the government trying to find ways to get efficiency and using technology.
I have to say, initially, the idea of DOGE using technology in a lot of these spaces is actually important because you can actually.
These are companies that are very big, that are dealing with millions of data sets.
So having artificial intelligence AI companies Building platforms to capture value there is really important.
Now the Dodge data set in our world that was released in healthcare was one of the largest healthcare data released by the government. So we have to give credit where credit is due. That is really important. That data set has advantages to a lot of people in my ecosystem. So it gives, it was a provider centered data.
So you have to think about the provider. When I say provider, I'm talking about a doctor, pharmacist, your dentist, information around those particular people. It's so rich if you are trying to sell into those networks. So I've seen a lot of people who have benefited from that. But it also shines, you know, a spotlight on where the other services are. So the Doge data set, 80,000ft, it's good. But as you really try to stick the landing, you start to realize that that data doesn't give you enough information or for you to fit into the solutions that we talked about earlier on. Can you fit into prepay just looking at the provider? No, you need to understand the patient. You need that claim level, patient data so that you can say to the insurance company, this particular doctor has oral provider has a very flaring, you know, bring red light of like this is really outlier behavior. But then how do you drill down to actually recover the dollar? So you needed that independent, you know, so that's where we come in, which is you can do statistical analysis, but there's logistical, there's clinical analysis that you can use you to actually pull down the dollars. So when I go and do a pilot with a client and I say, hey, here's 6.6 million, we really work hard to say in this state, if I save you 6.6 million, it all has to be one that you can bring back. You cannot just have noise or false positivity.
[00:18:02] Speaker B: Could you maybe like explain what exactly was in the DOGE data set for people who hasn't gone through it?
[00:18:07] Speaker C: Data set released was basically the patterns of spending and what usually were the billing codes that were being done by almost every provider, doctor, dentist, home care in the country. And the idea around that data set was that if you just, if you applied outlier analysis, you could immediately see things that stick out like a sorta right, like hey, these are the hotspots where we are seeing a lot of issues. So for example, one of the issues that has been in the news has been the case in, in, in Minnesota with behavioral care. We did one of our pilot where we handled behavioral care and it was a very interesting solution because there's a lot of fraud in it. Some people use it as quasi daycare, you know, because you have a kid that is autistic, they have to get care after school or. And then someone just comes in and builds them for eight hours, you know, so there's a lot of that where if the parent isn't at home and the plan is, it's been covered by a commercial plan. Hey, okay, abuse the system.
[00:19:16] Speaker B: Can you dive into that?
What, what did you, what was the case in that? What does in real life look like?
[00:19:25] Speaker C: So in that case is I could go in and say, hey, we can get diagnosis. So you're in an area. So the Minnesota case is a very interesting one where you had almost a 3,500% increase in, you know, behavioral health services being offered in a very short period of time. That's crazy. Growth in services, you don't get that growth even in revenue. Right.
So that was the first thing. So what it meant was once somebody could say, hey, my kid has, you know, behavioral health issues, you know, they've had autism issues, they can then qualify for behavioral health services. So let's take a look at. If you were, you know, I think our age is pretty fair to say that, you know, you could, you were allowed to go with a peanut butter sandwich to school, right? Now you're not even allowed to do it because it's so, you know, a lot of people with peanut butter allergies and you have to. That's the same thing that has happened to this behavioral health, you know.
[00:20:26] Speaker A: You know, what did you find in the data? Like, did you guys look at it? Did you find anything unique?
[00:20:31] Speaker C: Oh, yeah, there was significant. There's a lot of abnormal increase that isn't just explained by. Because you have to think about how lines of businesses grow. If you're an insurance company and they give you a two year data set, their membership hardly changes a lot, correct? Yeah, right. They are not increasing their membership significantly by, you know, 10, 15%. You know, in some data set you have got like 70,000 in this year. Instead, maybe next year it's 72, 73. But you're not getting those big jumps. What you get in the data set is just a huge amount of services being built. And also it's not even proportional to the number of people that are coming in. So what it tells me, hey, there are some people either that are billing for services that are not actually happening. Right. So. And there is also no audit on some of these claims. So somebody can bill 24 hours.
[00:21:31] Speaker B: I see we had a case in
[00:21:33] Speaker C: one of the data sets we just looked at in, in the state of Georgia, and there was, you know, a provider who built for 72 straight hours behavioral health services.
[00:21:46] Speaker B: And why is that? Why is that out of ordinary?
[00:21:50] Speaker C: Because usually if you are going to provide services in the home, you are not staying with them for 24 hours with the kid. So some of the. But the thing around it is that people can bill and remember. This is the interesting thing about behavioral health with our work. You know, in home care, every other claim type happens in a centralized location. You go to your doctor, you go to your pharmacist, you go to your dentist. They don't usually come to your place. In home care, the provider is moving. So if someone is claiming to have given me 1720 hours of services, but they've traveled through multiple zip codes and multiple areas, I see physically impossible. And be able to go to all these places and provide the services for the timestamps that you are giving us. What is also great is that the government has done a lot of work. There's what is called the Cures act, which requires every provider to have a mobile phone and ping a GPS to say, aim at this location. But that also can be spoofed, right? So there's a lot of fraud out there. And I think it's right that we are focusing on it right now.
[00:23:05] Speaker A: So people are submitting fake invoices is one big aspect.
[00:23:09] Speaker C: Invoices, kickbacks, kickbacks, as I talked about, like, I always laugh sometimes because it's an actual interesting story that friends, how
[00:23:16] Speaker A: does the kickback happen? What's the mechanism? Like, is it a fake bill?
I like, I like the example of like, hey, I'm the pharmacy. I got these doctors. Like, how are they actually doing the kickbacks?
[00:23:28] Speaker C: So the kickbacks is because usually you have a rendering and a referring provider. Rendering is the one providing the service. Referring is the one sending, you know, the claims. So in the pharmacy doctor situation, it's actually interesting because having to rent a place as a doctor, multiple rooms is very expensive, right? You need to pay 10,000, 20 plus, you know, $20,000 to rent a month. If you have your own office, the pharmacist that owns that place and is a, you know, his pharmacy downstairs can say, hey, every claim has to come down to the person downstairs. So let's say now there's a client who has a service that might not be offered by that pharmacist downstairs. Yeah, supposed to be referred to a cheaper place that is supposed to Be, you know, a pharmacist.
I'm going to force that claim to go to my pharmacist down there because I get a kickback out of that.
[00:24:29] Speaker A: How do you, how would you detect that, that type of fraud? Right. So they own a building. It seems like it's a difficult one.
[00:24:35] Speaker C: So what you, what we usually do with our analysis, we do what is called provider member density. So we tend to look at how each doctor is interacting with the patients around them. And if we see certain claims coming from one type of other provider, say the pharmacist, just sending a certain type of claims with an unusual set of volume that can also stick out once you start doing the analysis. So that's an example of kickbacks. There's a lot of kickbacks in durable medical equipment because the person who recommends you to get a wheelchair is not the person who actually sells you the wheelchair.
[00:25:14] Speaker B: I see.
[00:25:14] Speaker C: So there's money to be made.
[00:25:15] Speaker B: I see. So really it is just individuals instead of massive businesses that are affecting a lot of this waste abuse. Right. It's, it's a systematic problem. And then you.
[00:25:27] Speaker C: Big problem, right.
[00:25:30] Speaker B: You, you said that one of your latest clients is that you're starting to sell into the government. Could you just kind of talk about the regulations or the difficulties or how easy is it to sell to the government?
[00:25:42] Speaker C: Well, the government is slow to buy. That's, that's, that's, that's, You know, they are, they also do a very good process of when they do an rfp. So there's what is called an RFI process that they usually start with, which is basically soliciting for ideas. They'll submit an RFI and then companies will run for that rfi. And usually the good thing about the government is usually the contracts are pretty standard. So you're being paid maybe upfront a certain amount of money. You're not doing performance fees. So these are things that are budgeted, you know, when they make an rfi. So now companies come in with all their suggestions and then the government looks at all those ideas and then produces what is called an rfp, which is called the request for proposal. So that process usually can take, you know, six to 10 months to a year. It's not always straightforward as well, because there are a number of things that, you know, can, you know, can win you the day. Sometimes it could be having a product that is distinguishable. So for us, with our very focus on home care, we are able to go to other bigger clients, big companies that are working with huge Federal contracts and say, hey, you can be the differentiator in your bid.
[00:27:02] Speaker A: Yeah. So what are the biggest. Do you focus on every category? Are there certain categories that you don't focus on? Right. You talked about home care, you talked about pharmacies, you talked about medical devices.
Right. Do you focus on all of them?
[00:27:16] Speaker C: So we are focused on home care because we think just home care.
[00:27:19] Speaker A: Okay.
[00:27:20] Speaker C: However, one of the things that home care allows us to do is that we deal with a lot of medical claims. Because one of the issues that you see in the air in the industry is a lot of this data is siloed. Right. So you are having pharmacy data, not communicating with hospital data, but people's journey of care doesn't operate like that. Right. You go to your doctor and maybe your doctor sends you to your pharmacist, but all of those things are happening within a shorter period of time. So one of the things that with home care that is exciting is we get to see all the claims that come in from the hospital.
So a big part of the fraud that we catch is called patient location conflict, where the patient is in the hospital, but someone is still bailing for medical services or for personal care. They're coming in and saying, I'm doing laundry and I'm cleaning. And that's especially personal care. Personal care is ridden with fraud. It's almost 20 times more than pharmacy, medical, or anything in home care. Because with personal care, the providers or people who provide care are usually people that are not licensed. Right. So if you think of people. Right. Care in the home. So you're having a home care agent saying, oh, you know, I don't care whether the person is in the hospital or not. I'm just going to bail and say, hey, you know, I went and cleaned the house, I washed, I did laundry, I provided. So there is a lot of that. That is.
[00:28:43] Speaker A: Wow.
Hold on. I want to make sure I got this. It's so you're like. It's literally, you have the record of when someone is in a hospital, and then all you're doing is matching that to the billing dates that other people are submitting for that person when they're not available in their home. How come the government can't simply, like, match that up? Like, that seems really basic, Right. So I'm coming from the perspective, like, I run a little company and, like, I know that, like, this. This bill is completely bogus. Right.
[00:29:14] Speaker C: Like, so we talked about the problem of the data being siloed, Right? Okay.
[00:29:20] Speaker A: Okay.
[00:29:20] Speaker C: That's the biggest Problem. So this data is siloed. So sometimes you have to request where
[00:29:24] Speaker A: the hospital data is over here and then the other data is over there. Okay.
[00:29:27] Speaker C: Home care is that some of the data also cannot come in at the same time. So for example, a home care agency have a revenue cycle window that means they get paid within a week. Hospitals usually submit their claims every 30 days.
[00:29:43] Speaker A: Yeah, yeah, that makes total sense.
[00:29:44] Speaker C: I have to match that data set and have a look back period. But also, here's the interesting thing. This is where I was going to go deeper into the doors. Yeah, the overall system is great. But then when you look into those information, there's a lot of policies, there's a lot of, you know, you really need to look at the data and understand how health care is actually delivered. Because it's not just looking at an anomaly like that and saying, oh, you know, somebody was at home. They can be also something when somebody comes into the hospital in the morning and maybe they get discharged later in the evening and we have to figure out what time was that service delivered. So there are nuances to when you start analyzing the data. There are things that sometimes people can have modifiers, which means they get authorization. So you cannot. That's where the skill set, that's where our company comes into play to say, hey, we have been working on this data set for over three years. We have built edits and algorithms configured for the entire United States. And we can go into any platform and give you actionable savings with little false positives in a very 30 to 60 days. Right.
[00:30:53] Speaker A: I want to hear how much in terms of dollars, how much fraud have
[00:30:57] Speaker C: you detected for companies so far? We believe that we are over 900,000. 900 million.
[00:31:04] Speaker A: 900 million, yeah. And then how much has been like actually recovered?
[00:31:09] Speaker C: So some of the ones that we have had are in the process of recovery. So the recovery process is also very interesting because a certain look back period. Correct. Because again, remember, in healthcare you're dealing with three different ecosystems. You have got the insurance, which is the payer, you have with the doctors and the providers, dentists, etc, and then you've got the patients, so you don't want to cause any abrasion. So if you start going back with a look back period of four or five years, unless you, if it's like, you know, kickbacks, etc. And you're sending them to jail. Right. In most cases your look back period is just two years. And a lot of these insurance companies, think about your local insurance company in a smaller state they don't have enough people to go and do these investigations so they end up just focusing on higher dollar claims. Let me look at the claim that goes into the hospital, but the home care claim that happens every day is a hundred dollar claim they don't have. I can't pay my data scientist, you know, $10,000 a month for them to just look at $5 claims or a $10 claims. Right. So that's where we and the edits, the algorithms and also a little bit of the human intervention on the clinical side can optimize our solution and our focus for these clients and say hey, give us your data set and we can become part of your engine in how you analyze data and give good reporting. So we have dashboards, they can report now to the government and they can do all sorts of things. So this is an area right now that a lot of innovation and I do think if you were to put a number, I honestly believe in the US healthcare system you are looking at almost, you know, probably $1.5 trillion. Remember they spent almost about fraud, waste and abuse.
[00:32:59] Speaker B: Fraud, waste and abuse.
[00:33:00] Speaker C: Almost close to almost 8 to 10 trillion dollars a year in health care.
[00:33:05] Speaker B: So the U.S. economy say 10% of that is probably at least wasted if it's not fraud.
[00:33:12] Speaker C: I know of cases that I've done in POCs and working with clients where there is a durable medical equipment store that is in the middle of nowhere.
[00:33:23] Speaker A: Why is it so easy to get away with?
[00:33:25] Speaker C: Because there's a lot of data in the system and they isn't. You know, a lot of these processes were manual for, for and by the time you catch the fraud is 10 years down the line when somebody does
[00:33:35] Speaker A: a very quick, the scale is big, the data sets are disparate, it's manual, it's like it's a number of different things. But like yeah, it's very logical actually. Like, and people have figured this out and they've been gaming it for a long time.
[00:33:50] Speaker C: Yes, for yeah, like I always laugh like you know political. It's always like, you know, I want to eliminate fraud worse and abuse. And it almost is like a throwaway line so that people can get away to the next question. I think the focus here is important. I think they are real dollars. But I also don't think it's just like in healthcare, in finance. I think if you look at housing, you know, fema, fema like the, the Fed, you know, it's an insurance program. Right. You know, when people, there's a lot of places where if you bring good technology into how those services are being delivered, you can possibly avoid people that are going to be wasteful or downright commit fraud.
[00:34:30] Speaker A: Yeah, well, as we saw, DOGE was ultimately not very popular if they'd gone
[00:34:35] Speaker C: after the fraud directly. Look, I, I think.
[00:34:38] Speaker A: Tell me more about that. Why? Why do you think that was? Yeah, I'd love to hear your perspective
[00:34:41] Speaker C: on this, because I think they went out, they went after things that didn't matter. Like, okay, if you're going, you know, if you go after USAID alone, USAID alone is an organization, might have flaws or there might be issues that you might want to think about how they spend money, but they're not a corrupt organization. Right.
So by getting rid of it and say, I'm getting rid of usaid, but at the same time, I'm trying to fight fraud in health care. Those two things are different.
So you end up having people that would normally support the letter. Nobody wants fraud worse and abuse in anything they're doing because it's costing them money. And then, you know, you had all the other nonsense around, like getting rid of organizations that, you know, instead of reforming them, which is what DOGE should really be about. You know, we're going to create efficiencies, we're going to create technological innovations to make sure that, hey, we get more for our dollar. If that was their, you know, their strict mandate and they wanted to go in and help these employees become more efficient. Efficient and train them to identify fraud, waste and abuse so that they're clean. They are not having false positives in their work. I think it would be popular.
[00:35:49] Speaker B: Right.
[00:35:50] Speaker C: I think there were some sideshows that didn't meet the mission.
[00:35:54] Speaker B: They should have gone after the problem, not the people, basically. Right. Like, they kind of went after the people a little.
[00:36:00] Speaker C: Yeah. Like I, I said, USAID is a very simple example. You know, a lot of people might have different opinions about the organization, but nobody actually thinks that it's a corrupt situation. Money laundering, let's get rid of it. So it gives you the impression that you are throwing the baby in the bath water. Right. And you're not being surgical when you are doing a lot of these things. And the chaos around that is what companies like us are trying to avoid. When we work with clients, we take significant care of their data, we take significant care of our results. Because I don't want to put a 25% false positive rate into my client's workflow.
[00:36:39] Speaker A: Exactly.
[00:36:39] Speaker C: So that is the kind of discipline that if DOGE had really Focused on, I think the technology, the embracing of AI they would have had more rob and long and in leadership too, you know, a little bit of empathy goes a long way. You know, walking around with a chainsaw. Well, if you're firing people might not be the best look when you're doing everything. So I think there are a lot of things that you have to just think about when you're looking at do
[00:37:06] Speaker B: you still think we still have time to fix all of those problems? Are we too late?
[00:37:10] Speaker C: So one of the issues that I'm hearing, I was talking to one of my, one of my. One of our investors and a very guy who is sold. He sold his Fraud wasn't Abyss company for $5 billion. He's invested in my company and he's a really great guy. And one of the things that he was worried about was sometimes it feels like the fraud wasn't abuse right now he's a little bit of a political tinge to it. Right. Like, you know, it's like the oath letter is grant. I wish you'd send it to all 50 states because, you know, that's that, you know, if he sends to all 50 states, I think there's this structural reform that benefits everyone in Canada. We're seeing that as well with, with governments trying to get reforms. But I think that's some of the issues that I'm starting to hear when you go for, you know, you go to a conference and you start having a cocktail with people on the sidelines. So when I am working with my company, I want us to make sure that they understand we are a professional company. We are going to the depth of analyzing, making sure the rules, the policies. You know, we are not integrating noise into your workflows. And we are not going to make sure that somebody who should be getting care is rejected, you know, because we take great care of that. So I think there is room for that.
I applaud what the government is doing right now to put a spotlight. I wish they would just send 50 letters to the whole country because then whoever is in the next administration, we start creating the seeds of like these are good practices for the government to run. We need to do basic audits on outlier analysis, understanding how are they kickbacks in our system. And I think that could be, you know, a good legacy for doge. If again focus on the actual how do we get to the dollars and how do we do it quickly?
[00:39:02] Speaker A: I like this idea. I think like I, I like this conversation. So let me summarize what I think I heard. Because I think it's a really great suggestion. It's like, look, there are some really obvious things that are totally fraudulent.
[00:39:13] Speaker C: Right.
[00:39:14] Speaker A: So if you're in a hospital and if you're at home and there's a conflict, it's really clear, like, oh, that logic extended to Doge. If they had just gone after very obvious things that are really easy to, like, demonstrate to the public that this is clearly like a waste. It's clearly abuse. Right. You can't be in two places at the same time. We have the data. Like, we know where you are. Right. They just kind of like focus on that. I think they would have won a lot more.
[00:39:39] Speaker B: Yeah.
[00:39:40] Speaker A: Support from the public. I do think it's important to say that, like, I think there's so much fraud and abuse going on in these areas that there are a lot of people in the. Let's call it the ecosystem who do benefit. So I think, like, finding a way to get the public support.
[00:39:54] Speaker C: So there's an interesting thing, Greg, that.
[00:39:56] Speaker A: Yeah, go ahead.
[00:39:57] Speaker C: I think you'll like this.
[00:39:59] Speaker B: So, okay.
[00:40:00] Speaker C: In rural areas, sometimes some of those providers or doctors can get more political sway because it's that they know they have leverage with the insurance company. Right. So that's where some of this happens. So, you know, you find 100 million, but the insurance company goes in and says, hey, you know, these are. This is the only hospital that we have in this.
[00:40:21] Speaker A: It's the only option in Kansas or whatever. Yes.
[00:40:24] Speaker C: Yeah, exactly.
[00:40:25] Speaker B: So I see, I see.
[00:40:26] Speaker C: They say, hey, you can only audit me once or twice a year. If we lose this contract, if we lose the client, then really, they say that you can only audit me.
[00:40:34] Speaker A: They give like a rule, like a number.
[00:40:36] Speaker C: Yes. That's what happens. Sometimes there are providers that can become too big to kick out. Yeah. If you let me give you an example. If I'm actually shocked by that.
[00:40:43] Speaker A: I mean, I'm not shocked about that, but I am shocked by that. I'm sorry.
[00:40:46] Speaker C: Well, but I'll give you an example. If your biggest hospital in the city is committing fraud, you're not going to get rid of the hospital in most cases, if it is a regular, you know, you know, Anu. Peter. Poor. Whatever. Doctor. Doctor Answer, whatever. And they are out there and they did that, they would be excluded and then they wouldn't be able to do business with the government and build with the government. But now you're dealing with the biggest hospital that might have been billing wastefully a lot of money. And here's the thing. Most providers don't have a lot of money. So let's say your local hospital over billed you by $20 million in a year. You cannot just go and take $20 million from them. It doesn't work that way.
[00:41:28] Speaker A: They don't have it just lying around
[00:41:29] Speaker C: on the floor, hey, how do we continue providing services? It's a little bit of like, that's why understanding the ecosystem. So when you said why didn't, you know, why didn't you just go and buy point? Why isn't Dodge like, hey, let's just go after everyone? It's because there are nuances to how those relationships work and you have to be careful as to like, hey, you know, sometimes there's provider abrasion. We don't want to have patients drop off. So it's a very interesting, you know, it's an interesting area. I'm, I think this is going to be the biggest part. Today, JD Vance shared his first, you know, kind of like, you know, meeting at the White House and they actually found a special prosecutor, you know, to try and. Because one of the ways they think a lot of fraud happens is that as we mentioned, some of the people in responsible, in power, right, aren't doing their jobs to see like, hey, why did we have a huge increase in behavioral health in thousands of percent over 3,000, you know, in a short period of time, that should have been easier for the top law enforcement officer of the state to say, hey, maybe let's, you know, let's take a look here, what's going on here. But, you know, so I think there's a lot of stuff that's going to come out. Who knows, there may be criminal, you know, prosecutions that come out of this, but I think fraud wasn't abuse is going to be with us probably going into the2030s. There's a lot of demand for fraud, Western abuse. If you go out to companies in healthcare, in other parts of government, you're seeing that, you know, people are trying to solve this problem. And I think that's a good thing, but I think we can do better.
[00:43:10] Speaker A: Yeah, I, the excellent point. I mean, I just keep thinking over and over in my head how like the Pentagon can't pass an audit. Like it's, it's like it's public. Like this isn't like some rumor or some like secret thing. It's like they literally are saying, no, we cannot account for how we spend all of the money and we're not really interested in accounting for it. Just too bad, right? Like Which I think is terrible, personally, that we can't, you know, have some level of scrutiny. I agree with you about, like, doing it the right way, showing empathy. Like, people have leverage too. Right. So, like, you don't want to, like, hurt the best provider of military technology in the United States. Right. Like, but we are not. We need to find a way to, like, yes, passing an audit sounds like a very reasonable thing to me. Right.
And I'd be very comfortable. Tell them that, like, I think, like, aiming to have an audit is good. Like, I would think we should do that, you know.
[00:44:00] Speaker C: Well, I think there's just so much money in the system, you know, there's just so much money in the system. You can look into a stat, you know, and you know, in some of these big stats, you're looking at over hundreds of billions of dollars that are being administered. And you know, and remember, you know, one of the things that we have a bias towards is efficiency in the system. So some of the process that we are talking about actually requires maybe human interlope in the process, and then that creates a business. You know, it's no longer just, hey, I want to be kind or anything. There's also a business conversation you want to have. How long do we want to keep this? What's the cost of us to really do fraud, waste and abuse? So you start to see how somebody looks at, like, hey, if I am able to investigate five providers and it costs me $50,000 a week, do I want to investigate a hundred providers? Nah, you know, I'll let some slide, you know, and I'll pick. Right.
[00:44:59] Speaker A: There's a cost to everything.
[00:45:00] Speaker C: Exactly, exactly. So there's always that cost benefit analysis also that's happening internally. So that's why if you can automate as much as possible and put everything in that prepay, you know, where you can get to the dollars without a lot of human, human clinical, you know, input, then things move faster. So it's not just, hey, there's. Is there corruption in the system? Yes. Is there provider issues? Yes. But also, this is efficiency. If I am the payer, I want the most efficient and fastest way to get the claim out there.
[00:45:35] Speaker B: So let's talk about the technology for maybe 5 minutes. Has AI meaningfully created efficient systems where we can start tackling these problems in leaps and bounds?
[00:45:46] Speaker C: This is one of. This is the golden age of AI. You know, I think the companies that can integrate not just efficient coding models that you can get through AI, but also really understand how to create a flywheel effect around the problem because you can generate a lot of kind of like analytics, but is it scalable? That's the other thing. And is it incremental in value? Because next time you might be dealing with a different set of providers. So what cachiot AI for us we do is for our clients.
[00:46:18] Speaker B: We use.
[00:46:19] Speaker C: We automate all the rules and edits across the country as they are happening in real time. So now when the client comes to me, I can provide the same edit with an update six months later and also additional dollars that I based on that increment. So I'm always adding incremental value ways in which you can implement the air. Now, there is also a part where, you know, when you are. There's a clinical part which requires a human being, but a lot of that as well is using generative AI. So, you know, you, you can ask inputs, you have dashboards that are very efficient to use. But this is. Yeah, companies that are focusing this AI is really opened the door for significant improvements in this space.
[00:47:03] Speaker A: Super interesting.
[00:47:05] Speaker B: Yeah, there's so many questions, but if you kind of, I guess, have anything you want or if you want to see the government kind of improve, what would that even look like? What can they provide you to let you do your job better?
[00:47:20] Speaker C: Well, I think right now I think you want the government to create the focus. So, you know, and I think this is why I'm applauding the government currently. Because the government is the biggest corporation in the country. And if they show a certain focus on something, it's, you know, companies pay attention, right. To what the government is doing and where they're paying attention.
They're, you know, spending most of their time. So I, what I would really want is to have the letters. Dr. O sent 50 letters. That's the first thing that I would want because then it would really put the spotlight on the entire country and then we actually benefit everyone. Right. And then this is the first thing. And then if once that happens, I also think, you know, make the RFP process much quicker, much faster and open to new companies. You know, a lot of times there's a bias where, you know, a lot of these RFPs, and I understand, you know, these are big contracts, so the government needs to know that you can deliver the service and you're not going to.
[00:48:19] Speaker B: I see.
[00:48:20] Speaker C: But I think part of it is also opening up innovation, just like in the military world. Right. You see them opening up drone companies that are not legacy, you know, Boeing or other, you know, they're bringing in companies to create and we're now seeing the benefits of that sometimes on the battlefield. And that's the same thing here, right. The government has to be open to having conversations to growing companies like us, right, and say, hey, you guys have a focus on this. Let's really work with you. And in some of these tests, show us what you can do and we can expand across the entire country.
[00:48:51] Speaker A: Yeah, some. I mean, so many great points. Awesome. This has been such an incredible conversation.
[00:48:57] Speaker B: This is awesome.
[00:48:58] Speaker A: Yeah, you make, you make. You make me feel like what I do is not very important. Like, I'm really proud of, like we're doing. I think it's like, it's a really great service. I think it's really important, right? Like, it benefits everybody at the end of the day to be able to, like, detect and reduce fraud, waste and abuse in any. In any system, particularly the healthcare system, right? Where we know healthcare costs are really crushing people across, across the world, right? Not just the United States, not just in Canada. Health care costs are everywhere.
[00:49:24] Speaker C: Canada, just keep going.
Let me tell you something. Canada is one of the worst at this because our insurance companies, they don't handle a lot of the claims because a lot of it is covered by the government. The medical cost, you know, only $48 billion is spent, you know, by commercial plans. So if you think, if you are manualized, $48 billion and you have a slice of that for pharmacy, dental plans, etc, you have hundreds of billions of dollars in pension funds and other things. So the resources sometimes that are allocated to some of these services is bad. And one thing, if I were to speak to our Canadian audiences and even stakeholders is, hey, we need to have a better reporting system. The US Is so good at reporting instances of fraud, waste and abuse years, you cannot even figure out who has been banned or not in the system. And I think that reporting, that standard of transparency allows people to see, like, hey, there's opportunities to create innovation in this space and we can add value.
So I think the US is, I would say, ahead by quite a bit in terms of how they report, how they focus on this issue. And one of the incentives, again, is because some of these big healthcare companies make billions of dollars through the government and healthcare spending. So it's an incentive for them to actually, you know, find efficiencies and savings.
[00:50:44] Speaker A: Awesome.
[00:50:45] Speaker C: This is. This is great. Now you have experts in fraud. You know, you can like.
[00:50:50] Speaker B: Yeah, okay, go for it.
[00:50:55] Speaker A: I was gonna say add it to the list of all my Twitter expertise, right? Like on Twitter, you become an expert on every topic, whatever's popular, whether it's fraud, waste, abuse, geopolitics, AI. Right. Everybody on Twitter is.
[00:51:10] Speaker C: Well that's, that's the interesting thing about the Twitter when this came out, right, Because I saw people on Twitter, you know, come up with more gels and
[00:51:16] Speaker A: then they do their own analysis.
[00:51:19] Speaker C: I hope you don't go out to the, to the guy's office tomorrow and start harassing him because you don't know anything. Exactly. What's really happening. You're just looking at me like hey dude, you're stealing my money or something like that. I know, but I think that's good though to start having because again, you know, how you get better is a focus.
You know, there's a lot of innovation out there. We think with our first prototype blockchain we have that. We think we are probably because as people start getting their data sets on their phones, on all of their things, you can now be able to authenticate at point of service where you are and that sets, you know, the chain of custody secure before money ever goes out. And it's actually could potentially be referenceable to a blockchain where now they've immutable characteristics. So the industry isn't yet there and we think there's still another step where you can really lock up a lot of these services and make them more secure.
[00:52:14] Speaker B: So you're actually making a case for the application of blockchain.
[00:52:19] Speaker C: Yes, yes, yes.
[00:52:20] Speaker B: Crypto might make a. Crypto on blockchain
[00:52:23] Speaker C: might make a comeback for his good use cases.
You know I always said like when we started the company I told you like with outlier vengeance we were looking at self sovereign identity. I think it's just like everything if you put the right applications I think you can get the issue. The problem with healthcare is that the insurance company doesn't want to give you your data. If they gave you your data blockchain would be immediate because the payer doesn't care because now you get your medical record, you get to validate your claim in real time and then it works. The only thing is, you know they make a lot of money with the data set. They just don't want to give it to you yet they have the data. Awesome.
[00:53:00] Speaker B: Well this has been awesome. I really appreciate your time. Thanks for coming onto the show.
[00:53:05] Speaker A: Yeah, thanks for coming on. That was fantastic. Next time in Toronto we'll be sure all about we should do something dinner again.
[00:53:12] Speaker C: Well, I'm planning as the weather gets better I'm going to be hosting, you know Paul and Kelly for, you know, a Wagyu.
Okay.
[00:53:22] Speaker A: Okay.
[00:53:23] Speaker B: All right.
[00:53:24] Speaker C: We're looking forward to that.
[00:53:25] Speaker A: Sounds fantastic. Paul, any announcements we have for our audience or.
[00:53:29] Speaker B: Yeah. So we will be hosting another in person event in or you will through vibe your SaaS on May 12th in San Francisco.
I'll be there.
[00:53:41] Speaker A: Excellent. Yeah, we'll be doing another event.
[00:53:43] Speaker C: San Francisco is my favorite city.
[00:53:45] Speaker A: Yeah, you should.
We've got that one set up for May. We have Snowflake in September.
There's already a bunch of people who've signed up. And for those that aren't aware of it, it is a startup pitch competition and a mixture of founders can meet with different investors, venture capitalists, corporate development people and get to have a conversation about like what the landscape looks like. Maybe they're even interested in investing your company. We've had 150 applications come through, so I expect to see some really excellent startups. And again, we will pick five, just like we did last time, and five startups will get to pitch. And Paul, I have a special guest who will be coming on as a judge. I haven't yet revealed who this person
[00:54:26] Speaker B: super excited about that is going to be.
[00:54:29] Speaker A: And, and yeah, it's going to be, it's going to be fantastic. And oh, last thing I'll say too is like, thank you for everyone who has been here with us. We're almost up to 150 subscribers now. Right. We're about to cross 125,000 views on the channel. It's nuts. Like, I, I can't believe it. It's exciting. And yeah, if you want to subscribe and follow and put funny comments below our videos, please keep doing so.
[00:54:52] Speaker C: Perfect.
[00:54:53] Speaker B: Thanks a lot. Thank you. Thanks again, Nisu.
[00:54:56] Speaker C: Thank you guys. Cheers.
