 Percolate on a particular mix of odd and inspiring news headlines in Windy's Coffee House. Newsmakers with a pen shot for the unknown, unexplained and unusual share their experiences with UFOs, ghost encounters, near-death experiences and more for your own unique blend of Windy's Coffee House curious. And now, here's Windy. Glad to have you along. There's so much going on this year. It's just really, really already going by fast. Now we're going to go into the truth or consequences of lying. They're going to detect whether you can lie. Converis, the eye detect, my guest, next on KCMO Talk Radio. KCMO Talk Radio, Windy's Coffee House. Thank you for tuning in. I think you're going to find this interesting. These are new products, new technology, and we are really going into, I guess, future. Future escape when we're talking about this one. This is called, and we're just going to do an overview, something that is better than a lie detector. It's called eye detect, all right? And if you are not familiar with how a lie detector works, because not everybody is, and we haven't had to undergo all this, but what this does is give you an almost an immediate, an account of what's going on with your eyes. And maybe you've seen it in, there's, I think, remember Stargate. Is it Stargate or something like that where they have a little character and they put the eye on them and the little camera detects whether they're lying. It's some kind of totally fantasy, sci-fi fiction. Well, guess what? Not anymore. And so to talk about this, I have one of the authorities here, because I wanted to figure out what direction it was going and what we could count on, because this could be something that's a game changer just in the realm of any kind of legal, in terms of, you know, if you're putting the polygraph off the chart and putting this in there, then it also goes into human resources. It goes into all sorts of technologies that we will soon find antiquated as a result of this new product. Todd Mickelson is here and he is with the company Converis. Is it Converis or Converis? It's Converis. Okay. Yes. All right. Todd, first your background. What do you do and what is this product? So I am an entrepreneur. I've been involved in several startup companies, mostly in the high-tech area over the last 30 years. And this is the latest company that I've been involved in. I'm the CEO and we saw a great opportunity to take some science that came out of the University of Utah by the same professors that digitized or computerized the polygraph in 1992. They began research in this area of alternative ways to detect if someone is being deceptive or if someone is lying. They were seeking something that is non-intrusive, something that's accurate, something that's immediate, something that's non-biased. In essence, an alternative way to detect if someone is being deceptive. And what they found is the eyes are, in fact, the windows to the soul by analyzing changes that occur in your eyes as you answer a series of questions presented to you on a computer. We can detect with a high degree of accuracy if you're being deceptive about the answers to the items that are presented to you. So to compare it with a person, a human, who is looking at just to try and detect whether somebody is lying has about, according to your statistics, 54% for detecting a liar. So that's like a coin flip, 50-50. That's correct. There are a lot of individuals who presume that they can tell if someone is lying based on their facial expressions, based on their mannerisms. Do they look up or right or look down to the lower left as they answer a question? The science and the studies that have been done on that type of approach indicates that on average, people are only 50% to 54% accurate in terms of determining if someone is lying in that situation. The exception would be a parent or a spouse perhaps who really knows the individual and can detect those tales. What we do is we analyze changes using an infrared camera in the eyes that you can't see with the naked eye. And it's based on the premise that when we lie, it takes more mental effort, more cognitive load, if you will, to tell a lie than to tell the truth. And that increase in mental effort, even though it's slight, is exhibited through our eyes. It's exhibited by them dilating up to a tenth of a millimeter. So once again, something you can't see with the naked eye, but with an infrared camera and eye tracker that's capturing these minute changes up to a hundredth of a millimeter at 60 times per second using a computer algorithm we can detect. We can first of all capture those minute changes during the course of presenting a series of simple questions to the individual on the computer screen. And then we can analyze that data that's captured instantaneously with an algorithm that tells us with 86% accuracy or more whether or not the person is being deceptive about that topic. And how does that compare with a polygraph? So the polygraph has a range primarily because with a polygraph you have a human being that is presenting the items and analyzing the results. In the case of a really broad set of questions like what would be presented in a pre-employment screening test on topics such as, you know, have you used illegal drugs say in the last five years? Those are referred to as screening exams and the published results from the American Polygraph Association indicates that the accuracy is between 72% and 82% for a screening exam for a specific incident exam or an investigation. Something that is more specific like were you at the scene of the crime when it occurred? That range is anywhere from 80% to 90% accurate. So we're 86% accurate across the board and it's consistent. We don't have any inconclusives. So that's another difference in the case of the polygraph. They have up to 15% of the tests that are given that come back as inconclusive where there just isn't enough data to make a decision one way or the other. In all of our tests that we have performed in the field and in the lab to verify what our accuracy is we make a determination one way or the other in all cases. Okay, so if somebody comes back and says, but it didn't work, it malfunctioned. Well, so if anyone says, you know, these approaches are 100% accurate, they're not being truthful. These tools measure physiological changes against the baseline and in some cases for one reason or another that we can't explain. Some people react physiologically different than others and that's why they're not 100%. They really should be utilized as a data point for making a decision as opposed to a definitive decision upon which you take action. So, you know, we don't get it right every time but it certainly is a data point and we believe it's more consistent and it's more fair because it's in essence a sensor that captures the data and then a computerized algorithm that determines whether the person is credible or deceptive. Okay, and the reason I find this so fascinating is because this is probably, you know, a template for future types of tech that will give us a lot more information about our world than we have had available prior to these types of programs but this is a game changer. It's a real game changer. I've got to take a break but we're going to come back with Todd Nicholson who's talking about Converis and the product that we're talking about is eye detect but there's another one called identity detect. All right, and we're going to get into that too and find out for one thing, you know, what situation you would be facing if you had to take one of these little tests and what, you know, what that might reveal all across, I guess, our government, our human resources and just general basis of determining who you're going to hire for the next program, corporations. There are a lot of uses I can see with this. So, hang on, and again, Todd Nicholson, my guest, talking about the new AI using some form of computer program to determine whether or not you're lying or you're telling the truth. All right? Back next with more of that on Wendy's Coffee House, KCMO Talk Radio. KCMO Talk Radio. Hi here, I'm Wendy, my guest. We're getting into the truth of the situation. Truth or consequences? Truth or lies? Deception? Are there ways to get around these kinds of things when people, you know, you know, darn good and well, they are not telling the truth and you don't want to call them on it, but it's really obvious. And is there another way that you can do that gently and still forcefully enough to say, see ya, you know, and I'm not going to say the word, but you know, the BS detector, how's that? Well, there's a pro, actual product out now. Converis is the originator of this product. It's called iDetect. It is better than a lie detector, a polygraph, okay? A better lie detector than a polygraph. Maybe I should say it that way. My guest is Todd Mickelson and he's going to tell us about the product and more, there's another one called iDetect. These things are coming on the scene. They are being incorporated into current use, whether it's government or corporate or all across the globe. This is just the beginning and that's my view of it anyway because these things are so effective and they take the middleman, the human factor out of it by, you know, are you having a bad day at work? Did you read it wrong? Did you misplace the equipment? Did we have a lightning strike? That kind of stuff. Todd, this is, I think in so many ways, a real game changer in terms of getting, in corporate use or in government use, intelligence use, a product that will give you a very extremely accurate result when you're trying to figure out if somebody's telling a lie or telling the truth. I'm kind of frightened, to be honest. Maybe I don't want to answer these questions. Intimidated, how's that? You shouldn't be intimidated. Hopefully, you know, our goal in producing iDetect was to help protect countries, corporations and communities from corruption, fraud and threats. And those are the areas in which iDetect is being used today. We have over 400 customers across 40 countries running tens of thousands of these exams today. And I can just give you a few examples in each of those areas where iDetect is being used. First of all, to protect countries or us as people, we have some governments that are using iDetect to verify if people have ties to terrorism. Because it's a product that is consistent and computer-based, it can scale, it can be used on hundreds or thousands of people. Very quickly, the test is less than 30 minutes, and basically the person sits down in front of the computer where there's an infrared eye tracker, and they are presented with the same questions that they'd potentially be presented with in order to get a visa. Those questions would include things such as, do you have ties to terrorism? And with iDetect, within minutes after the test is completed, we can know whether or not the person potentially has ties to terrorism. It's an actual test that is being used today by some governments who are wary of people coming into their country that may have ties to terrorism. We have corporations and law enforcement here in the U.S. that use it for screening potential hires, so it is more of a pre-employment screening situation. What are they asking? They're asking whether or not, in the case of law enforcement, the individual has issues with illegal drugs whether they have specifically consumed illegal drugs in the last two years or last five years, depending on what the role is. They also will ask whether or not they have been involved in any serious crimes, crimes that would be characterized as a felony or a class A misdemeanor. And finally, where they have worked, where they have gone to school, and where they have lived. Things that go into doing a proper background check if they haven't been truthful about some of those pieces of information, then the background check won't be as effective. And so those are the typical things that they're screened for in a single exam. So if somebody was to embellish on their little curriculum on their resume and say, well, I got a Ph.D., well, maybe not that high, but I went to a certain school that's better than another school and I finished the program or I also am fluent in certain other abilities. And then as they're talking and responding to those questions, you're going to know that doesn't match without even looking at the piece of paper. That's correct. Now, obviously, the tests need to be constructed in a way that there's clear differentiation in what you're asking. We all know that people lie or embellish their resume, for example. They may make it sound as though they were the key contributor on a given project that generated significant revenue for the organization. In reality, they were a member of the team, but they weren't the leader perhaps. Those kinds of things are hard to detect. And so we focus on more discrete things such as, you know, did you lie on your application specifically about where you went to school? Oh, lovely. What degree you received, where you lived, or where you have worked. And those are more cut-and-dry things that we can test on that the system will register properly. And we know with a high degree of accuracy if the person is lying. Now, the other thing that we didn't get into the intelligence that identity detect. Now, is that different than eye detect? It is, yeah. So eye detect uses an infrared camera, as I mentioned, to capture up to 60 measurements per second as someone is sitting in front of a computer answering these questions that are presented to them. And they're simple questions to which they answer true or false. Identity detect is a product that we designed that could be run in an online environment or via a mobile app specifically to verify if the person is who they purport to be. And as you can imagine, that scenario is more specific. It would apply in a situation where perhaps I'm getting set up with a financial institution with a bank account or with a means of transferring funds from one account to another. And it's really important that the organization confirm my identity. It may be a government organization that's interested in whether or not the person asking for a visa to travel to that country is in fact who they purport to be. And so the test needs to be, you know, no more than three minutes. In some cases, you know, as short as a minute. And it needs to be something that can be done over a standard web browser or a mobile app on a smartphone. And so we don't have an infrared camera that can track changes in the eyes. Rather, we capture subtle variations in their motor nervous system as they interact with either the app or the web browser to where they're entering the information. And much like our eyes change, when we lie, there are physiological changes in our motor nervous system that affects how we respond to a form that we're filling out when we're lying about the information that we're putting in that form or how we navigate a web page and click on various buttons to respond to a series of questions that are being presented to us. And it's those subtle variations that we capture. Okay, we're going to take a quick break. I'm going to come back in a minute. But the press release I saw, Identity Verification Technology by Converis only takes three minutes, okay? So maybe you can address that when we come back out. Because that's amazing. Amazing. It's that quick. Okay, so hang on. Again, we're talking about Converis and truth or lies or maybe consequences. Next on Wendy's Coffee House. Can we get what we deserve? KCMO, talk radio. Hi there, I'm Wendy and my guest who's with me talking about lies, deception, dishonesty and truth. And they've got a product that will figure this out very quickly with pretty much... I mean, this is really accurate. 80% or 86%, I'm sorry, for Converis, the eye detect. And that is just mind-boggling to see that this technology is... And I'm sure it's going to get even better as it continues. This has only been... This is pretty much relatively new, what, the last three or so years. It was released in 2015. But we're talking about the kinds of things it can be used for. And when I was looking at one of the press releases that came out earlier, that Identity Verification, this is a little different than the eye detect. The Identity Verification takes three minutes. I guess I should say, how does that change things now with the current arsenal of identity, whatever we have at our disposal, Identity Verification systems? Is that like a leaps and bound forward? Well, I think it is. There is a lot of challenges today with people's identity being stolen. And there are new technologies that have come out that help with some of that. Most of those technologies, however, don't address the core issue of, is this person really who they purport to be? So I can steal someone's password. It can even create a fake federal, ID that I present that has my picture on it, so it looks like me. And those kinds of things can be beaten. What we've done with Identity Detect is provide a way in which, in under three minutes, we can present either a series of questions to someone on a web browser or through a mobile app that they respond to that will help us to know with 90% accuracy or more whether or not that person is really who they purport to be. And imagine, if you will, coupling this capability with some of the other traditional methods. Organizations are now moving more towards biometrics using facial recognition, using fingerprinting, even an iris scan. Well, that biometrics is only as good as your ability to know from the get-go whether or not that person is who they purport to be. Because if you take their fingerprints and they're representing someone else, obviously those fingerprints are going to be validated each and every time that person presents him or herself. What you want to do is capture at the same time you capture the fingerprints whether or not they are who they purport to be and then couple those two concepts together so that you can then just use biometrics moving forward to verify that the person is who they purport to be. With this, unbiased, nobody's there. There's no middleman. Again, this is technology. This is all computerized. That's correct. There is no test proctor like you'd have with the ID Tech test to make sure the person is following instructions and sitting in front of the device. In this case, literally it is embedded in the workflow of your process. So if you're applying for a visa to enter the country and the government agency that is verifying your details, they can embed identity detect in the context of that visa application process and present you with a series of questions to verify if you really are who you purport to be. The same would be true for a banking transaction where today you may get a text message that comes back saying, did you just perform a given financial transaction and before the financial institution allows that to go through, they wait for confirmation back from you via text. Someone could have spoofed your phone, logged in as you, and be responding as you. Identity detect could be coupled with that capability where there's an additional step in the confirmation process and we present through that application a set of simple questions that you respond to or a form that you fill out and we capture your motor nervous system subtle differences as you're responding and we compare that against your baseline and can now in less than three minutes if it's a forms-based prompting then it's less than a minute whether you are who you purport to be. Okay, I'm assuming that this technology is basically improving as you begin to roll out these kinds of programs because I think identity detect was after the eye detect and do you see these things changing significantly in the next five or ten years? Absolutely, I mean on a monthly basis we come out with new capabilities and improvements in the accuracy as we run more and more tests and capture more data associated with real-world situations we can use machine learning to harness that data and improve the overall algorithms and the accuracy. Are you seeing people switch over now to eye detect versus a polygraph? Yes, in certain situations where organizations would really like to know if people are being deceptive but yet a polygraph is too intrusive you get hooked up to various cables if the exam is done properly it's a minimum of 90 minutes that just isn't feasible in some situations whereas eye detect is a non-intrusive presentation of items that people are used to responding to on a computer they're filling out a job application for example presenting a series of simple questions to you that are pertinent to the job and you simply respond by interacting with the software and while that's happening we capture it. I felt less intimidated when I was reading the list of things if I would be filling out topics related to bribes cybercrimes, drug trafficking and money laundering unauthorized transactions I think I can answer those questions That's correct It won't nail me on that I think in terms of game changers what you're talking about with this technology and coming online other things that will probably follow that very quickly I think you're going to be very popular So people do want to connect It's Converis, right? It is Converis.com Truth validated is a nice mantra That's really the goal We believe that with truth societies will be safer organizations will be more successful We're helping create an environment of trust and amongst our customers who have deployed iDetect in a broad way they've seen that to be the case They get different classes of applicants people who know that they're coming to work for an organization that rewards and cares about the integrity and trust of their employees and that has an effect overall on their success because studies that have been shown where there's fraud involved indicate that in 70% of the cases that fraud is associated with someone internally with an organization So if you can cut that out that plays a major role I appreciate you talking with me Thank you for your information Converis, truth validated, my guest Michelson, thank you