← Lead With AI
Lead With AI
This AI Prevents Murders While Protecting Privacy at 145,000 Locations
This AI Prevents Murders While Protecting Privacy at 145,000 Locations
Loading summary
Transcript64 lines
- [00:00]
A
What if your security system could sense danger before you even knew something was wrong? In this episode of lead with AI, I sit down with Alex Verkutios, CTO of eCamm, the company behind Blackout, a real time AI surveillance platform deployed across over 140,000 cameras and counting. This isn't your typical motion detection. We're talking about AI that flags threats faster than the human eye can react, sometimes in under a second, and is already transforming how cities, corporations, and critical infrastructure think about safety, privacy, and what it means to be secure. If you care about where AI is going or who's watching while you're not, this conversation is for you. Let's get into it.
- [00:53]
B
Welcome to lead with AI. I'm Dr. Tamara Nall. In each episode, we will take you behind the scenes with visionary leaders shaping the future of AI across public and private sectors. Join us as we explore groundbreaking projects and innovations that are transforming industries and making a real impact on people's lives. Let's dive in.
- [01:22]
A
Hi, everyone. How are you? It's Dr. T, your host with the most, on lead with AI. And today I'm so excited because we're talking all things security with Alex Vakutios, CTO of ecamm. Alex, how are you?
- [01:37]
C
I'm good, I'm good. Thank you very much for having me.
- [01:40]
A
Awesome, awesome. So, you know, security is such a big deal these days, and that's why I'm really excited about, about this conversation. In fact, we even had an intruder in our home about a few months ago. So I'm always like, I just have this scary bubble around me these days. And he just walked right in, I walked down the hallway and I jumped up. So we're all safe. Yeah, we were home and I jumped and went down the hallway and I was, who are you? Get out of here. And he looked at me like he owned the house and just kept walking. So thankfully we're safe. But I am so excited for this conversation just because it is near and dear, given my recent experience. So, Alex, I just like to start with what I call start with a spark, and that is, you know, every guest has some core essence to himself or herself. So tell us about who you are at your core and why you're so passionate about ecamm and all of your different products.
- [02:48]
C
Oh, man, that's a lot unpacking on that one, for sure. So for me, I mean, at my core, very family oriented. I have two daughters, happily married, have been for 20 years. And being a family man, I do tend to lean towards family. Environments, safe environments. When you're with your kids, you want to make sure that where you're going is safe. You had your own scare recently on this. And so I went to school for computer sciences and such. And so I had a bit of experience with that in the past. I worked with some family friends that had security businesses, traditional businesses like guards. And as you know, I'd be talking with them at gatherings and such. And the idea would come to me that maybe technology can be applied in this in a great way to create efficiency, provide a lot of just general optics into what's going on in different areas and make it cost effective enough where customers would be able to use it at multitude of sites if it becomes ubiquitous. Well, now you've created very safe environments. Right. And that was kind of the idea on where it started.
- [03:57]
A
Okay, amazing. And then talk to us about. Because our guests love this, talk to us about a moment where a customer uses Blackout. You know, it particularly would have been AI powered. And it just changes everything for them. They're. They're in that moment of, oh my gosh, where has this product been all of my life? Talk to us about one of those, wow, holy smokes moment.
- [04:20]
C
Sure, yeah. So, I mean, as the example I gave you in the previous question, that was 14 years ago that we really started playing with that AI. So it was early days.
- [04:29]
A
Wow.
- [04:29]
C
In the last 14 years, you can imagine that's a lot of evolution in this product and in this technology as a whole. So nowadays what happens is like we're, you know, Garda World is a large security company. Right. Offers in 86 countries around the world. It's a big business. And so we have a lot of customers now, commercial customers that come to us, they want to secure their environments as we do so. And the kind of, the first thing to your question is like, what's the holy smoke moment? Well, it's the amount of content that we actually generate for them. So what, what? From our perspective, every camera that's at a location is basically a set of eyes. It sees the same thing that our human eye sees. Right. What makes humans unique is that we can interpret all kinds of information that, you know, a traditional camera you'd buy off a store shelf won't be able to interpret. When you apply artificial intelligence specifically, if it's, you know, advanced AI, it can start determining all kinds of things for you. And so what we do is we generate content for the customer based on all of that that we observe. And I don't think any of our customers are prepared for how much information they're going to get from us when we turn our service on at their sites.
- [05:45]
A
Wow. And then how do you ensure that it's not, you know, data overload or content overload? Like, how do you do that? Because I can see myself. I mean, you think about how many seconds are in a day and how many frames that are available, et cetera, I could easily be overwhelmed. How do you ensure I'm not overwhelmed and I get to know exactly what I need to know or know what I need to know when I want to know it?
- [06:06]
C
Yeah, no, that's. That's the golden question. We. We call it managing expectations upfront with the customer. So, like, you know, the sales rep at the, you know, onset of. Of us talking with them about the service, we'll explain to them this is what we can do, this is what we can generate, or, you know, this is what you may or may not be interested in understanding of your site. Right. And customers will talk to us. They'll tell us, hey, we. We want to stop crime, we want to stop theft, we want to create safe environments for staff. You know, we want to stop loitering. They kind of communicate that to us. And then we tailor the service to do everything. Like, we'll do everything in our work, but we will only generate content for them to view based on what they want to see.
- [06:45]
A
Got it, Got it. And so then how does that work? If we want to, like, look at the brain of Blackout and. And open up the hood, what will we find underneath the hood in terms of how all the magic happens?
- [06:59]
C
Oh, boy. This is. Yeah, 14 years of work on this. So, I mean, there's a saying to the man with a hammer, everything looks like a nail. So I always try to get into, like, the. The minutiae of it. All of our mathematics are tailored pretty much around derivatives and integrals. So we like to scale things down, scale things up. So our AI is based a lot on computer vision and using that kind of math for our AI systems. Blackout is like one of a multitude of different technologies we've created that we've patented actually within Garda World. Blackout's role is to basically take the camera view because the camera sees like we see. Right. It's infinite in its depth.
- [07:43]
A
Although I have bad eyesight, so I don't know if mine is infinite.
- [07:47]
C
Yeah, fair enough, fair enough.
- [07:49]
A
Gets me sometimes. So.
- [07:52]
C
So we need to be cautious that we're not kind of overshooting our customer locations. Right. So what Blackout does is it uses AI to kind of determine what areas we're interested in looking for for the specific customer. And then it gets the AI to only focus on those areas. And then we have other technologies like Integra, which is tip of the spear. We, we just applied for patents on this about a little over a month ago, actually. And Integra basically was trained on about six and a half million images on how to upscale in real time the, the actual resolution of images that we gain from our customer sites. So now what's happening is, right, we have customers that have like older camera systems that, you know, they, they don't, they can't afford to spend, you know, a lot of money to upgrade their system. And so we can actually upgrade their system using Integra and then apply AI for object detection. So we're we're just, you know, seeing a lot of ways to use it. Right now. There's, there's about six primary methods that we use. AI.
- [08:52]
A
Awesome. Awesome. Now, I mean, I'm sure a lot of customers come to you. You're the cto, so a lot of customers come to you. A lot of your employees and team members come to you with these out of world stories about how people are using your technologies and they're safer or it avoiding something. Tell us one of those stories that, that gave you chills so that we can have chills too over. Because of your technology, we are in a safer world.
- [09:21]
C
Oh for like. Yeah, so we're a security monitoring company. Right. So we, we've seen it all. I, I've honestly seen it all. The ones that I think I appreciate the most are when we, when we deter loss of life, when we stop loss of life on one of our customer locations, whether it be an attempted assault or an attempted murder, which we have intervened with in the past, or you know, whether it's just stopping an altercation, let's say, that could have ended up where someone really gets hurt. I think those are the ones that are the most meaningful to me. That's where you start to, how do I put this? Like, we're in this obviously for profit. We're a business. We're looking to make money and provide our service. But when you do that, it really, really brings the humanity out in what we're achieving.
- [10:12]
A
Yeah, yeah.
- [10:13]
C
And it really gets celebrated the most within our organization, I think, is when we, when we protect people, that's what we celebrate the most. Yeah.
- [10:21]
A
Yes. I love that. I love that. And you are, as you, as you said, like there's this computer vision, intelligence how does the company think about Ethics and how you preserve, you know, like ethical observation and, you know, videotaping or what have you and make sure that the human stays at the center of it all? How do you think about that?
- [10:46]
C
Yeah, so that's a great question. Ethics is paramount to us. It really is. And I know any organization would say that in the world of AI right now, because it would be very taboo not to. But the reality is. Right, but the reality is it really is with us. So for instance, we actually anonymize all data that we ascertain from our customer locations. In other words, we never track or even look at who the individual is. We're not interested in identifying who the individual is. What we want to know is that there is an individual at the site. And so the anonymization of all this data means that we're not singling people out and we're not causing any undue hardship to an individual beyond what they've done to themselves in the court of law. Right. So if you get apprehended because you're stealing, that's something that's carried out by the police through apprehension and so on. But we are not ostracizing people by identifying them for the general things that people do. It's completely anonymous data, and it always will be. Facial recognition will never be something that now we deal with in our organization for those reasons.
- [11:57]
A
Okay, and I was just going to ask that question. So like, if there was a murder, but you're saying then yes, they can actually get a. I guess a subpoena for the video, but you're then removed, is then goes to the police or whichever officials. Okay, got it.
- [12:14]
C
That is precisely. Yeah, yep.
- [12:16]
A
Okay, got it, got it, got it. Now let's think about the future. Where do you see Blackout and your other technologies in terms of shaping the world? I used to say three years from now, but my gosh, it could be tomorrow. Like what will Blackout and all of your different companies and products be known for in terms of world impact?
- [12:37]
C
Oof. Geez. Okay, so we do think about this. We kind of planning things out as we go. So artificial intelligence, computer vision, specifically, the way we use AI is very good at identifying objects. And the hardest part of what we do is to identify the object. Once we have the AI trained to do so, then what we could do is start looking at what patterns objects commit to within the field of view of the camera. So 2030, let's say. Right. So we're five years out. Our subset of objects that we can detect very accurately is going to increase. And then what we're going to start doing is being able to really get into the nuance of the patterns that we see within these locations. And really, patterns just lead to business analytics. So our customers now can get a whole new wealth of knowledge of what happens on their site. Because what it'll start to feel like, Tamara, is imagine if a person like you or me sat 247 watching one camera, and you had a different person for every single camera that we were monitoring, which right now is about 145,000 cameras. So you got to. You can imagine.
- [13:50]
A
Right, but that's not employment, but probably boring employment, but yeah, employment.
- [13:54]
C
Yeah, that would be the most boring employment I could possibly imagine. But. But it'll feel like that to the customer because we're going to get that level of intelligence from AI on every camera that we're looking at.
- [14:07]
A
Oh, wow, that's amazing. That's amazing. And you said five years from now, but it sounds like you're already here.
- [14:12]
C
Yeah, we're making, like huge leaps right now with what we're able to provide our customers with, honestly.
- [14:18]
A
Yeah, that's amazing. Now just tell me, who are your customers? Because we have a. A breadth of listeners that, of course, are in corporate America. They have their own businesses, of course, they have their own homes and places that they live. Who are your ideal customers? Is it B2C? B2G? B2B? Tell us so that, you know, people can apply what they're hearing here to their everyday world.
- [14:42]
C
Sure, yeah, yeah. So it's primarily B2B. I'd say, like the vast majority, 99.9% is other businesses. Basically any business that has property or assets that they want to protect. We do everything from nuclear facilities to car dealerships to multifamily dwelling. We cover the whole gamut, really. A to Z coverage in the commercial space.
- [15:06]
A
Yeah, got it. Okay. So for those of you who are listening, if you have a business, if you work at a business and you, um, want to try them out and try something different, install something. If you don't, then you need to reach out. Which leads me to my next make us believers question. You know, we have very curious customers. Our listeners, rather guests. Oh, editors. Cut, cut, cut, cut, cut me do that again. Okay, well, Alex, that gets me to my next question, which is around make us believers. So our listeners are very much like, you can tell me that, but I want to experience it for myself. So what can. If, you know, if I work for a car dealership, or if I work for a franchise, what can. Or what would you recommend that they do this week to kind of try out your products?
- [15:56]
C
Yeah, so, I mean, like, we, we demo our products on a regular basis for, for customers all over the nation, all over the US And Canada. So, I mean, giving us a call, having someone come out and being able to demo the product, show you what it is, how it works, and then you can see for yourself. But, you know, seeing is believing in this case. And like I said, our customers are never really prepared for just how much content, like what happens on their sites that they don't know is happening because they're not seeing it.
- [16:22]
A
Right.
- [16:23]
C
So that, that's where the magic is for sure.
- [16:26]
A
Right. And should they go to your website and if so, what's that? Should they call some number?
- [16:32]
C
For sure? Yeah. My apologies. Yeah, so it's ecamm.com, ecam.com, super simple. From there you can get the contact phone numbers for your region, or you could just send us an email and we'll be reaching out and take it from there.
- [16:46]
A
Awesome, awesome, awesome, awesome. Now, I also have this question that I always ask, and I call it from One Genius to Another. And that's where my last guest wants you to share a perspective to a question that they have. So that question is, how many years from now will it take before an AI movie creating fully from prompts? How long will it take for that to win an Academy Award?
- [17:14]
C
Yeah, that was. I. That's. That's a really good question. Okay, I'll answer it in two parts, if you don't mind. Okay, so, yeah, well, we'll, we'll, we'll. I'll answer it. Like, if it was possible, how long would it take? I think honestly, within, I would say two, three years, you'll be able to prompt an entire movie and get something very compelling and hard to distinguish from reality. I do believe that that's possible. But the second part, the reason why I wanted a second crack at this is because I think the Academy Awards would never allow it. I think the arts will want to protect that creativity of, of mankind. And so, I don't know. I mean, maybe the Oscars will have like an AI category.
- [18:06]
A
Okay. You know, ethical way to do it. Right. Like, just have the categories can be revealed that that's how they created it.
- [18:14]
C
Yeah. And some guys come out and they're like, yeah, we prompt engineered this. And, you know, they get their Oscars and, like, that'd be cool. I think I, I'd be interested in that. But they will always have that division. But I really. The core of this question is how long will it take for AI to be able to be able to be capable of doing that? And I'm really saying two to three years. We're already, like, some of the videos that people are seeing. It's indistinguishable.
- [18:36]
A
It's crazy. Absolutely crazy. You know, and there have been some things out where I feel like the excuse going forward when stuff. When stuff comes out, they're gonna be like, oh, that's AI generated. That's not me. That was AI generated. So. So let's go to our bonus rapid fire. So I'm gonna quickly ask you four questions, and you're gonna quickly respond, okay, sure. You up for it? All right, good. Most overrated tech trend.
- [19:01]
C
For overrated. Most overrated tech trend. Oh, this is tough. Tamir. I could spend a while thinking about this one. Hmm. Oh, the cheese. I wouldn't know where to begin. I hate smartwatches. I think that that's something that's overrated. Yeah, I like a bit of traditional in. In timekeeping, so that. That's like just a personal opinion, but. But for actually, like, overrated, like, let's. Let's talk, like, now globally or maybe across North America. Technology that's overrated. I'm not. I think Bitcoin is. Is one of the things that I think was. Was fairly overrated.
- [19:44]
A
Okay.
- [19:45]
C
I know it has its, you know, merit right now from an investment perspective, and it's holding value, but I. I don't think it's doing anything particularly beneficial for society and what it's providing. Again, a personal opinion, but.
- [19:59]
A
Yeah, and that's what we want. The listeners want to hear your. Alex, your personal opinion. All right. Most under hyped AI trend or breakthrough under hyped.
- [20:10]
C
Some of the mathematics that are coming out for artificial intelligence, specifically in their use with quantum computing. It's not quite as mainstream for general public as of yet. Obviously, like, me and my circle, we talk about it and we read about it and, you know, read the research papers. I think that the implications that come out of that, when, you know, the question you asked me from one genius to another, where you said, how long before this happens? Well, so the reason I say two to three years, and even that would be a stretch to me, is because when quantum computing becomes viable and you apply artificial intelligence modeling and mathematics to that level of computational power, I mean, it's really crazy to think about what the outcome of that could be, what the output of that will be. And so currently underrated. I don't think it'll be underrated very long. It's going to become more mainstream. People are going to start to understand that it's a big deal. Yeah.
- [21:16]
A
Yeah. And I. I can't wait for that to happen because you're right. Like, in our circles, we talk about it all the time before, just like, you know, that person that isn't exposed to it all the time, because it does impact their lives. And that's one reason why I do have this podcast, so that people can hear about these topics, hear about these products, and see it in their everyday, you know, going in and coming out, etc. So. Okay, got it. Quantum. One book everyone should read on the future or on AI.
- [21:47]
C
Oh, geez, what would I go with on this one? So, like, I mean, I've read some that are just like, you know, that'll bore you to death because they're just more on the engineering side. But. But one that was good just to understand, like, how. How this kind of all came about to be and like, what it means and like, the philosophy behind it. Like, you asked a question on ethics, which I think is fantastic. We really should be putting as much thought into the ethics and philosophy of AI as we do into the actual engineering side. So there's a book called Nexus A Brief History of Information Networks from the Stone Age all the Way to AI. Yeah. And that book, it just for, I don't want to say layman's, because, like, that doesn't counter rate term, but for. For the average person that just wants to gain a bit more understanding, that book was good because you start to understand that AI is an extension of logic that we've already been using computationally for the last 67 years. Right. We're just getting better at doing it. And that's what that is about.
- [22:52]
A
Gosh, that sounds fascinating. I tell you, I get a book recommendation every time I have a guest, and my book catalog is becoming like this big, so I got to figure something out, maybe get the Cliff Notes of all these books. But. But Nexus, I'm going to definitely write that down. What is your boldest, biggest, most extreme AI prediction for the future?
- [23:14]
C
Oh, geez. And that isn't Fire and Brimstone?
- [23:23]
A
Well, I mean, if we need to be afraid, we need to be afraid. Whatever it is, we want you to tell us in your authentic voice what you truly believe the biggest, boldest prediction is.
- [23:34]
C
I mean, we are going to have to certainly put a lot of restrictions around the Use of AI because it just has such profound implications for the future of humanity. And so there's always talk about like Murphy's Law, right? Murphy's Law is anything that can go wrong will go wrong if given enough time. And that last piece is so important that if given enough time, because AI, you know, at some point in the future when it has a singularity and actually becomes, you know, its own entity, if that ever happens. You know, some believe it's not possible, some believe it is, but if it were to, time would not be a factor to that entity. The same as time is for us. We as humans tend to think about things in lifespans, you know, very unoften. Do we think beyond our own lifespan into the future? Few people do, but not the majority. And AI is going to think about forever because it's not going to really have those same restrictions. And the unfortunate part of that is it could lead to not so great things for humanity if we don't ensure that we have the right checks and balances in play for how we use AI.
- [24:47]
A
Now, is there a place in, where you, you would prefer AI not be involved? Is it like, you know what, well, you did say your watch, you did kind of say, you know, you prefer that, but is there something else where you're like, you know what, let's just keep AI away from that.
- [25:03]
C
Oh, geez. I like, I'm not saying it is in a big way in this category yet, but I absolutely don't like the idea of AI being used to augment managerial or decision making processes for humanity as a whole, whether that be in a corporate environment, a political environment. I think that the morality and ethics of, of human nature are, are, are fairly sound within reason. And I don't, I don't know if I necessarily like the idea of a machine kind of making those kinds of decisions on our behalf.
- [25:40]
A
Right. Yeah. Because you always want to make sure that there's some critical thinking that's going into it and particularly from, from as a human in my development and skills, you know, or my children's development, is because I want them to be able to critically think through issues and not just run the chat GPT or some other LLM or some other AI tool to like come up with the solution. So yeah, I, I definitely, definitely get that and I can tell you, well, one place that I definitely wanted to be in and is definitely in security systems. So I wanted to thank you and the entire ECAMM team for keeping us all safe and to also say that I've Thoroughly enjoyed this conversation. So where do we go to stay in touch with you? We go to ecamm.ecam.com to learn about the product. But how do we find you and eCamm? Social media and your product? Social media, yeah.
- [26:36]
C
So for me, you can send me an email@alex vracudioskam.com you could look up my LinkedIn. Contact me there as well. I'm very much into the AI kind of sphere here in Canada and the U.S. so, I mean, there's lots of conferences and stuff that I like to attend and would love to meet people to talk about this kind of stuff, both the philosophy and engineering side of things.
- [26:59]
A
I love that. The left brain and the right brain. Well, Alex, I so enjoyed this conversation. Everyone, you heard it here. First, stalk him at all his conferences and afterwards walk up to him and say, hey, I heard you're on Lead with AI podcast, so just follow him there. Thank you so much for talking about Blackout and all of eCamm's AI driven technologies. And again, thank you for keeping us the world safe.
- [27:25]
C
Yeah, thank you, Tamara. This is fantastic that you do this and it was a pleasure joining you here today.
- [27:30]
A
The pleasure was mine. All right, everyone, until next time, Lead with AI. See you soon. Bye. Bye.
- [27:37]
B
Thanks for tuning in to Lead with AI. I'll see you next time as we continue exploring the cutting edge innovations shaping AI across the public and private sectors. Until then, keep Leading with AI.