“As this, the flywheel continues to evolve in digital transformation for various segments and industries, we’re seeing an evolution also of concerns. Concerns about our individuality and our personal information and our personal data and things like this. So, hard to solve.”

Intro:
Welcome to the Agile Digital Transformation Podcast, where we explore different aspects of digital transformation and digital experience with your host, Tim Butara, content and community manager at Agiledrop.

Tim Butara: Hello everyone, thanks for tuning in. Our guest today is Peter Evans, CEO of Xtract One Technologies. They’re a leading technology-driven threat detection and security solution which prioritizes the patron access experience through the usage of artificial intelligence. 

In today’s episode, we’ll be exploring how modern-day problems require modern-day solutions, obviously, with a special focus on physical security and other practical applications of artificial intelligence in the physical space. Peter, thanks so much for joining us today, welcome to the show. Want to add anything here before we jump to the questions?

Peter Evans: Well, thank you very much for having me today. This is a very topical subject. We hear about weapons, threats, and concerns, and issues in the news every single day. This is not a problem that’s going away. So, I welcome the opportunity to talk to you and your listeners about how we can use digital transformation technology to address one of society’s biggest problems.

Tim Butara: Awesome. Yeah, as you say, it’s definitely an important topic. But before we dive into the specifics of threat detection and everything that you introduced, I want to take a step back and hit you with a kind of a chicken-or-the-egg type of question. And I’m wondering, what’s really the relationship between problems and solutions in the modern era? 

And what I mean is, for example, a certain groundbreaking solution to something may cause a totally new set of problems in a totally different area, and these areas will then require their own solutions, which may be then problematic somewhere else – what are your thoughts here?

Peter Evans: Boy, that’s a very complex question to unpack, there’s a lot behind that. I think that as a society we’re much more informed than we might’ve been in the past. The internet has given us access instantaneously and immediately to soundbites of information that allow people to somewhat feel that they are informed individuals, when all they’re reacting to is 140 characters in a tweet.

And so, what we find, for example, is that technology does provide an answer to many problems. Technology solutions scale – versus, say, for example, a labor-based business model, which does not scale very well. Software and technology can infinitely scale and can address much more pieces of information, as an example, and correlate that information for an outcome, than the human could.

The flipside of that, though, is that because we’ve got informed humans, we’ve got informed opinions. And those opinions tend to be more and more polarized these days. Let me give a simple example – facial recognition. It’s hard for a police officer who’s manning a venue, a large stadium, for example, to know everything about every individual that’s coming in.

But digital information allows us to understand, for example, you, Tim, everything about you. Google knows more about your shopping habits than your wife probably does. And so the digital insights are available, can be presented. We can look at things like what your characteristics are, your typical patterns of buying, your typical behaviors. We can look at things like your face and I recognize you and identify you as a patron who’s been to the venue before, or a patron who’s been banned at a venue before.

But now you raise concerns and new problems. What about access to personal and private information? How is that information being used? Is it being used correctly and with good intent, or is it being used maliciously? We’ve all experienced overzealous marketers who’ve taken advantage of our data and start spamming us too much. That’s just a small problem, that’s an annoyance, versus something that could present a risk to a business.

As this, the flywheel continues to evolve in digital transformation for various segments and industries, we’re seeing an evolution also of concerns. Concerns about our individuality and our personal information and our personal data and things like this. So, hard to solve. 

When it comes to a chicken-and-egg problem, what I also find is, as humans we tend to bias ourselves towards what we’ve done in the past. There’s an old saying, nobody got fired for doing what they did before. And when we look at physical security, we see this to be very true.

More often than not, the people are worried about, for example, in the United States weapons in schools. And what happens? People say, let’s hire more guards. But the guards are, again, a labor-based business model that doesn’t scale. A guard can’t be everywhere at all times getting insights, such as video cameras could be. And so, we start to see the same sort of approach to solving new societal problems with old historic methods that no longer apply.

I’ll use the same example, one last example here, Tim, with a walkthrough metal detector. That was a technology that was built 50, 60, 70 years ago for the penitentiary system, for the prisoners. To make sure prisoners weren’t carrying little pieces of metal that they could open up the handcuffs with.

Here we are in a society that has a large number of weapons, a large amount of societal unrest, a large number of people with homelessness and medical issues and these sorts of things. And we’re now trying to protect ourselves with a technology that wasn’t built for today’s expectations. So we need an organization or company, a society, a business, to change our mindset. And maybe not think about a chicken or the egg, but think about instead making a great vegan salad or something.

Tim Butara: That was a great intro. And yeah, basically, changing the mindset is the crucial thing to effectively digitally transforming, to effectively leveraging these new technologies, and to effectively leveraging the solutions to new problems without creating new problems. And also, another thing that, even if a physical or an old system is robust and tried and true, the longer it’s in existence, the more opportunities will be for people to beat it, to game it. 

I recently read this futuristic book – which wasn’t actually happening in the future, from our point of view it’s happening in the past, but from the point of view of the publication it was happening in the future. And one of the bad guys was carrying glass knives, which weren’t detected by any metal detector, any weapons detector, not even– they had all these advanced technologies, totally next-level stuff, but he outsmarted everyone because he had glass weapons.

Peter Evans: Yes. Oh, it’s true. And there’s actually an incident that occurred – I can’t name the venue because I think they would be embarrassed – but someone was stabbed with a ceramic knife. And the metal detectors don’t detect that, and unfortunately that individual died. And it was not a large knife, it was a two-and-a-half inch knife. But it was a ceramic knife and the metal detectors didn’t catch it. 

It’s also one of the challenges you’re seeing with some of the pro sports leagues. You know, the old metal detectors cause people to divest of everything in their pocket. And it’s fascinating, people might have pepper sprays or mace, little cans that are attached to the keychains; they put the keychains in the bin, and the organization that’s doing the security screening sees that and confiscates it. The plastic mace container would’ve gone through the metal detector just fine; the individual didn’t think about the fact that when they took their keys out and they put them in a bin, they were a challenge.

The real challenge to that is, to your point, well, the leagues are looking at this and saying, well, we detect a number of prohibited items because of the divesting process that occurs. People are unaware that they’re putting things in a bin that will get confiscated. But are you really stopping the person with malicious intent? 

Because the person with malicious intent, like your conversation, your comment, is going to bring the glass knife in. They’re going to find a way to bring in plastic weapons and things like this. Which then says, ok, what’s the maturity of the technology and where is the technology going in the future so it actually can detect these things?

Tim Butara: So now we’re coming to the meat of our discussion. So, how can technology solutions, modern-day stuff, modern-day solutions basically – how can these prevent incidents such as somebody dying after getting stabbed with a ceramic knife?

Peter Evans: Yeah, that’s a great question. First off, we need to understand, if we look around us, every aspect of our lives has changed. The fact that you and I are doing this podcast together and we’re in different parts of the world, that’s been enabled by digital technologies – the power of the internet, the power of fiber optics, the power of high speed access, all sorts of applications that ride on top of this.

Fundamentally, digital technologies, whether we’re talking about things like high speed access and fiber optics, or AI and ML, and other tools – they’ve changed the way that we shop. How many people go to a physical brick and mortar store versus shopping online these days? They’ve changed the way we bag. They’ve changed the way we educate. My daughter did two years during Covid entirely virtually of her education – and she actually did much better, her marks were better.

Digital transformation has changed every aspect of our lives. So it’s fascinating to me that when we look at things like physical security, it has not been adopted in the same way and the same manner. And again, you think about the promise of digital transformation – infinite knowledge and infinite data that can be correlated together to provide insights.

The events of 9/11, 22 odd years ago in New York City; there was enough data out there – and people found this by going through forensic analysis after the fact – there was enough data out there to absolutely identify that there was a threat that was about to occur. And they could’ve prevented and preempted that with correlation technologies. Those correlation technologies exist today. And you can use these insights to fundamentally change the way that we do physical security.

I know that when you go to an event and you buy a ticket, there’s enough information out there to say, we know who you are. Are you using a fake credit card? Are you exhibiting characteristics of a human trafficker? Are you using a false identification? We know all about you before you come to a venue. 

And the insights can be used to identify you at the venue, to track you at the venue, to watch when you walk into the venue, using technologies like ours, to see if you’ve got some sort of a weapon, if you’ve got appropriate credentials, these sorts of things. These are data insights that can be used to get preemptive in manner versus reactive in manner with physical security. 

And that’s the shift, that’s the change. If we want to solve the threat issue, you have to become preemptive, and that means you need access to reams of data that are correlated in an automated way.

Tim Butara: I love the point about preemptive versus reactive. This should be the modus operandi for pretty much everything, right. It’s better to prevent a disease rather than to cure it, or to cure its symptoms.

But also, you mentioned that it’s fascinating and it’s maybe surprising that, even though we’ve been going through digital transformation and we’ve been embracing digital technologies, and it’s been changing every aspect of our lives, these changes haven’t been so prominent in the physical.

And I’m betting that this is because the digital space is something completely new, and it needs new– it hasn’t had best practices and frameworks and stuff like that established for it, whereas the physical world kind of has. We’ll not argue about their effectiveness and their success right now.

But it doesn’t come as naturally for us to just assume that something that appears to be working should be changed now with the advent of digital technology, rather than something that is totally new, obviously needs something new, and it’s in the digital. So I think that this is a major factor here also.

Peter Evans: Yeah, it definitely is. One of the things that people need to realize– there are all sorts of reasons that digital has not been adopted in the physical security space at the pace or at the expectation that we want. Are the products mature enough? Are the assorted products integrated in a manner where I’ve got a nice dashboard?

Let’s use a venue like a sports stadium. I have insights upon approach, insights upon entry, I have insights while the patron is inside the facility. Are all those parts integrated together into a complete holistic solution to address an end-to-end security need? It’s still a lot of individual components that haven’t yet been fully integrated yet.

I think there’s also a little bit of, kind of strange, coming back to your chicken and egg. A lot of folks look at this and say, well, if I bring in all these digital technologies, am I going to be displacing my security guards? Is the resistance by the security guards to say, I’m going to be out of work?

Well, in fact, there’s actually an interesting idea here that– we’re a big believer in “human in the loop”. Digital technology is very good at looking at hundreds, thousands, millions of data bits of information, and correlating that down to two or three insights. Humans are very good at looking at three or four bits of data, and coming to a conclusion very quickly. 

How quickly do we size up an individual when we see them on the street, just based on two or three pieces of information? We may be wrong, but humans are actually very good at that. So the combination of digital insights to create the digitally enabled security guard who has advanced knowledge and advanced information, it actually enhances and turns the security guard into a profession, as opposed to a job.

Tim Butara: Well, but also, we talked about video cameras in the beginning. What good would it do if you implemented video cameras, and then you wouldn’t have a security guard watching the video cameras, right? I mean, we’ve had it happen before. It’s like, oh yeah, this new thing was introduced which maybe people fear that it’s going to displace them because it’s doing the job for them. But, no, now you’re just able to more efficiently do your job, basically. And this leads perfectly into the next question, right, about AI.

Peter Evans: Let me touch on that, cause it does lead into the next question. So, the challenge of video camera systems is, a lot of cameras were deployed to get insights and a perspective on what was happening around the outside of a building and the hallways, these sorts of things. But the challenge then becomes, the average human will tune out after about 27 minutes, lots and lots of studies show this, if they’re sitting there looking at a wall of video screens with all these video cameras all day long. It’s not a very enjoyable job.

So, AI can actually parse through that, coming back to the analogy, an AI can parse through all those video screens all day long and then just highlight the one or two anomalies. This is new, this is unique, we haven’t seen this before. Why are these five people standing by this back door when five people have never stood beside that back door previously?

So, you eliminate the need for monotonous work staring at cameras. At the ISC trade show in Las Vegas last week, there must have been at least a good 30 or 40 companies specializing in AI and video cameras, with that intent and that purpose, to augment the security solution with good AI.

And it’s there, that’s what AI is good for. You know, we named our company Xtract One because we’re in the business of extracting out of billions and billions of data points, or millions, or thousands, or whatever the number is, the one critical piece of information that you need in order to take a preemptive step in protecting your business.

AI on cameras does that very very well these days; whether you’re looking for a lost child, whether you’re looking for a particular person who’s a bad actor who’s been banned from a stadium before, whether you’re looking for a weapon on a person’s body as they’re walking through a system, and not alerting on their cell phone and their watch and their keys. This is the value of AI, to provide that kind of insight and scalability.

Tim Butara: That was an excellent point. And before we talk about other uses of AI I just want to mention a caveat here, because I think we can’t really move forward without this – kind of the downside of this, and you probably already know what I’m getting at – it’s how the data collection is done and how the AI is programmed.

Because we’ve seen examples of, especially when it was the case for somebody maybe who was not caucasian and was misidentified through facial recognition and then maybe got a sentence that they didn’t deserve, they got an unfair sentence. So this is also one aspect of it that we probably shouldn’t ignore, and we should probably invest more in developing AI more responsibly, developing these technologies with all of this in mind to minimize the risks of something like that happening.

Peter Evans: Yeah, exactly, you’re exactly right. AI, like anything, can have biases depending on how it’s trained. Whether I’m training my three year old child over there or training an AI engine, they become biased to what they learn. And originally, to train AI engines, people would look for a lot of data.

And you can imagine a scenario where somebody’s saying, we want to create an AI engine on video cameras where we can detect a fight. So how do you get a lot of videos for fights? Well, you might go look at Hollywood movies, and say, well, there’s lots of fights in Hollywood movies, let’s go run those through and identify those.

But Hollywood movies may have a bias that a person of a certain age, a certain gender, a certain ethnicity is the person who’s always in the fight. Now your AI systems are biased. So you have to be very careful. So, as developers, you have to make sure you’re creating objectivity in these systems; and, more importantly, making sure that, if you do use things like machine learning to train your engines, you’re putting those engines in a place where you’re getting a broad cross section of society.

Tim Butara: Ok, glad that we got this out of the way. So, if you do develop AI properly and responsibly, what other practical benefits can it have in the physical space, besides what we already mentioned before?

Peter Evans: Yeah, well, the big thing for me is shifting from a reactive to a preemptive mode in security. We mentioned that earlier, but that applies in many ways. It can also apply not only in the security side but to the patron experience side. Oftentimes what we see is, let’s call it the patron experience, whether it’s a venue, whether it’s a bank, whether it’s a shopping mall, whether it’s a school, a person’s experience oftentimes is in conflict with security.

We see it every day when we think about, my bank just went– and Salesforce just did this, for example. I logged into Salesforce the other day and now I need an app on my phone to validate and approve. It’s one more step in a multifactor authentication, a process. And every time we add another step in the multifactor authentication, it becomes an irritant. My experience goes down.

And so, security, for example, and experience have always been in conflict with each other. I can’t have a great experience if I’m adding in more and more and more layers of checks and checks and checks. Digital transformation offers the promise of actually making these two things work in harmony.

So while I can use digital insights to also get preemptive through the knowledge of who you are when you come to the venue, I can use that same knowledge to enhance your experience. So, for example, you go to the venue, I might be able to recognize you and say, you’ve been here five times before. Every time you’ve been here you’ve always sat at the 100th level; we’re going to send you to the suites this time and thank you for your repeat business.

So I can actually look at this not only in terms of making security better and less of a hassle, but I also can use it in terms of improving my security operations; improving my concessions; improving my food and beverage sales; answering questions like, why is one lane twenty minutes to go buy a beer and another lane’s two minutes? And what do the digital insights tell me about that in order to change the flow of my business patterns or a particular event?

I might find that on a Tuesday night, I’ve got one set of experiences and one set of lines and sales and commissioning and sales of food and beverage – another night it’s completely different. Having all those digital insights allows me to continuously finetune my business operations to make that patron experience better and drive more revenue. So, for the first time we might actually find that security is not just a cost to a business, but the insights provided can actually be a revenue generator.

Tim Butara: Yeah, it’s interesting that now, we’re enabled to have both experience and security. Because what you alluded to earlier is basically, friction kills experience. I guess that even people who really don’t like friction and don’t like multifactor authentication, probably after one security breach they would be much more favored towards it. But, yeah, this has been a great discussion, Peter– sorry, you wanted to say something.

Peter Evans: No, I 100% agree with you. There’s an old saying, security is always too much until it’s never enough. And it’s fascinating to me how often people are saying that, we can’t do this, we’re fine, we didn’t have a hack last year; but all of a sudden, when they do have a hack, they overspend. In this case, there is a benefit in overspending because the digital insight will actually make the entire business better, not just be looked at as a cost for the business.

Tim Butara: Well, as I said, this has been an excellent conversation. And I just want to return to the beginning of our discussion and take a forward view of it. And we already kind of talked about this when we talked about the need to develop more responsible, more ethical, non-biased AI. 

But in the context of the chicken or the egg problem that we spoke about in the intro, I’m wondering what you think about, how should we think about and build these digital tools, digital solutions to minimize the risks of these tools creating new problems down the line, as we discussed?

Peter Evans: Well I think, like with any technology adoption that any company undertakes, I would always say: do your homework. Do your homework, do your homework, do your homework. Do pilots, talk to others, talk to people who’ve deployed, learn the lessons. There’s always surprises with any new technology, whether it’s a digital technology, a physical technology, cyber security protection, new revenue generating enhancements on your website – do your homework. Don’t believe marketing people. Because they’ve got a different mandate than you have. Do your homework; run pilots, run trials. 

But also think about not only what the solution can deliver in terms of value today, but does it create a platform for adding more value tomorrow? The iPhone, when it first came out, lacked a lot of features. It didn’t have a GPS in it, there was no idea of a Google Maps or MapQuest or anything like that. But the platform itself held a promise of what the future could look like.

So, there’s do your homework, but then, as commercial organizations, invest in new technology, cause it helps accelerate the flywheel to get to not only what the requirement is today to be solved, but also what the future holds too. The small innovation tech startup world needs investment in terms of purchase orders in order to be able to accelerate delivering the full promise of the value of digital technology.

Tim Butara: So, long term thinking is one of the priorities here?

Peter Evans: Absolutely. Short term and long term in balance. Make sure you’re not buying technology that’s just a throwaway and solves the problem for 18 months, look for something that’s got the promise of the future also. If I’m looking to buy a digital technology, for example, AI on a video camera to watch for weapons outside my premise, for example – what else can that platform do? Does it integrate to your existing systems? Does it integrate to your overall alarming system? Can you use that same digital insight and that same AI to solve other business problems, or is it just a single-threaded application?

Tim Butara: Well, I think we’ve just gone full circle. Thanks so much, Peter, this has been really great. A lot of really good and valuable insights here which I’m sure our listeners will get a ton of value from. Just before we jump off the call, if listeners would like to reach out, or maybe learn more about you or Xtract One, where would you send them to?

Peter Evans: Well, our website is always a great place to start, www.xtractone.com. And if anyone wants to get in touch with me personally or any of my members of that staff, there is a contact me form that can be filled out right on that website. There’s videos there, there’s content, there’s information, but if they’re looking for more, just fill out the contact me form, and generally we get back to people within about 24 hours.

Tim Butara: Awesome, we’ll make sure to include all relevant info in the show notes. And, Peter, thanks so much.

Peter Evans: Thank you so much, I really enjoyed it. Take care.

Tim Butara: You too. And to our listeners, that’s all for this episode. Have a great day, everyone, and stay safe.

Outro:
Thanks for tuning in. If you'd like to check out our other episodes, you can find all of them at agiledrop.com/podcast as well as on all the most popular podcasting platforms. Make sure to subscribe so you don't miss any new episodes and don't forget to share the podcast with your friends and colleagues.