LPRC - CrimeScience – The Weekly Review – Episode 208 Ft. David Thorp (Cobb County PD)
Episode Date: July 8, 2025In this episode of the CrimeScience Podcast, LPRC's Director of Research, Cory Lowe, sits down with David Thorpe from the Cobb County Police Department. The conversation explores real-world crime prev...ention strategies, the role of law enforcement partnerships, and the importance of data-driven approaches in reducing crime.
Transcript
Discussion (0)
Hi everyone and welcome to Crime Science. In this podcast, we explore the science of crime
and the practical application of this science for loss prevention and asset protection practitioners
as well as other professionals. Hello everyone, my name is Corey Lowe and I'm director of research
at the LPRC and I'm joined by David Thorpe who is with the Cobb County Police Department.
David, if you could just start off by giving us, tell us a little bit about yourself and your background.
Yeah, absolutely. Thanks for having me, Corey.
I've been with the police department for about 23 years now.
Prior to getting in law enforcement, I actually did work in the retail space.
Shortly after getting out of the Marine Corps
in the late 90s, I found myself needing a job
while I was going to college up in the Boston area.
So I wound up working for two separate retail,
clothing retailers, about a year and a half for one,
about two and a half
years for the other and then I find my way down to Georgia about four and a
half five years later and that's where I was introduced to policing and so I
started out as a patrol officer as most officers do and I really started to I guess have desire to do more with cases
than just write incident reports and hand the incident reports over to
detectives and so I began diving into on-scene investigations much more
deeply and fortunately for me back then we had access to some pretty cool technology.
It's hard to believe that about 18 years ago in 2007 our department was demoing some vehicle-borne
cameras that were connected to a license plate recognition system. And I was able to use that technology to help identify and locate stolen vehicles, as well
as doing some traditional investigations.
So I really started to have a passion for technology-led policing back then and had
some success when I was transferred over to a few detective units. I wound up working for a
property crimes detective unit at my local precinct after I had been with the department
for about four or five years and really just started to cut my teeth there, get a lot more
experience, and I was transferred up to the crimes against persons unit and I was assigned
to the robbery unit.
And I continued to use a lot of these technology
driven policing methods, facial recognition,
different LPR programs,
and had some significant success with it,
as well as being introduced to some GPS tracking technology,
which obviously comes in handy
when you're working on a robbery unit during the daytime, bank robberies and all. And so I want to spend a significant amount
of time in investigations and then spend some time on the road as a sergeant,
helping develop and train officers in the use of some of this technology.
I found myself back in an investigative unit,
supervising some grand detectives.
Then I got promoted to lieutenant,
went back to the road, and then in
about seven or eight months,
there was an option for me to come up to
the real-time crime center and we were just developing
an RTCC and looking to staff it.
Due to some extenuating circumstances,
they needed somebody with some depth of knowledge.
And I wound up getting transferred up
to the real-time crime center and started within four days
of being notified.
So that's a pretty exciting time.
And I look back now on the history with my department.
And little did I know it at the time, but there were very deliberate steps getting me to this position.
Just having a deep knowledge of different LPR systems and being able to have some success with them,
as well as all the other integrated technology we use.
So I was very fortunate for that transfer.
Four years later, I've been there this whole time.
Awesome.
Yeah, and we had the great fortune
to have you come speak at Impact,
and you talked about two case studies with GPS units,
and those case studies were just absolutely amazing.
It was a very, very well received presentation
at the conference.
Now you've been with the RTCC for four years now.
We're now partnered with the
National Real-Time Crime Center Association.
It's a partnership that's turning out to be fruitful,
even though it's still pretty early days.
turning out to be fruitful, even though it's still pretty early days. But how has your role with the RTCC evolved since you first joined them, you know, four years ago? How
have things begun to change and what do you think are some of the greatest changes happening
in RTCCs? Well, for starters, when we first began the RTCC,
we were solely focused on integrating a lot of the technology
that we already had.
Camera streams from municipal cameras,
shared cameras from some other departments within our region.
And really, little did I know back then
how large the RTCC world could get or was.
But I think in a word, I'll use the word exponential.
I think there's been exponential growth in public safety technology since I started four
years ago in the RTCC.
So initially, it was just bringing those camera streams, as I said, from our own
county-owned assets and then branching out. And really the impact of an RTCC is not just,
in my opinion, not just those camera streams that you own, but community partnerships,
leveraging those partnerships to get integrated camera streams,
and light up the map, if you will,
and increase your footprint of community partnerships throughout your area.
It's a different challenge for every agency.
We have about 770,000, 780,000 residents within our county.
We share a lot of law enforcement responsibility with six or seven different police agencies.
When you think about almost 800,000 residents over 345 square miles in our county,
that's really one of our largest challenges,
is getting access to streams,
building partnerships,
and then maintaining those partnerships because they do require maintenance, they do require
frequent communication.
Technology is great, but oftentimes things go down, devices get unplugged, properties
change hands. And so it really requires a lot of hands-on time.
And so fortunately we have somebody on staff,
you know, RTCC that focuses on that.
So nowadays it seems like there's a lot
of different technology that at some point
in the past was disconnected from the RTCC.
And now I think we're starting to see the importance
of seeing everything on one map
to allow us to respond to things much more quickly.
Some of the access points that we have to this data
are standalone, and they do not incorporate into the RTCC.
However, the more tabs or
the more Cloud-based services you
have that are open separately,
it makes it very difficult
to see everything in relation to one another.
The FBI, I think, states something like 70 or 75 percent of
crimes are committed with the direct or
indirect use of a vehicle.
And so access to our Department of Transportation traffic cameras is imperative for the success
of anything we do really in the RGCC, not just for us, but also to support and serve the community. Because cameras capture direction of travel
for vehicles, they capture all sorts of attributes about the vehicle, and then using that in
conjunction with our license plate recognition cameras, we can usually identify a vehicle
or a registered owner pretty quickly. So managing all those different cameras, managing all those different integrations is one of the
biggest challenges we have, but it's a good challenge to have.
Yes, it is. Just thinking back, right now at the LPRC,
we're really focused on building out our fusion platforms.
And what I'm referring to here is, we're calling them fusion platforms.
There's a lot of different ones,
but there's Axon, Fuses, Motorola Command Central,
Wear, Flock OS, even 3SI's direct dispatch
and Seagraph from Genitech.
Those are some fusion platforms.
And what they do for both retail and law enforcement
is they fuse together all these sensors
so that instead of looking at all these different. So that instead of looking all these different tabs
and windows,
it's all in a single pane of glass or as close
to a single pane of glass and you can get
so that you can not be distracted by going to
have to switch between tabs or windows.
But then it's also there also fused to platforms because
they can fuse together both private businesses and the public sector
law enforcement and other agencies.
It's it's very difficult on audio to do this, but it's it's
like you have this overview. So say you've got that vehicle
that's moving down the highway and you've got a GPS tracker in
that vehicle. At the same time, you can start getting LPR hits
and opening up the traffic camera feeds and all of that.
So it's pretty exciting times in my opinion.
Are you all using the fusion platforms at your RTCC, right?
We do, we've got a FUSIS by AXDON is our contracted vendor
and we have several different license plate recognition companies that that dump LPR data into their.
Yeah, excellent.
What are you seeing your common pain points today? What are your greatest challenges in RTCC work? But I think. I kind of touched on it just before just before
we started talking about the different fusion centers, but
one of the largest challenges is is getting integrations for a
smaller agency that may not be the challenge, but I will tell
you for our agency that is one of the challenges getting
integrations, keeping them. Another
challenge for us is we in law enforcement, we are here to serve, we're here to provide that service
to the community. And sometimes our budgets lack, you know, I talked about the advent of, or the invention of all these different technologies and how things
have been growing exponentially. Now, it's like a kid walking into a candy store. You see all
these things you want. You just can't afford them. And so we have to be diligent about how we use
technology. We've had to sunset some technology that we just don't see a big impact on or it's
not making an impact for our department and thus it's not making an impact for the community.
And so we've had to pivot a little bit. So, you know, as an agency, I can tell you law enforcement
is obviously a government entity and things can move very slowly.
So budgets don't automatically inflate quickly.
It's a struggle to try to get the technology that you want to help support the community.
And so in summation, I'll say that sometimes we're slow to react and slow to adapt and
adopt new technology because there's just so much out there right now. And ultimately, when one vendor or one technology partner sees success in the marketplace with
another vendor, oftentimes there is a large overlap.
And so now one vendor who used to produce ABC widgets, now sees another vendor producing, you know, XYZ widgets,
and they want to start dabbling in that. And so there's a lot of crossover, there's a lot of
overlap in what technologies are available, and not every partner produces something that we need across the spectrum.
And so having the ability to talk to different vendors,
talk to different technology partners,
get demos and trial periods for these products
is imperative for our success
and to ensure that we're actually procuring something
that we need and we're not simply just keeping up with the Jones next
door because what suits our agency and our department in our community may not necessarily
be a good choice for somebody else and ultimately vice versa.
So those are the three things that I would point to as those pain points.
Yeah, I can definitely see all of that.
I think in retail specifically,
since we work on the loss prevention
side of things,
I think one of the challenges that
I've seen between real time crime
centers on the retailers is kind
of the scope with which each works.
You know, you're with Cobb County and
so you've got to interface with all these different businesses, you know, you're with Cobb County and so you've got to interface with all these different businesses. You know
citizens, you know all these all these different
organizations, things like that.
On the other hand, you know, a retailer like Target or Walmart
might just have a few locations in your area and thousands of
locations elsewhere that they have oversight for. So Walmart might just have a few locations in your area and thousands of locations
elsewhere that they have oversight for.
Sometimes what I've seen is that there's a disconnect between
your folks who are doing
the community outreach and trying to get people integrated or
talking to people at the store level when
it's the very top of the organization that you really have to
be talking to to get the integrations to happen.
So can you speak a little bit more to that and how that creates challenges with working with some of the more technologically advanced retail companies that have some really cool stuff, but getting with the right people can be very difficult.
Absolutely. Sometimes the analogy I use is like a puzzle. Sometimes you find the right piece in that box and it just goes together easily.
Other times you could be searching for hours for that one specific piece of that jigsaw puzzle.
It takes a lot of effort in some cases to get that piece in the right place.
I will tell you that we have waited patiently for up to 18 months working with a retail
partner at a shopping plaza to incorporate with them.
But ultimately, not everybody is aware of what a real-time crime center does.
In the grand scheme of things, RTCCs are fairly new.
Some of the large agencies have been around for 10 or 12 years operating with a real-time crime center,
maybe even 20 years, but these smaller agencies
don't have the wherewithal or even the space for that RTCC.
And so community knowledge is just not there.
And so that is one of the challenges of trying to get,
excuse me, an integration is that at the ground level,
when you walk in the front door to a brick and mortar retailer,
whether it's a small local or regional retailer,
or even a large nationally known retailer,
you're working with them every single day to prevent shoplifting and organized
retail crime and to increase safety and security.
And they want to partner with the police department, right?
I mean, just like you said, it's a one-to-one conversation, and they couldn't think of anything
better to partner with the police department because they have a passion for the same things
that we do. Then at some point in the conversation, the concept of an RTCC and what it means travels up the chain on their side,
and it gets lost in translation or it dies on the vine.
Working with our, not just retail partners,
but working with these, as you said,
fusion center companies,
these RTCC companies,
their technology companies, to have larger conversations with corporate leaders, whether
it be legal or loss prevention or asset protection, is very important. So, you know, thinking of it as
kind of a one-two punch, we are operating at the ground level, and they are operating oftentimes at the corporate level.
And between both conversations, we've seen pretty good success. But obviously, it can be frustrating to see people in a retail environment wants a partner, and then somebody winds up shutting it down because they're, like I said, there a something that's lost in translation, or somebody conjures up the idea of surveillance.
And in most real time crime centers, I can tell you in ours, specifically, we're really incident
based. So when an incident happens, that's when we start accessing the cameras. But I think I
remember one conversation that we had on teams call with you a while back was the
question that you would pose to a retailer is, can we have access as a law enforcement
agency to all your cameras all the time, every day, all day?
And the overarching response was a one word answer, no.
If you want a two word answer, you're going to have to bleep it out.
But it, no. Right, if you want a two word answer, you're gonna have to bleep it out, but it was no.
And then fast forward to a hypothetical incident.
Well, it's what if a shooting happens
or what if there's a missing child
or what if you have violence at your store?
Would you want officers to immediately be able to access
that to develop information in real time?
And the answer is a polar opposite.
It's a 100% absolutely.
Well, where do we meet in the middle?
And so success stories certainly speak volumes in sharing those success stories.
And once one retailer has some success in that sharing of video, identification of a suspect, appreh identification of a suspect,
apprehension of a suspect,
they're gonna start seeing some of those lost numbers
go down, the shrink numbers go down,
and hopefully they're sharing that information.
And through just natural organic conversations,
and obviously I have to highlight LPRC,
I was very impressed with the conversations
that I had down there at the impact
conference and the integrated conference a few months ago. And a lot of conversations that we had
locally in RPD post-conference came directly from conversations we had at that conference
and some of the success stories that we shared. So those are the things that are the
challenges for us. But I think as more retailers come online, then it will be an obvious choice,
an obvious solution for these other retailers. So it's the, you know, I always use the term FOMA,
right? The fear of missing out.
When you see somebody else having success with it, you want to have that same success.
And so it's a national inclination that once word gets around, that we'll partner with
them.
Excellent.
Those partnerships are absolutely essential to success of real time crime centers.
At the LPRC we're always talking about intelligence building.
You can't do it with everything being in silos and
we were talking about technologies earlier.
Personnel and people and organizations having those that those connections
between organizations is just just absolutely essential.
So that's what we've been trying to do at the connections between organizations is just absolutely essential.
That's what we've been trying to do at the LBRC is bring together
the real-time crime centers and the retailers so that we can have
these higher level discussions to make some of this happen.
Sure.
I think that the future is bright for real-time crime centers.
I hope so.
I want to talk about-
I think so. real time crime centers. I want to talk about a few of the technologies that you are,
were you seeing the most success? What some of those are? And can you describe some of the
successes that you had with some of the technologies that you've been using?
Yes, absolutely. I will say really the bread and butter of our real-time crime centers is any integrated camera.
It doesn't just have to be an LPR camera, but it could be a camera from a DVR or an NVR and a retailer or other business.
LPR is obviously a big component. And then one other technology that we've had a lot of success with is GPS tracking technology.
As you mentioned earlier, in my presentation, I shared two success stories we've had with
two separate technology companies that do GPS.
And what's great about the GPS technology is, and I'm thinking about this not just from
a law enforcement angle, but also from that
time when I was in retail, a little short story is that I was a retail manager and we
had apprehended a shoplifter.
And that shoplifter was in the office with me.
We had escorted him into the office and he wound up pulling a knife on me.
And so that's not a place that anybody
wants to be. And so shoplifters and organized retail crime suspects can turn violent in an
incident once they know they've been caught. And the great thing is that in modern retail
establishments, nobody has to actually confront the suspect.
That GPS device, if it's stolen,
when it's stolen, either on one item or it's simply on an item,
and there are 100 items that are stolen in a large-scale smash and grab,
it'll ask for the apprehension to take place as far from the store as possible at a time and
location that's convenient and safe for not just law enforcement, for the suspect's apprehension,
and also for any bystanders. So the GPS tracking technology is really great. And again, we've had success with it. The two success stories I shared were with large retailers
where there were some women that had come in
and bear hugged, you know, a rack of clothes.
And they're always after one particular brand,
whatever's popular and whatever's, you know,
selling on the open market.
And so they came in, scooped up a bunch of property, and then fled.
And using the GPS tracking technology is great.
I remember working as a robbery detective tracking
die packs and stolen cash from banks 15, 20 years ago.
And we didn't have LPR back then.
We didn't have integrated cameras with our DOT.
So we had no idea what vehicle these guys were in.
So the integrated technology is a layer on top of the GPS technology.
So it allows us to build information and we can use those two technologies in concert
with one another.
And then you've got LPR, not just the DOT cameras, but you've got the LPR footage, you've
got a crystal clear image in most cases of the vehicle, the attributes of the vehicle,
and hopefully the license plate.
And so in both of these instances with these GPS tracking technology companies, we were
able to identify the vehicle.
We were quickly able to receive photographs from the retailer,
because this relationship, if you will,
Corey, has already been established through a partnership,
and so I'll call it a digital handshake.
The retailers want to see success,
the GPS tracking companies want to see success,
and we certainly in law enforcement want to see success.
And so that digital handshake is imperative
and it's already established.
And when something's stolen, everybody knows what to do.
So in a traditional law enforcement environment,
it could be days, if not weeks
before we
get images of the suspects because somebody's not in the store to provide them or the district
loss prevention manager is on vacation or away and they can't access it.
So for us to have a picture of the suspects fleeing the store with what they're wearing. It's almost like a immediate show up or a lineup. You've got information
that you now convey to an officer. So in both of these cases, we wound up sharing
information, working with license plate recognition software to identify intersections and cameras from the store to two or three jurisdictions away, 15-20
miles, we're able to corroborate information, get that information out to
officers and apprehend them and in most cases recover the merchandise. And then
that's just the tip of the iceberg though. We're able to not just identify them, but now all of those retailers
that these suspects have been targeting know who these people are.
Those retailers share the information through that intelligence sharing platform.
And all of a sudden, all the retail incidents across a region starts to light up, and you
start connecting the dots and understand that these women have been targeting several stores
in a metropolitan area over the past two or three months, and then those cases can go
from being unsolved to solved.
So that's just a high-level view of what some of those success stories are.
But if we have time, I'll tell you about another success story we had that didn't involve the
GPS tracking.
So let me know if you have any questions about those two scenarios I just told you.
No, I think those are amazing wins.
I'd love to hear about the next one here about sales volume.
Yes. amazing wins. I'd love to hear about the next one here about sales volume. Yes, so this happened about two years ago and really it's one of the things that retailers need
to do and law enforcement real-time primary centers need to do is establish that data pipeline.
Certain information that's conveyed to a law enforcement agency is important. A lot of times,
a retailer, whether it's a mom and pop retailer, a small regional retailer may not know that
the local police department has all these additional resources now in a real-time
primacenter, three or four people working together, not just officers out in the field
working together, but two or three people sitting next to each other
working in concert with one another.
And getting information out is important.
So in this particular case, it was a computer digital electronic store, and they were familiar
with this offender.
The offender was wanted from a previous, for a previous theft. And the manager employee got on the phone and relayed specific
information about that, who the person was, what he or she was wearing. And when they were confronted,
the suspect then left. And so they hopped into a vehicle and it was a brown or black Chevy Malibu. And so the first thing we did
at that point was get on some of our DOT cameras. We accessed them and I could see the vehicle
leaving the parking lot. And so I continued to access cameras down that surface street,
down that corridor. I think I accessed three cameras. Traditionally, an officer's
going to get dispatched to the store. But if you put a dot on the map, if you get on the radio and
say, hey, that vehicle is leaving, you're confirming what happened. You're giving real-time
information to that officer. And even though there's no GPS tracking technology in this case,
it was the correct, accurate accurate relevant information being relayed
from the retailer to law enforcement and then us doing something with that in real time.
And so over the course of about two minutes, we access four or five cameras. I updated
the officers to where the vehicle was going. And now I've got a dot on the map where this
officer is going to go. Like that old phrase that Wayne Grespie
has often quoted as saying, don't skate to where the puck is, skate to where the puck is going.
So previously the officer goes to the store and he's going to take a report and that's not exciting
and then he's going to turn the case over to a detective and that detective is going to be three
or four days behind because that's just the nature of the beast. But in this case, I was able to tell the officer where the vehicle was in real time within
a matter of milliseconds of it passing an intersection. And the officer got behind the
vehicle and put information out on the radio that he was there. And then he made a traffic stop.
And now we can safely apprehend that suspect about three miles from the store. The officer can wait for backup to detain him and you've identified the suspect, you've fulfilled
your commitment to the community, you've fulfilled your commitment, you've essentially closed the
loop and you have restored an incident and a retailer back to its original upright position, right?
I mean, you've completed the project.
And so relevant real-time information is important, and accurate information from the retailer
is important.
Yes.
So that's actually where I was wanting to go next.
In loss prevention and ORC investigations, the focus for the last couple of years has been on
increasing.
Increasing. The collection or improving the collection of
intelligence about organized retail offenders and groups.
And you know that's made those cases actionable, but for prosecution because all that's happening on the investigation
side, not on the real time response side.
When we're thinking at the at the LPRC, we're very interested
in, you know.
Yes, we want actionable intelligence, but the question
that at the next question is actionable by who right? We want to get the right intelligence to the right people at the store level. We want to get the right intelligence to the right people at law enforcement.
And then we also want to get the right intelligence to the right people in investigations. Right. So, you know, if we have that real time intelligence and, you know, there's all these factors to drive law enforcement response. You know, law enforcement to take a call
for service much more seriously and going to have a better
response if they know.
What's going on? Whether there's something's actually they come
from responding to the call right? No one wants to waste
their time and shut to something that one's no one's
enthusiastic about that, but if they're getting information directly
from a colleague at the law enforcement and law enforcement,
that's something they there's a little bit more credibility to
that. Right? When they say hey, skate to this this place,
because this is where the offenders going that's much more
actionable and probably going to be much more fruitful to them.
Officers are just as rational as anyone
else in those regards. So very good stuff there.
What other factors do you think drives law enforcement or do you
know drives law enforcement response? I mean, when you get
two calls, what are the things that are making you say or
making law enforcement
say I'm going to respond to this one versus this one?
Well, first and foremost, any call that comes into an emergency communication center, a
911 center, is going to be triaged.
And we have, as most agencies, if not all, have priority responses.
Something that's a threat to life,
something that's an emergency is going to be prioritized much
higher than an average alarm call.
And so think of four boxes, if you will.
And every time a call comes in, the dispatcher
has a choice to make about what the call is
and which box that's going to fit in. We're not talking about the type of call, whether it's a
theft or a burglary. We're talking about the priority. So they put it in one of those four
boxes, if you will, and based on the totality of circumstances from that now-owned call,
the totality of circumstances from that now-owned call. Has it just occurred?
Did it occur previously?
Oftentimes, retailers have policies and protocols where they're not to call for an hour.
That can really change the response of an officer and just an entire shift.
If the crime is in progress,
well, then obviously the suspect is on scene.
Think of it as solvability factors.
Our detectives operate in solvability factors and oftentimes,
by design, if there are not enough solvability factors in a case,
the case will not get assigned
because there's just so much legwork
that would have to be done.
And most of that work would be not fruitful.
And so is the suspect on scene?
Is there a good description of them?
Are they currently offending?
Have they previously offended? All those things again make a difference. And so what I love about the
real-time environment is that to me it doesn't matter what the
offender is doing. It could be a shooting or an armed robbery or it could be
somebody who just shoplifted some crest white strips
or a bottle of Windex.
To us, since we are essentially one step removed
from that environment,
all that's just the cameras no matter what.
And so we can see a lot of impact,
not just in the violent crime,
but we can see a lot of impact in smaller crime
because that still affects retailers.
It affects large retailers,
it affects mom and pop retailers.
And so for us, it really doesn't matter
what the priority of the call is.
We can still access the same data sets
and provide the same information if you've stolen $10
or if you've committed a heinous felony and injured a person and then fled. So for me, that's that's one of the exciting things is there's always work to be done.
We're just not sitting around waiting for the big show or waiting for some heinous criminal event to occur. We can we can do a lot with many of the calls for service.
Thank you very much. So so far we've covered, you know, history. So starting off with your background and some case studies and you know the types of things that you're doing today at the RTCC.
Now I want to talk about where things are going in the, you know,
about where things are going and the opportunities and partnerships and technologies that you're most excited about. So let's just start with technology. Is there anything out there that
you've seen that is just like, oh, this is going to just change the nature of real-time crime center
work if this really scales? Anything that comes to mind in terms of technologies
you're excited about?
So I will tell you about two things that I'm excited about.
One, and I know a lot of people talk about it,
and it's kind of painting with a broad brush,
but AI and artificial intelligence
and the machine learning aspect, that's
been in a lot of this technology for a number of years.
But what's great about it is it's becoming more accurate.
If my suspect in a call for service is driving a red pickup truck,
I want to be able to identify which cameras in a matter of
20 minutes have picked up a red pickup truck
that matches that description. And so in real-time environments, accessing camera streams,
you know, one after another can be very cumbersome, and it can be slow because you access it,
you have to make a determination
if the vehicle that you're looking for already passed that point, you can rewind it, and
then meanwhile you have to worry about accessing another camera stream.
So obviously having somebody else to assist you is imperative to the success of that incident. So AI and being able to identify the path that a vehicle took,
attributes of the vehicle,
and provide me with a bunch of different pictures
so I can identify, you know, that has to be,
to me, that has to be a human element involved.
It's not just AI saying, go stop this vehicle.
It's saying, hey, we come through X amount of video
and deductive reasoning and the algorithms tell us
that it's going to be one of these five vehicles.
And then the human element comes in.
And then what AI has just done is done searches
that would take a human,
probably 10, 15 minutes to do in a small scenario,
much longer in a large-scale scenario maybe.
Now you've got this buffet of options and you look at all the attributes
and there's something about the vehicle that's gonna stick out.
And so we also talk about AI in the realm of security
and safety and if you've got a pharmacy
that closes down at eight o'clock at night,
there should be nobody in there at one o'clock
in the morning and this AI can tell me
whether or not there's a person in the building.
can tell me whether or not there's a person in the building. And so that immediate report
to a human in a real-time crime center or law enforcement agency, that will generate access to camera feeds. And then we can corroborate whether or not somebody's actually in the business.
We talked about priorities before. That's the difference between an alarm and we'll find
out what happens when we get there and a response which is, hey, there's somebody actually in the
store. And then all the other things that are connected, LPR, DFT cameras, all those would be
populated with the same information, the height of the
person, whatever they're wearing.
And so that search capability is very important.
And so the other thing I'll say that I'm really excited about is just there are a lot of siloed
assets out there, not just that we don't have access to, but even within our own law
enforcement agency that we have to access. And so having a technology that brings all those together
allows us to make decisions much quicker. It puts that buffet of options in front of us in a matter
of seconds instead of this laborious long time, long-term searching by a human.
And so we can do a lot more with less,
much quicker in that environment.
And so I'm excited about the AI being laid on top
of some of these camera streams,
and then also tying all those different disparate
data sets together and being able to leverage
small information that was kind of tucked away and that we didn't know about and have
that be relevant in a real time scenario.
So I actually want to go back a few questions now.
You're talking about tying things together in real time and the
difference it makes to be able to see you know that
that person who's at a chain drug store at 1 AM
in the morning.
That's not supposed to happen and go back to those
response factors.
Well, what's going to drive a different response?
So say you've got two chain drug stores.
One you have sensors on the site that say someone's
at this location there inside the building. Maybe maybe you know
alarm has gone off and you've gotten notification of that
versus another one where you have you know not only do you
have the alarm, you know someone's in the building, but
you also interior cameras showing what that person is
doing on the inside. How is that going to if you, you know, you've got one person to respond
to one of these incidents, which one are you going to choose and why? Well, the answer is obviously
easy, right? So we would wind up responding to the incident that clearly shows there's an offender in
the store. The only person that potentially could be in the store at that or the pharmacy at that that late hour could be a cleaning crew. But with access, we know that they're there.
You know, I mean, they can come in, they can enter certain key code on the keypad and then that that bypasses, you know, any, any AI. But if somebody
But if somebody throws a big rock through the window, or the front glass door, that's going to generate response.
And so it will be a much more heightened response.
If we have units available, they would respond to the crime in progress because it's validated.
So there's an alarm company, I can't remember the name of it, but they've been advertising in the local area for years, and they always talk about their alarm systems, their key panels, or their alarm panels are actually have a,, you have an officer that responds.
But if you can say, hey, we have live video, the alarm system says we have live video of
a person that's actually in here taking silver off the mantle of the fireplace, then that's
going to provide more information.
Officers are going to get excited to respond.
And the priority, ultimately, of the call for service
is going to be much higher.
And so, again, the answer for your question
is we're going to want to respond to the one that
has the person in it.
And so, ultimately, we just need just a little bit more
information than we've had to make a difference.
And the reason I brought that example up
was because a couple of years ago, we were dealing with a group of gang members who were stealing Dodge products, and they were driving around in Challengers and Chargers and Durangos, and they were breaking into pharmacies at 1 o'clock and 2 o'clock in the morning. And we would get the alarm call and we would respond. And we inevitably would arrive to a pharmacy
that had a broken window and all the pill bottles were gone.
And it was always an uphill battle
to identify who these folks were.
And just a little shift, just a two millimeter shift
in the way we respond can really make a big impact
in identifying three or four people.
Now you've got a vehicle that's associated with them. And so we ultimately would share that
information with our law enforcement partners and in this case a retailer and that would have an impact.
Yes. So we've gone for a pretty good while now. I did want to mention one more thing that at the LPRC
we're increasingly thinking in terms of this connect,
detect, affect web of where we're trying to
connect all these sensors together and then being able to
know where all of our sensors are relative to each other.
And then finally, where all the personnel
and people who can actually respond to an incident are.
So not only in this case,
having all these technologies tied together
helps you to understand where the threats are
and things like that,
but you can also begin to understand
where the resources are relative to that threat.
So if you have that rich intelligence
about the crime in progress,
and you know that there's a unit nearby,
with all of that real time intelligence
that you have about not only the threat,
but also the resources available to address the threat,
you can really make those connections a lot quicker
with that real time intelligence, right?
Absolutely.
Yeah, so that's, I think that that's where I wanted
to wrap up today.
David, it's been an absolutely fantastic conversation today.
So thank you very much for joining us again.
We really appreciated having you at the Impact Conference
to speak to all of our members there.
And we're really looking forward to the work
we're gonna continue to do with the National
Real-Time Crime Center Association, of which you are an important part.
So thank you very much, David.
And thank you for joining us.
You're very welcome.
Thank you for having me.
I appreciate it.
Thanks for listening to the Crime Science Podcast, presented by the Loss Prevention
Research Council.
If you enjoyed today's episode, you
can find more crime science episodes and valuable
information at lpresearch.org.
The content provided in the crime science podcast
is for informational purposes only
and is not a substitute for legal, financial, or other
advice.
Views expressed by guests of the crime science podcast
are those of the authors and do not
reflect the opinions or positions of the Loss Prevention