October 20, 2025 • 22 mins
Episode: 00289 Released on October 20, 2025 Description: In this fifth installment of the Real Crime All The Time series on Analyst Talk with Jason Elder, Nikki North joins Jason to discuss how artificial intelligence (AI) is shaping the future of Real-Time Crime Centers (RTCCs). Nikki shares firsthand experiences with facial recognition, vehicle detection, and integrated camera systems that accelerate investigations while emphasizing that AI is a lead, not a conclusion. From license plate readers and convoy searches to predictive alerts and policy compliance tools, Nikki discusses practical ways analysts can use AI responsibly without losing the human touch. The conversation also explores how technology affects data volume, ethical considerations, and what the next five years may hold for analysts as AI continues to evolve. Nikki also provides updates from the National Real Time Crime Center Association (NRTCCA) conference. She introduces a new organization for civilian staff, the National Association of Professional Staff in Public Safety (NAPSPPS). She ends by sharing exciting personal news about becoming a new mom. [Note: Description produced by ChatGPT.] Get to know more about Nikki by listening to his episode on Analyst Talk With Jason Elder: https://www.leapodcasts.com/e/atwje-nikki-nichole-north-%c2%a0real-time-crime-analyst/ Name Drops: Related Links: https://nrtcca.org/ https://www.axon.com/products/axon-fusus https://www.napsps.org/ https://www.theiacp.org/resources/document/artificial-intelligence-and-policing https://www.justice.gov/opa/pr/department-justice-releases-report-facial-recognition-technology https://www.leapodcasts.com/e/analyst-talk-angie-stroda-from-corporate-world-to-crime-analysis/ Association(s) Mentioned: National Real Time Crime Center Association; National Association of Professional Staff in Public Safety Vendor(s) Mentioned: Contact: https://www.linkedin.com/in/nikki-north-196377239/, nnorth@flaglersheriff.com; nnorth@nrtcca.org Transcript: https://mcdn.podbean.com/mf/web/k5j5amjma2pbhhuy/RCATT05_transcript.pdf Podcast Writer: Podcast Researcher: Theme Song: Written and Recorded by The Rough & Tumble. Find more of their music at www.theroughandtumble.com. Logo: Designed by Kyle McMullen. Please visit www.moderntype.com for any printable business forms and planners. Podcast Email: leapodcasts@gmail.com Podcast Webpage: www.leapodcasts.com Podcast Twitter: @leapodcasts
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
Welcome to Analyst Talk with Jason Elder.
(00:01):
It's like coffee with an analyst,or it could be whiskey with an
analyst reading a spreadsheet,linking crime events, identifying a
series, and getting the latest scoopon association news and training.
So please don't beat that analystand join us as we define the law
enforcement analysis profession.
One episode at a time.
How are you doing?
Analyst, Jason Elder herewith another LEA podcast Deep
(00:23):
Dive, real Crime all the time.
And with me as always is one Nikki North.
Nikki, how we doing today?
We're doing good
Jason, how
are you doing?
I am doing very well.
It is been a whilesince you and I chatted.
We had last did it , winterin the spring, and now
we're beating the heat inFlorida and now it's getting into
(00:44):
the the Florida fall weather.
Absolutely.
It's definitely startingto get a little chilly.
Yes, but Florida cold.
That's right.
All right.
What are we talking about today?
Let's do some AI today, I think.
Sounds good.
I know that's definitely a hot topic.
Yeah, that's a pretty big topic.
What's your message that youwant to send AI as it relates
(01:07):
to real time crime centers?
I
think the biggest thing is there'sso many different variations of
ai, which is a whole conversationin itself for another day.
But just to really drive home the fact ofAI is just a lead for an investigation.
You cannot make AI your whole case asgreat of a technology component as it is.
It's just a lead.
You still have to do actualinvestigative work beyond what
(01:30):
a computer is telling you.
Yeah, and I can see wherethat would be easy to do.
Just go with what a AI tells you.
Plug in a picture to facial recognition.
It's like, oh, this is a 99% match.
Perfect.
That's all I needed.
No, that's not the case.
Yeah.
Well, I would imagine you need othercorroborating information that it
(01:51):
might just point you in the rightdirection, and then once you get
more evidence, you're like, oh, okay.
You can then corroborate or downplay.
That tip that you, I guess it wouldjust be like any other tip that the
real time crime center might evaluate.
Correct.
Exactly.
Like you said, it'sliterally a pointer system.
That is exactly what it is, is when yousearch a name and you're trying to find
(02:11):
leads, it's the same thing just withimagery or keywords or whatever it may be.
Exactly.
When you're talking about imagery, isthat getting into facial recognition?
Is that getting intomaybe graffiti tattoos?
What are we talking about there?
I feel like that's like the whole conceptof like, AI is, when you hear it, everyone
just thinks like, oh, like you havechat GPT and Gemini, and all this stuff,
(02:35):
but there's also facial recognition.
There's also mm-hmm.
Detection stuff, likeshow me all the red cars.
Show me all the truckswith this in the back.
So I know that's a lot of what differentplatforms are working for is as soon as
they hear ai they think it's the samething when there's just so many different
components to what AI actually is.
Yeah, no.
At your department, do youhave a particular AI model
(02:58):
or program that you use?
So we have some facialrecognition programs there.
We have a couple of those at least.
But we use at.
Pretty much daily, I wouldsay, as good pointer systems.
And then we also have integratedinto our fuss platform.
We have some of the AI on our actualcameras too, so our traffic cameras,
some of the other ones to help us detectlike suspect vehicles or anything.
(03:20):
It's a good preliminary checkof, Hey, we just had a car break.
We're looking for a red sedan.
Cool.
We have a camera in that park.
Let's just run it through real quick.
Oh, there's nothing like, we'renot just gonna stop there though.
We're still gonna lookand manually make sure.
But it's at least could save you afew seconds, could save you a few
minutes, could save you a few hours.
Mm-hmm.
And it's been a minute.
Remind me about Fugitive.
(03:42):
So Fuss is basically our platformthat integrates everything into one.
So like we have our city trafficcameras in there, we have all of
our license plate readers in there.
We have some of our own deployedcameras in there, our body
cameras, which we can live stream.
Pretty much anything that we couldthink of to try to put into one
map for us to be able to accessas well as everyone in the agency.
And then of course, everyone hasdifferent permissions within it
(04:03):
because we do have like our schoolcameras and that kind of stuff too.
Yeah.
Now are you.
So fancy that you can haveAI running automatically.
'cause some of this stuff I could justsee license plate readers or the cameras
stuff that it just picks up automaticallythat AI could evaluate as opposed to
(04:24):
an analyst saying, okay ai, we wantthis specific thing to be evaluated.
There's definitely programs outthere like brief cam and things along
those lines that can do that for you.
Right now we personally aren't using any,being a smaller agency, it's just not
one of the things we've needed thus far.
Mm-hmm.
But there's absolutely abenefit to stuff like that.
I know a big one could be if you havelike a known narcotics location that
(04:46):
it triggers telling you if there'slike an increased amount of traffic
in front of that location, you couldautomatically get notified and then
notice, send resources that way.
Like you said, vandalism, ifyou know you're having vandalism
in your parks after hours.
You all of a sudden get two kidsshowing up, it could notify you.
And that way to send someonethat way and check it out,
geez, oh yeah, that's, that soundslike a lot of potential there.
(05:09):
A lot, a lot.
Maybe some potential false positives orsending you down the wrong path, but that
definitely has some potential there too.
At Lace curb some of the, some of theissues that a department may have.
Absolutely.
That's what we've, so far, being allcivilian staff, we really don't try to go
the proactive route as much at this point.
(05:31):
Mm-hmm.
We mostly try to stay reactive that way.
You kind of still keep that reflectionfrom your community of we're only
accessing the technology based onwhat you people are calling in.
We're not using it just from what we'reseeing or catching or anything like that.
It's because someone called in.
Yeah.
Now, do you have any goodstories about using AI in, in
the real time crime center?
Trying to think of a good one here.
(05:51):
I know like I said, facial rec isprobably our biggest AI contributor that
we use to get us a lot of good leads.
Like, we'll find people that,it gives us a good match.
It'll say like, it's a 99% match, andas soon as you start digging into them,
if it's, you're looking at 'em for aburglary, sure enough, when you start
looking at other databases, oh look,they have a history of 15 burglaries.
Probably a pretty good chance.
This is gonna be our suspect.
Okay.
Now we'd to figure it outwhat their vehicle is.
(06:13):
Oh look, their vehicle's never been in ourcounty except for the day of this crime.
Yeah.
Yeah.
So I'd say facial wreck isdefinitely our most top used.
We have used, of course, like some ofthe features within our fuss platform
though to find like specific colorsof vehicles and stuff like that.
I haven't had like some crazy likeridiculous success story that I
can think of right now, but it isstuff that is used almost daily.
(06:33):
Yeah, that reminds me.
I recently had on the show, Angie Strodafrom Hutchinson PD in Kansas City.
She had a really cool case in whichshe had A-A-A-T-M robbery series.
Mm-hmm.
And she used license plate readerand identified which license plate
(06:57):
was all in the same location.
Oh, yeah.
Yep.
During all the ATM
Yep.
Robberies, and then she was thinkingthat it was outta state, so she threw
out the, all the in-state ones and,and then I think, and whittled it
down to just one vehicle, which wasreally cool to use the data like that.
So I can, I can imagine in thatscenario too, that AI would be helpful.
(07:20):
If you ask the right question,okay, which ones are all in the same
location, or, which ones have hitall these cameras, gimme that list,
and something like that.
You could even take it a stepfurther with some of the AI too, and
do what's called a convoy search.
And basically see too, like if there'sa secondary vehicle that's like
consistently traveling with that vehicle.
So in like for example, when we had an ATMburglary a few years back we identified,
(07:45):
we had a truck that got stolen and wewere able to kind of figure out that
truck traveling with the other truckbased on using those kind of searches.
So we looked at like specific sitesand saw these two trucks together and
we're like, okay, something suspicious.
This is probably the truckthat tried to pull the ATM.
Then here's the stolen vehiclethat's also linked to it.
Yeah.
I mean, I can, it's interesting becauseI think obviously AI is just really
(08:09):
being implemented and I, I, I feel as thetechnology gets the, such that everybody
has basically their own AI module.
I'm not using that right.
Correct term, but everybody.
Each police department's gonnahave basically their own AI
module running on their system.
That in terms of the real time crimecenter world, it's okay adding a name
(08:36):
and then it's the AI is making decisions. Identifying all the sources, which
ones are the best for the question thatyou're asking it, and returning a list
based on the history and and whatnot.
So instead of the analysts going throughand checking through 27 databases,
(08:57):
this AI will go through, run a name,plate, or location, and then spit back
best chance of being the most useful.
And like what I see the future ofthat even being too is say you have
a known shoplifting ring and they'rehitting targets all across the state.
It could get to the point where you're,everyone has target cameras integrated
(09:19):
then, and one target store plugs in thatname and it alerts you when they walk
into another target store or somethingbased on the facial recognition.
Could be the future ofstuff like that too.
Yeah.
Now do, now do you feel that the, thefacial recognition technology is that.
Accurate.
It's, we've gotten our, our resultstogether are normally pretty accurate.
Like if it's giving you a match that'slike 99%, it's normally pretty solid.
(09:42):
So normally within like the 96 toa hundred percent range, like we're
definitely looking at it to see, and thenonce you kind of start getting lower than
that it's a little more questionable.
Mm-hmm.
I mean, when we get results,they're normally pretty accurate.
So, I mean, it seems spot on so far.
I think it's only a matter of time asthey figure out more and more sources
to where to pull pictures from too.
'cause a lot of it rightnow is only booking related.
(10:02):
So if you've never been arrested,it's a little harder sometimes.
Mm-hmm.
There are some social media ones,but you have all these avatars
that people are using now for theirprofile pictures and stuff like that.
So it's not necessarily foolproof quiteyet, but I'm sure that's something
that'll continue to advance too.
Yeah.
I guess what about using ai.
I guess maybe not typically, or maybe,maybe stuff that's isn't picture related.
(10:23):
So it's not related to thelicense plate reader or a camera.
It's so
I've been doing a lot an example of
that
with different AI stuff lately.
So I mean, there's allkinds of ones to think of.
There's some where.
It's even just policy based of, A lot ofpeople use power DMS for their policies.
I know that's what we use and it isalmost impossible to actually try
to search for a keyword in a policy.
(10:44):
So there's certain platformsthat are basically saying like,
Hey, we're C is compliant.
He can pull all of our policies inour program and we'll make it so
it works for accreditations too.
And it basically comes tochat GPT for your policies.
Then there's other ones that go a stepfurther and integrate legal statutes.
And then there's other ones whereyou could plug in your case report
and say, Hey, are there any otherleads I could have followed up on?
(11:06):
Did I miss something?
Was there somethingelse I could have done?
What charges can I do?
I mean, there's so many differentvariations of stuff that you can
do with it for timelines even.
I know that's another commonrequest for crime analysts is
to help with making timelines.
There's programs out there that youbasically plug in a case report.
I trust to generate a timeline for youand you just kind of mess around with
the pictures and the wording and makesure everything lines up correctly.
(11:27):
Yeah,
no, I, I think there's, there'sdefinitely a lot of potential
for documentation using ai.
Mm-hmm.
, And this goes hand in hand with what wehave talked about previously with, which
is just the enormous amount of data.
That analysts have access to now.
Mm-hmm.
Right.
I am always flabbergasted when Italk to different analysts and, and.
(11:51):
They how much different data sets,databases, resources they have access to,
and in the real time crime center world.
And I think we had this,which might've been the world.
Our, one of our first conversationswas, when to say when, when to go?
When are you going down arabbit hole that's not gonna
have a return on investment?
How do you know when to stop?
(12:13):
In terms of all these different leadsand tracking the different sources.
And so I think because of theenormous amount of data that is at
the disposal of these real-time crimecenters, it's, it's only fitting
that you're gonna have to have an AI.
To help process all that data.
(12:35):
Yeah, because as you get more and moreof that technology too, you figure you
used to get a shoplifter, you maybe gota video, but if you didn't have any kind
of vehicle description or anything, oreven if you did, your best bet was you
could kind of go into the DHMB database.
And look up, make and model andhope you found the right vehicle.
Mm-hmm.
Now you have these license platereaders that you can search if
they were in the area at that time.
You can search based on descriptions,you can search based on stickers,
(12:59):
even like, there's so many differentvariations of what you can search now.
So it's making it that cases thatused to just get closed out so
quickly, that's not the case anymore.
Now there are investigative leadsthat could be followed, that weren't
able to be previously followed.
Yeah.
No, I I think there's, there'sobviously a lot to that.
And for the listeners, if youhave a, a good story you want
(13:19):
to share reach out to us.
Email me at elliot podcast@gmail.com,or just comment on one of
our social media sites.
We're on LinkedIn Facebook, you know.
So just just find us and , giveus a story nikki I guess anything
else for the good of the order?
In terms of the topic?
(13:39):
I think that's about it for the topic,like I said, I mean, if you wanted each
specific one, that's a whole separateconversation itself, but I think that
definitely is a good general coverageof just a glimpse into the AI life.
Yeah.
And it's common., One of thequestions I like to ask on the
podcast is the return on investment.
What is something that an analystcan study now that's going to be
important five years from now?
(14:00):
And it's, AI is definitely goingto be there five years from now.
So it's it's definitely somethingthat, it'll be interesting to see how
it matures and how it's used in, infive years or as we get to the 2030s.
Absolutely.
'cause I know even since I startedlooking originally when we started
the crime center five years ago,compared to now, there's so many
(14:21):
more programs advertising it.
Yeah.
Now that's a question.
Do you think that potentiallycould put the real time
crime centers outta business?
Right.
I feel like
it's a such a mixedanswer because yes and no.
'cause you still need that humanelement to verify what you're getting.
Sure.
Because I know.
So things like D's reads O's and B's, so.
(14:42):
There's still always that elementas it continues to improve.
Who knows, maybe from a hundred yearsfrom now, the technology might be
that good to knock it out of business.
Yeah, well, I would, where I guess I,my mind was going, I didn't even go to
that part of it where I was going withis as patrol officers or investigators
have access to that at their fingertips,that there's not that need to go to
(15:04):
the central hub of a realtime crimecenter because all of that is at their
disposal, maybe in their vehicle.
I think if it gets efficient enough,maybe I know even for now, like
for example, like we gave accessto everyone in our agency to get to
fuss, to get to the traffic camerasand we still get daily requests
to get video, the traffic cameras.
Yeah.
So I think some like maybe if itbecame like a quicker process.
(15:28):
And it was literally like a chatGPT, like export the video from
this time and date and it did it.
Yeah.
Then maybe you might get there one day,but I think a lot of people are used to
having that piece to fall back on andstill need for the time being at least.
All right.
For at least for now, theemployment future is fine with
the real time crime center?
Yes, I would say so.
(15:49):
All right, let's let's justfinish up with some updates here.
So what do you want to promote today?
So we'll just go on toa couple quick things.
So I know we just recently had theNational Realtime Crime Center Association
conference up in Cobb County, Georgia.
I unfortunately did not attend thisyear, but I heard a lot of great things
and it was a sold out conference yetagain, so that's awesome to hear.
To the point.
Pretty much going forward.
(16:09):
They're always gonna be inconvention centers now because of
the numbers we've been looking at.
So next year we'll be over inPhoenix, so they're working out some
details with the Phoenix ConventionCenter and all that over there.
Now that the comp do you,is done on this side?
I'm kind of looking more into actuallydeveloping a Southeast region now.
So I am the Southeast region director.
I've kind of laid off of it just 'causeI knew the conference was here this year.
So starting around January, I'm gonnabe looking for some people to get
(16:32):
more involved and hopefully have havesome smaller level events beyond,
besides just the conference too, sothat way people can't make it over
to the west part of the country.
You still have optionsto get some training.
Yeah.
And back to the conference, do you knowhow many people attended this year?
I
know it was over a thousand.
Wow.
Nice.
More than they expected.
Nice, nice.
(16:53):
And I guess in terms of the, your regionalassociation, I know it's the Southeast
United States, but if somebody's listeningto this, maybe wants to get involved or.
Or join.
What states , does theregional part include it
spread pretty far?
Do not quote me 100% on this.
I have to officially verify becauseI know it was something we're still
working through, but I believe itstarts pretty much over at Louisiana
(17:15):
and makes its way up into Virginia.
So it's a pretty wide span ofwhere exactly it goes to all.
All right.
And of course all the way down to Florida.
Alright, nice.
And then as always, we'll leavecontact information for Nikki and some
additional links in the show notes.
And then you there, oh, there was anotherassociation you wanted to promote?
Yes.
This one's relatively newer thatI recently learned about and just
(17:38):
definitely starting to get a little moreinvolved with them and learn some stuff.
It's the National Association ofProfessional Staff and Public Safety.
And I mean, it ranges.
It's a good basically it's to those ofus without a badge that are essentially
civilians, which is the term we're tryingto stray away from, is opportunities
within law enforcement trainingsto just kind of give you something.
(17:58):
So, I mean, it's a widearray of people though.
Like it could be executive assistants,human resources, crime analysts.
It's just kind of that extra piecewithin the agency to have opportunities
to train and connect and to includesome leadership training too, because
it is a little different being aprofessional staff leadership position
compared to a sworn leadership position.
Yeah, well I'm sure it would probablybe beneficial to get this same room as
(18:21):
HR staff and to realize that they arehuman and they have a job to do and that
with them, you not knowing who they areevery time you get frustrated with it.
Hr you swearing under your breath.
Exactly.
So I'm sure they appreciatethat side of it too.
Yeah.
It does sound like an interestingcombination of people I, I think it's
(18:44):
would be fascinating that diversity,because there's obviously opportunities
to learn from those various aspects.
Mm-hmm.
Like, it's everythinglike crime scene, it.
Your dispatchers almost anythingyou could think of is what
they're looking to target.
Mm-hmm.
All right.
And you said they weretheir conferences in Miami?
(19:06):
So they have a one dayevent coming up here.
Okay.
Next month in November in Miami.
And then they have a bigger conferencethat they're working on as well.
All
right
now, and then they're starting tomake their way to some of the events.
So they're at IACP andjust different stuff.
They're starting to get involvedwith the National Real Time
Crime Center Association.
So that was actually how I learnedabout them was from partnering the
crime center Association with them.
(19:26):
Yeah, no, that'll befascinating to read up.
On to see what their, their mission is.
While I like the diversity, it isalso kind of a lots of players means
that that can get pretty heavy.
Correct.
That you can't really have directtrainings necessarily then.
'cause it can't be targeted directlyat analysts when you have all these
(19:47):
other different positions in there.
Yeah, yeah.
But it's always.
Always a good networking opportunityand an opportunity to learn about
some other aspect of public safety.
'cause you never know when it'sgonna be important as an analyst.
Absolutely.
All right, Nikki, and hey, beforeI go, congratulations on being
a mother for the first time.
Yes, thank you.
(20:08):
That's why we had such a gapbetween spring and fall was I was
busy with my little guy, so yeah.
Now we're getting backinto the routine again.
All right.
And it was it wasn't avery simple pregnancy.
He was a little early, right?
Yeah.
He came two months early and spentjust shy of a month in the nicu.
So.
Definitely took us on an adventure.
Oh man.
Now it, it's, that might be TMIdid do, do premature babies running
(20:32):
either one of your families.
So you and your husband?
They do.
Actually his mother was a premature baby.
Yeah.
But ours was more directlyhe was an IVF baby.
We had been trying for years.
Oh, okay.
So from IVF, your chances ofpreeclampsia were increased, so Oh, okay.
Felt preeclampsia.
Yeah.
Well see, that's all kinds of new stuffthat I, that I learned, but, mm-hmm.
(20:52):
Obviously, I'm glad that the wholething worked out and that it was
certainly scary there for a while.
But how old is he now?
He
just turned six months on the ninth, so,
man.
Yeah.
, He's probably already a little mobile andhe's probably looking to be walking here
he is rolling like crazy andtrying to figure out how to crawl.
(21:15):
So yeah.
He has been, he was very mobileinside too, so he will, once he
gets moving, he will be moving.
Yeah, yeah.
No, that's that's certainly fun.
Quite an adventure ahead ofyou, so congratulations again.
Well, thank you.
All right, Nikki thankyou again for this time.
Always look forward to chattingwith you, , real crime all the time.
(21:36):
Until next time, take care.
Thank you for making it tothe end of another episode of
Analyst Talk with Jason Elder.
You can show your support by sharingthis and other episodes found
on our website@www.podcasts.com.
If you have a topic you would like us tocover or have a suggestion for our next
guest, please send us an email at elliotpodcasts@gmail.com next time analysts.
(21:59):
Keep talking.
Welcome to Analyst Talk with Jason Elder.
(00:01):
It's like coffee with an analyst,or it could be whiskey with an
analyst reading a spreadsheet,linking crime events, identifying a
series, and getting the latest scoopon association news and training.
So please don't beat that analystand join us as we define the law
enforcement analysis profession.
One episode at a time.
How are you doing?
Analyst, Jason Elder herewith another LEA podcast Deep
(00:23):
Dive, real Crime all the time.
And with me as always is one Nikki North.
Nikki, how we doing today?
We're doing good
Jason, how
are you doing?
I am doing very well.
It is been a whilesince you and I chatted.
We had last did it , winterin the spring, and now
we're beating the heat inFlorida and now it's getting into
(00:44):
the the Florida fall weather.
Absolutely.
It's definitely startingto get a little chilly.
Yes, but Florida cold.
That's right.
All right.
What are we talking about today?
Let's do some AI today, I think.
Sounds good.
I know that's definitely a hot topic.
Yeah, that's a pretty big topic.
What's your message that youwant to send AI as it relates
(01:07):
to real time crime centers?
I
think the biggest thing is there'sso many different variations of
ai, which is a whole conversationin itself for another day.
But just to really drive home the fact ofAI is just a lead for an investigation.
You cannot make AI your whole case asgreat of a technology component as it is.
It's just a lead.
You still have to do actualinvestigative work beyond what
(01:30):
a computer is telling you.
Yeah, and I can see wherethat would be easy to do.
Just go with what a AI tells you.
Plug in a picture to facial recognition.
It's like, oh, this is a 99% match.
Perfect.
That's all I needed.
No, that's not the case.
Yeah.
Well, I would imagine you need othercorroborating information that it
(01:51):
might just point you in the rightdirection, and then once you get
more evidence, you're like, oh, okay.
You can then corroborate or downplay.
That tip that you, I guess it wouldjust be like any other tip that the
real time crime center might evaluate.
Correct.
Exactly.
Like you said, it'sliterally a pointer system.
That is exactly what it is, is when yousearch a name and you're trying to find
(02:11):
leads, it's the same thing just withimagery or keywords or whatever it may be.
Exactly.
When you're talking about imagery, isthat getting into facial recognition?
Is that getting intomaybe graffiti tattoos?
What are we talking about there?
I feel like that's like the whole conceptof like, AI is, when you hear it, everyone
just thinks like, oh, like you havechat GPT and Gemini, and all this stuff,
(02:35):
but there's also facial recognition.
There's also mm-hmm.
Detection stuff, likeshow me all the red cars.
Show me all the truckswith this in the back.
So I know that's a lot of what differentplatforms are working for is as soon as
they hear ai they think it's the samething when there's just so many different
components to what AI actually is.
Yeah, no.
At your department, do youhave a particular AI model
(02:58):
or program that you use?
So we have some facialrecognition programs there.
We have a couple of those at least.
But we use at.
Pretty much daily, I wouldsay, as good pointer systems.
And then we also have integratedinto our fuss platform.
We have some of the AI on our actualcameras too, so our traffic cameras,
some of the other ones to help us detectlike suspect vehicles or anything.
(03:20):
It's a good preliminary checkof, Hey, we just had a car break.
We're looking for a red sedan.
Cool.
We have a camera in that park.
Let's just run it through real quick.
Oh, there's nothing like, we'renot just gonna stop there though.
We're still gonna lookand manually make sure.
But it's at least could save you afew seconds, could save you a few
minutes, could save you a few hours.
Mm-hmm.
And it's been a minute.
Remind me about Fugitive.
(03:42):
So Fuss is basically our platformthat integrates everything into one.
So like we have our city trafficcameras in there, we have all of
our license plate readers in there.
We have some of our own deployedcameras in there, our body
cameras, which we can live stream.
Pretty much anything that we couldthink of to try to put into one
map for us to be able to accessas well as everyone in the agency.
And then of course, everyone hasdifferent permissions within it
(04:03):
because we do have like our schoolcameras and that kind of stuff too.
Yeah.
Now are you.
So fancy that you can haveAI running automatically.
'cause some of this stuff I could justsee license plate readers or the cameras
stuff that it just picks up automaticallythat AI could evaluate as opposed to
(04:24):
an analyst saying, okay ai, we wantthis specific thing to be evaluated.
There's definitely programs outthere like brief cam and things along
those lines that can do that for you.
Right now we personally aren't using any,being a smaller agency, it's just not
one of the things we've needed thus far.
Mm-hmm.
But there's absolutely abenefit to stuff like that.
I know a big one could be if you havelike a known narcotics location that
(04:46):
it triggers telling you if there'slike an increased amount of traffic
in front of that location, you couldautomatically get notified and then
notice, send resources that way.
Like you said, vandalism, ifyou know you're having vandalism
in your parks after hours.
You all of a sudden get two kidsshowing up, it could notify you.
And that way to send someonethat way and check it out,
geez, oh yeah, that's, that soundslike a lot of potential there.
(05:09):
A lot, a lot.
Maybe some potential false positives orsending you down the wrong path, but that
definitely has some potential there too.
At Lace curb some of the, some of theissues that a department may have.
Absolutely.
That's what we've, so far, being allcivilian staff, we really don't try to go
the proactive route as much at this point.
(05:31):
Mm-hmm.
We mostly try to stay reactive that way.
You kind of still keep that reflectionfrom your community of we're only
accessing the technology based onwhat you people are calling in.
We're not using it just from what we'reseeing or catching or anything like that.
It's because someone called in.
Yeah.
Now, do you have any goodstories about using AI in, in
the real time crime center?
Trying to think of a good one here.
(05:51):
I know like I said, facial rec isprobably our biggest AI contributor that
we use to get us a lot of good leads.
Like, we'll find people that,it gives us a good match.
It'll say like, it's a 99% match, andas soon as you start digging into them,
if it's, you're looking at 'em for aburglary, sure enough, when you start
looking at other databases, oh look,they have a history of 15 burglaries.
Probably a pretty good chance.
This is gonna be our suspect.
Okay.
Now we'd to figure it outwhat their vehicle is.
(06:13):
Oh look, their vehicle's never been in ourcounty except for the day of this crime.
Yeah.
Yeah.
So I'd say facial wreck isdefinitely our most top used.
We have used, of course, like some ofthe features within our fuss platform
though to find like specific colorsof vehicles and stuff like that.
I haven't had like some crazy likeridiculous success story that I
can think of right now, but it isstuff that is used almost daily.
(06:33):
Yeah, that reminds me.
I recently had on the show, Angie Strodafrom Hutchinson PD in Kansas City.
She had a really cool case in whichshe had A-A-A-T-M robbery series.
Mm-hmm.
And she used license plate readerand identified which license plate
(06:57):
was all in the same location.
Oh, yeah.
Yep.
During all the ATM
Yep.
Robberies, and then she was thinkingthat it was outta state, so she threw
out the, all the in-state ones and,and then I think, and whittled it
down to just one vehicle, which wasreally cool to use the data like that.
So I can, I can imagine in thatscenario too, that AI would be helpful.
(07:20):
If you ask the right question,okay, which ones are all in the same
location, or, which ones have hitall these cameras, gimme that list,
and something like that.
You could even take it a stepfurther with some of the AI too, and
do what's called a convoy search.
And basically see too, like if there'sa secondary vehicle that's like
consistently traveling with that vehicle.
So in like for example, when we had an ATMburglary a few years back we identified,
(07:45):
we had a truck that got stolen and wewere able to kind of figure out that
truck traveling with the other truckbased on using those kind of searches.
So we looked at like specific sitesand saw these two trucks together and
we're like, okay, something suspicious.
This is probably the truckthat tried to pull the ATM.
Then here's the stolen vehiclethat's also linked to it.
Yeah.
I mean, I can, it's interesting becauseI think obviously AI is just really
(08:09):
being implemented and I, I, I feel as thetechnology gets the, such that everybody
has basically their own AI module.
I'm not using that right.
Correct term, but everybody.
Each police department's gonnahave basically their own AI
module running on their system.
That in terms of the real time crimecenter world, it's okay adding a name
(08:36):
and then it's the AI is making decisions. Identifying all the sources, which
ones are the best for the question thatyou're asking it, and returning a list
based on the history and and whatnot.
So instead of the analysts going throughand checking through 27 databases,
(08:57):
this AI will go through, run a name,plate, or location, and then spit back
best chance of being the most useful.
And like what I see the future ofthat even being too is say you have
a known shoplifting ring and they'rehitting targets all across the state.
It could get to the point where you're,everyone has target cameras integrated
(09:19):
then, and one target store plugs in thatname and it alerts you when they walk
into another target store or somethingbased on the facial recognition.
Could be the future ofstuff like that too.
Yeah.
Now do, now do you feel that the, thefacial recognition technology is that.
Accurate.
It's, we've gotten our, our resultstogether are normally pretty accurate.
Like if it's giving you a match that'slike 99%, it's normally pretty solid.
(09:42):
So normally within like the 96 toa hundred percent range, like we're
definitely looking at it to see, and thenonce you kind of start getting lower than
that it's a little more questionable.
Mm-hmm.
I mean, when we get results,they're normally pretty accurate.
So, I mean, it seems spot on so far.
I think it's only a matter of time asthey figure out more and more sources
to where to pull pictures from too.
'cause a lot of it rightnow is only booking related.
(10:02):
So if you've never been arrested,it's a little harder sometimes.
Mm-hmm.
There are some social media ones,but you have all these avatars
that people are using now for theirprofile pictures and stuff like that.
So it's not necessarily foolproof quiteyet, but I'm sure that's something
that'll continue to advance too.
Yeah.
I guess what about using ai.
I guess maybe not typically, or maybe,maybe stuff that's isn't picture related.
(10:23):
So it's not related to thelicense plate reader or a camera.
It's so
I've been doing a lot an example of
that
with different AI stuff lately.
So I mean, there's allkinds of ones to think of.
There's some where.
It's even just policy based of, A lot ofpeople use power DMS for their policies.
I know that's what we use and it isalmost impossible to actually try
to search for a keyword in a policy.
(10:44):
So there's certain platformsthat are basically saying like,
Hey, we're C is compliant.
He can pull all of our policies inour program and we'll make it so
it works for accreditations too.
And it basically comes tochat GPT for your policies.
Then there's other ones that go a stepfurther and integrate legal statutes.
And then there's other ones whereyou could plug in your case report
and say, Hey, are there any otherleads I could have followed up on?
(11:06):
Did I miss something?
Was there somethingelse I could have done?
What charges can I do?
I mean, there's so many differentvariations of stuff that you can
do with it for timelines even.
I know that's another commonrequest for crime analysts is
to help with making timelines.
There's programs out there that youbasically plug in a case report.
I trust to generate a timeline for youand you just kind of mess around with
the pictures and the wording and makesure everything lines up correctly.
(11:27):
Yeah,
no, I, I think there's, there'sdefinitely a lot of potential
for documentation using ai.
Mm-hmm.
, And this goes hand in hand with what wehave talked about previously with, which
is just the enormous amount of data.
That analysts have access to now.
Mm-hmm.
Right.
I am always flabbergasted when Italk to different analysts and, and.
(11:51):
They how much different data sets,databases, resources they have access to,
and in the real time crime center world.
And I think we had this,which might've been the world.
Our, one of our first conversationswas, when to say when, when to go?
When are you going down arabbit hole that's not gonna
have a return on investment?
How do you know when to stop?
(12:13):
In terms of all these different leadsand tracking the different sources.
And so I think because of theenormous amount of data that is at
the disposal of these real-time crimecenters, it's, it's only fitting
that you're gonna have to have an AI.
To help process all that data.
(12:35):
Yeah, because as you get more and moreof that technology too, you figure you
used to get a shoplifter, you maybe gota video, but if you didn't have any kind
of vehicle description or anything, oreven if you did, your best bet was you
could kind of go into the DHMB database.
And look up, make and model andhope you found the right vehicle.
Mm-hmm.
Now you have these license platereaders that you can search if
they were in the area at that time.
You can search based on descriptions,you can search based on stickers,
(12:59):
even like, there's so many differentvariations of what you can search now.
So it's making it that cases thatused to just get closed out so
quickly, that's not the case anymore.
Now there are investigative leadsthat could be followed, that weren't
able to be previously followed.
Yeah.
No, I I think there's, there'sobviously a lot to that.
And for the listeners, if youhave a, a good story you want
(13:19):
to share reach out to us.
Email me at elliot podcast@gmail.com,or just comment on one of
our social media sites.
We're on LinkedIn Facebook, you know.
So just just find us and , giveus a story nikki I guess anything
else for the good of the order?
In terms of the topic?
(13:39):
I think that's about it for the topic,like I said, I mean, if you wanted each
specific one, that's a whole separateconversation itself, but I think that
definitely is a good general coverageof just a glimpse into the AI life.
Yeah.
And it's common., One of thequestions I like to ask on the
podcast is the return on investment.
What is something that an analystcan study now that's going to be
important five years from now?
(14:00):
And it's, AI is definitely goingto be there five years from now.
So it's it's definitely somethingthat, it'll be interesting to see how
it matures and how it's used in, infive years or as we get to the 2030s.
Absolutely.
'cause I know even since I startedlooking originally when we started
the crime center five years ago,compared to now, there's so many
(14:21):
more programs advertising it.
Yeah.
Now that's a question.
Do you think that potentiallycould put the real time
crime centers outta business?
Right.
I feel like
it's a such a mixedanswer because yes and no.
'cause you still need that humanelement to verify what you're getting.
Sure.
Because I know.
So things like D's reads O's and B's, so.
(14:42):
There's still always that elementas it continues to improve.
Who knows, maybe from a hundred yearsfrom now, the technology might be
that good to knock it out of business.
Yeah, well, I would, where I guess I,my mind was going, I didn't even go to
that part of it where I was going withis as patrol officers or investigators
have access to that at their fingertips,that there's not that need to go to
(15:04):
the central hub of a realtime crimecenter because all of that is at their
disposal, maybe in their vehicle.
I think if it gets efficient enough,maybe I know even for now, like
for example, like we gave accessto everyone in our agency to get to
fuss, to get to the traffic camerasand we still get daily requests
to get video, the traffic cameras.
Yeah.
So I think some like maybe if itbecame like a quicker process.
(15:28):
And it was literally like a chatGPT, like export the video from
this time and date and it did it.
Yeah.
Then maybe you might get there one day,but I think a lot of people are used to
having that piece to fall back on andstill need for the time being at least.
All right.
For at least for now, theemployment future is fine with
the real time crime center?
Yes, I would say so.
(15:49):
All right, let's let's justfinish up with some updates here.
So what do you want to promote today?
So we'll just go on toa couple quick things.
So I know we just recently had theNational Realtime Crime Center Association
conference up in Cobb County, Georgia.
I unfortunately did not attend thisyear, but I heard a lot of great things
and it was a sold out conference yetagain, so that's awesome to hear.
To the point.
Pretty much going forward.
(16:09):
They're always gonna be inconvention centers now because of
the numbers we've been looking at.
So next year we'll be over inPhoenix, so they're working out some
details with the Phoenix ConventionCenter and all that over there.
Now that the comp do you,is done on this side?
I'm kind of looking more into actuallydeveloping a Southeast region now.
So I am the Southeast region director.
I've kind of laid off of it just 'causeI knew the conference was here this year.
So starting around January, I'm gonnabe looking for some people to get
(16:32):
more involved and hopefully have havesome smaller level events beyond,
besides just the conference too, sothat way people can't make it over
to the west part of the country.
You still have optionsto get some training.
Yeah.
And back to the conference, do you knowhow many people attended this year?
I
know it was over a thousand.
Wow.
Nice.
More than they expected.
Nice, nice.
(16:53):
And I guess in terms of the, your regionalassociation, I know it's the Southeast
United States, but if somebody's listeningto this, maybe wants to get involved or.
Or join.
What states , does theregional part include it
spread pretty far?
Do not quote me 100% on this.
I have to officially verify becauseI know it was something we're still
working through, but I believe itstarts pretty much over at Louisiana
(17:15):
and makes its way up into Virginia.
So it's a pretty wide span ofwhere exactly it goes to all.
All right.
And of course all the way down to Florida.
Alright, nice.
And then as always, we'll leavecontact information for Nikki and some
additional links in the show notes.
And then you there, oh, there was anotherassociation you wanted to promote?
Yes.
This one's relatively newer thatI recently learned about and just
(17:38):
definitely starting to get a little moreinvolved with them and learn some stuff.
It's the National Association ofProfessional Staff and Public Safety.
And I mean, it ranges.
It's a good basically it's to those ofus without a badge that are essentially
civilians, which is the term we're tryingto stray away from, is opportunities
within law enforcement trainingsto just kind of give you something.
(17:58):
So, I mean, it's a widearray of people though.
Like it could be executive assistants,human resources, crime analysts.
It's just kind of that extra piecewithin the agency to have opportunities
to train and connect and to includesome leadership training too, because
it is a little different being aprofessional staff leadership position
compared to a sworn leadership position.
Yeah, well I'm sure it would probablybe beneficial to get this same room as
(18:21):
HR staff and to realize that they arehuman and they have a job to do and that
with them, you not knowing who they areevery time you get frustrated with it.
Hr you swearing under your breath.
Exactly.
So I'm sure they appreciatethat side of it too.
Yeah.
It does sound like an interestingcombination of people I, I think it's
(18:44):
would be fascinating that diversity,because there's obviously opportunities
to learn from those various aspects.
Mm-hmm.
Like, it's everythinglike crime scene, it.
Your dispatchers almost anythingyou could think of is what
they're looking to target.
Mm-hmm.
All right.
And you said they weretheir conferences in Miami?
(19:06):
So they have a one dayevent coming up here.
Okay.
Next month in November in Miami.
And then they have a bigger conferencethat they're working on as well.
All
right
now, and then they're starting tomake their way to some of the events.
So they're at IACP andjust different stuff.
They're starting to get involvedwith the National Real Time
Crime Center Association.
So that was actually how I learnedabout them was from partnering the
crime center Association with them.
(19:26):
Yeah, no, that'll befascinating to read up.
On to see what their, their mission is.
While I like the diversity, it isalso kind of a lots of players means
that that can get pretty heavy.
Correct.
That you can't really have directtrainings necessarily then.
'cause it can't be targeted directlyat analysts when you have all these
(19:47):
other different positions in there.
Yeah, yeah.
But it's always.
Always a good networking opportunityand an opportunity to learn about
some other aspect of public safety.
'cause you never know when it'sgonna be important as an analyst.
Absolutely.
All right, Nikki, and hey, beforeI go, congratulations on being
a mother for the first time.
Yes, thank you.
(20:08):
That's why we had such a gapbetween spring and fall was I was
busy with my little guy, so yeah.
Now we're getting backinto the routine again.
All right.
And it was it wasn't avery simple pregnancy.
He was a little early, right?
Yeah.
He came two months early and spentjust shy of a month in the nicu.
So.
Definitely took us on an adventure.
Oh man.
Now it, it's, that might be TMIdid do, do premature babies running
(20:32):
either one of your families.
So you and your husband?
They do.
Actually his mother was a premature baby.
Yeah.
But ours was more directlyhe was an IVF baby.
We had been trying for years.
Oh, okay.
So from IVF, your chances ofpreeclampsia were increased, so Oh, okay.
Felt preeclampsia.
Yeah.
Well see, that's all kinds of new stuffthat I, that I learned, but, mm-hmm.
(20:52):
Obviously, I'm glad that the wholething worked out and that it was
certainly scary there for a while.
But how old is he now?
He
just turned six months on the ninth, so,
man.
Yeah.
, He's probably already a little mobile andhe's probably looking to be walking here
he is rolling like crazy andtrying to figure out how to crawl.
(21:15):
So yeah.
He has been, he was very mobileinside too, so he will, once he
gets moving, he will be moving.
Yeah, yeah.
No, that's that's certainly fun.
Quite an adventure ahead ofyou, so congratulations again.
Well, thank you.
All right, Nikki thankyou again for this time.
Always look forward to chattingwith you, , real crime all the time.
(21:36):
Until next time, take care.
Thank you for making it tothe end of another episode of
Analyst Talk with Jason Elder.
You can show your support by sharingthis and other episodes found
on our website@www.podcasts.com.
If you have a topic you would like us tocover or have a suggestion for our next
guest, please send us an email at elliotpodcasts@gmail.com next time analysts.
(21:59):
Keep talking.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.