Analyst Talk - Sean Bair 2.0 - AI, Analysis, and Officer Wellness - Leapodcasts

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 time.
How we doing Analyst, Jason Elder herewith a very special LEA podcast Deep Dive.

(00:23):
I wanna welcome back to the podcast, SeanBear the entrepreneur, to find out what
he's been up to since he was on the showand what advice he has for our listeners.
Sean Bear, how we doing?
I'm doing awesome.
Thanks for having me back and I'mexcited to chat with you over the
next little bit here and, and youintroduced me as an entrepreneur.

(00:45):
I'll take that badge, but I, Imiss you saying crime analyst.
So we'll maybe try and workthat in toward the end.
Yeah, yeah.
Well we can certainly talksome crime analysis as well.
So we, we got a lot of topics to talkabout, 'cause I want to pick your brain
about artificial intelligence, aboutanalysts role going forward with ai.

(01:10):
So there's, a lot.
To go over.
A lot has happened in the lastfive years since we last spoke, so
My goodness.
Yeah.
Wow.
Now you say that AI wasn't even athing commercially five years ago,
so we do have a lot to catch up on.
Yeah, no, , it seems science fictionuntil it wasn't science fiction.
Yeah.
Now we're trying to catch up.

(01:31):
All right, so co-founded a new companysworn AI to work with first responders on.
It's a well on the health platform.
So let's get into that, how you gotinto that and where it is today.
Okay.
So sworn ai, obviously the word swornshould be familiar to your listeners

(01:52):
three friends and law partners actuallywere doing some consulting work for the
Detroit Police Department, helping themwith, various administrative things.
And what they discovered in doing thatis that there was a, an immense need.
For somebody to finally pay attentionto the blight that is officer health,

(02:14):
both mental health, but also physicalhealth and, and emotional health
and, and all sort of aspects of that.
And they came across some alarmingnumbers that the average age or, or life
expectancy of a police officer is age 57.
Now I'm 55, so thatdoesn't compute for me.
Yeah.
And that they have five times the heartattacks that the national average is in.

(02:38):
It's an epidemic basically.
So it's a very stressful jobas in your listeners know.
Mm-hmm.
Not only for an officer, but everybodywho works in public safety, who they
might not have hands on cuffs, butthey read the reports and they're.
The dispatchers hear what's going on,and so everybody is impacted by it, by
the stress, by the, the intensity ofthat profession albeit such a noble one.

(03:00):
And so they started to see that thatneeded a solution they were doing some
things and, and trying to come up withways that they could help the department
and then public safety as a whole to justget a handle on it, to understand how
we can make them live longer and havehappier and healthier lives and lower

(03:21):
the divorce rate and all the things thatcome with it, lower the drinking rate.
And so they, they started toenvision that a piece of technology
was probably gonna be in play.
That it was, it was moving for that.
And so one of the co-founders ofsworn, interestingly enough, in
small world Adam, safer, Adam, safer.

(03:41):
Is the son of Howard Safer, the Saferhas got a great name in public safety.
Howard Safer was the former NYPD and,and New York Fire Department Commissioner
back in the day has since passed on.
And he was on the boardof directors of Lexus ns.
Sort of a roundabout way throughLexisNexis, I got to know Howard

(04:02):
Safer, but also during the acquisitionof Bear Analytics, I got to know
Adam Safer because he was contractedby LexisNexis to help do the
due diligence on Bear Analytics.
So Adam was the guy who was goingout there asking all of our customers
without me knowing about it.
Hey, what are these bare people?
Legit?
Is this come on, what's going on here?

(04:22):
And do net Promoter scoreand do all the things that.
That a Adam is, is brilliantat and capable of doing.
And so he reported back toLexiNexis and said, Hey, thumbs
up everybody's pretty happy withthem and you should move forward.
So, long story short, what happenedwas when they were now trying to
figure out this tech piece, theythought, well, I know this guy.

(04:44):
He's got some backgroundin law enforcement tech.
So Adam reached out to me and, and itwas at first just going to be a little,
can you help us out for a little bit?
And maybe help direct the techand sort of help us stand that up.
So a couple month gig and I'mlike, yeah, I can do that.
I don't think it'll be toomuch of a distraction for
the other things I'm doing.

(05:04):
And then the more we both, whenI say both, I mean the three
of them and me got into it.
And then we went into an IECP in San Diegoa couple years ago and I started seeing
all the same people I'd seen for the last20 plus years being in public safety.
It was funny because we had sort ofa follow on meeting after that and

(05:25):
we both almost simultaneously said,you really should go full time.
So I, I realized at that point, Hey,I still am passionate about this.
I want to be full on in let me,let me come on board full time.
And they were approaching me in thatsame call to, to ask if I would.
Come on full time.
So I joined and we're all equal partnerstrying to make this thing great.

(05:46):
And and here we are.
Alright, so
the AI piece of it, I'm on the website.
I'm looking, through the details of Yeah.
Sworn a ai and obviously there'sstudies and research that have been
done with officers over the years, butthen there's just the personal aspect

(06:08):
of, of it is like, yeah, what, whatit means to the individual officer.
So can you talk a little bit about like,the research versus AI and the combination
of that, that, that helps fuel this tool?
Sure.
And in doing so, I'll sort of describewhat it is we're trying to do, but I mean,
your listener's gonna start to see like,oh yeah, I, I get what you're doing there.

(06:29):
Oh, I've done that before.
There's, there's manyaspects as, as you know.
And your listeners know as well.
AI is just an umbrella term, artificialintelligence, and in all candor, Jason,
we've been doing this for a long time.
So if you go back to the early daysin crime analysis one of the very
first things that I did in attackworkstation was a neural network.

(06:52):
So being able to.
Reward and punish the, machine learningor, and the, the algorithm based on
the analysts expert weight and inputs.
So that's artificialintelligence just of itself.
That's, that's one component of it.
Neural networks.
And then of course, if you look at apolice report, and one of the things

(07:13):
we used to do is look and read thepolice read the police report and try
and extract out salient things from it.
So, because the, the RMS might nothave it or the officer didn't enter it,
you might have had to, as an analyst,as I had to do, read the report and
say, okay, what was it was a burglary?
How'd they get in?
What was the point of entry?

(07:35):
All right, well, how did theyget into the point of entry?
Well, they pride, they kicked, they,they did whatever they did or they
broke what was the point of entry?
It was a front door, it was a window.
And we would extract these elements out ofthe police report and create structured.
Either an SPSS or, or Excel or whateverit was, and then begin to mine that.

(07:57):
Using processes and, anddifferent statistical tools.
But that process of extractingout those elements is.
P, natural language processing.
And so Attack did thatback in the day too.
That was one of the things, each of thesethings is of course, trying to program
myself outta work or make my life easier.

(08:17):
But one of the earlier thingsthat we did was it could extract.
Based on regular expressions andnatural language processing context.
So we would look at a sentence in apolice report and see that the suspect
pride opened the window to gain entry.
Well, they would look at that sentenceand look for keywords like action
verbs, pride kicked, broke, whateverit was, and then look within context

(08:43):
bidirectionally of that to say.
Oh, there's the word window, orthere's the word skylight or door.
And thus be able to interpret thatthey must have kicked the front
door or pried the window or brokethe window, things like that.
So there's again, artificial intelligence.
Natural language processingis a component of that.

(09:03):
We were doing that too.
So everybody's beendoing that for a while.
We're just now it's more commercialized,these capabilities and one of which
are these large language models andbeing able to do things like a Chachi
PT buzz, where it's gobbled up theinternet basically, and you're able to
ask questions of the internet becauseof these really wicked cool vectors that

(09:26):
it's turned words into numbers and math.
And we get to do things likeask a question and it turns
my question into numbers.
And then it looks at the vastnessof the data that it has that has
also now been turned into numbersand it matches numbers, closeness,
and it says, well, you said this.
That's pretty close to this.

(09:47):
So here I'm gonna pull that out and I'mgonna use some natural language processing
to also make it pretty and make sense.
And here's that.
So one of the things thatsworn does is we've also
leveraged large language models.
So we've taken sort of the best ofthe best peer reviewed articles.
We've taken the, the newestresearch on health and wellness.

(10:08):
Again, when I say health and wellnesstoo, I mean the wholeness of it.
It's mental, it'sphysical, it's emotional.
It seems spiritual.
It's all the different things thatgo into making somebody a well
adjusted, happy, healthy person.
And what we've done is we've been ableto train the system to understand all of
that, that documentation going further,what we're able to do is say, okay

(10:33):
agency X, Y, Z, we have all this reallyrich content that'll give insights to
your officers and, and analysts andwhomever is gonna use the system and
how to be well, and what to do basedon your particular point in time or
where you're at in your health journey.
But we also enable the agencyto upload their own information.
So there's a second layer.
So the first layer is the sworn layer,which is all the, the sort of cleaned.

(10:57):
Health and wellness data anddocumentation that we have in our
system trained ready to rock and roll.
Then we have the middlelayer, which is the agency.
So the agency might have employeeassistant programs, they might
have wellness programs, they mighthave resources of, of phone numbers
and people and specialists andwhomever the, the department or

(11:18):
the employee could call in need of,of advice or crisis or counseling.
And then they can also uploadthings like policies and procedures
and, and different things that.
That are specific to the agency that wouldmatter to that individual, that employee.
And then finally, which I think is themost interesting thing, the individual

(11:38):
can upload their own informationto have the system train on it.
So there's three layers there.
So I now have the, the sworn layer, Ihave the agency layer, but then I've
also uploaded things of interest tome, particular diet plans that I want
to sort of follow workout plans that Iwanna make sure I'm adhering to basic

(11:58):
medical information that is gonna help itgive me more accurate information back.
So maybe something about my glucoselevels that need to be monitored so
that when it responds to me, it'staking into consideration the totality
of not only my interests and my.
Standing with my data, but the agency andas well as the sworn library database.

(12:23):
And so that's just one aspect of it.
So we do have a chat that they couldask questions of, but behind the scenes
what's happening is we've ingested otherdata that matters to that individual.
So if it's an officer, what we've doneis we've brought in CAD data, all right?
We brought in computer aided dispatchdata, and with that we've placed weights.

(12:44):
Here we go with the crime analysisstuff, we've placed weights on
those CAD records based on whatthe agency or the officer feels.
'cause it could be, it could be finetuned to the officer what they feel
is the impact of that call on them.
And of course, there's anelement of time in that as well.

(13:05):
So we get that.
I won't get too deep in the weedson the algorithms, but mm-hmm.
You, your, your audience is smart enoughto see where we're going with this.
So with that we havethe CAD data, so I know.
What the individual doesthroughout his or her day or shift.
Then we also have biometrics, sowe're sort of biometric agnostic.
I wear an aura ring, whichtracks sleep and fitness and

(13:29):
HRV and steps, and you name it.
It's connected to my, my Appleecosystem, which also provides
additional data and we connect to that.
So sworn is by volunteer basisconnecting to these biometric devices
and being able to incorporate thebiometrics along with their cad,

(13:50):
along with their own interest in thirdlibrary with the agency and was sworn.
And so what happens is behind the scenes,instead of it and us proactively asking,
or I should say reactively asking.
Chat, GBTE, why should I do this?
How should I do this?
Where should I do this?
What we're doing behind thescenes is we're finding moments

(14:12):
and opportunities to assist thatindividual with their health based on
what we've just seen with the data.
So there is again, some, some, it'snot necessarily, I would call it ai.
There's a lot of math and algorithmsthat goes into it with some of that.
But then there's also the AI aspectwhere it's passing the vectors to the

(14:36):
database to retrieve back the closestresponses to be able to use other AI
to cleanse and to focus the responseback to that individual and to give it
to the him or her at the right time.
I'm gonna take a breath here 'causeI've been talking a lot, but that, in
terms of an ai, you asked about it.
There's a lot of different aspects thatwe're touching when sworn says we're

(15:00):
sworn.ai, we're all in on it, and ittouches every aspect of our platform.
Yeah.
And I'm curious to know whatthe reception has been like
because the way you described it.
The way the O officer can volunteer thebiometrics, add additional information
it makes, I instantly think of, well,there's certain people that if they're

(15:25):
already healthy, maybe they're alreadytracking that and they might be
willing to go supply that information.
So you're getting people that are alreadyhealth conscious, putting information into
the system as opposed to somebody that'snot into that, maybe not into health.
Just starting out may not be as openor willing to supply that information.

(15:47):
So I'm just interested to see whatthe officer reaction has been to this.
It's been surprisingly receptive mm-hmm.
In that we are not getting pushback fromofficers saying this is big brother,
I don't want you having my stuff.
You're gonna just sellit, so on and so forth.
One of our promises to our, our customersare that their data is anonymous.

(16:10):
So the department will never seewho in the department has, you're
not sleeping good or their, their,their health is not being maintained,
or they went on or, or had some.
Real problems with certain calls.
We're not sharing that.
The department sees aggregate, theysee that, okay, overall the health

(16:31):
of our department is increased, it'sdecreased, it's remained stable.
There's, there's that sort of feedback.
But this is a system that issolely built to help the officer.
And the most important thingthat we could do is assure them
that only they see their data.
And so it's, it's all, whenI say voluntary, I mean it's
voluntary in the sense that theygive their biometrics and stuff.

(16:52):
Of course, the CAD data mm-hmm.
Is the agency can see that already.
But even the agency doesn't see our, ourweighting of that and our correlations
against the other variables we're doingwith that, that's all behind the scenes.
And the data is in fact pushed tothe officer not through an app.
It's pushed through the text.
So they get a text when there's somethingto pay attention to or to be aware of,

(17:15):
or something to help them out with.
They might have just experienceda, a, a series of, of events, let's
say, as I used to be back in theday sexual assault investigator.
I would go out to these, these sortof emergent scenes and my sergeant
would never know that I had hadthree of those in three weeks.

(17:36):
They're just not paying attention to that.
They're not seeing the longitudinalsort of data points there.
They know that I just went on one.
And who knows, they might besmart enough to, with 12 officers
realize, oh, I think bears actuallydone a couple of these lately.
Mm-hmm.
Or even other significant calls, maybenot those, but maybe like a drowning child

(17:57):
call or a 9 63 or a accident fatality.
Which can be really traumatizing.
Mm-hmm.
And so if we'd had a couple of thoseacross our shifts, or even just the death
a death investigation, just a dead body,those can also be somewhat traumatic.
That if those had sort of been throughoutmy last two weeks, I, as an officer,

(18:19):
frankly, never paid attention to that.
'cause every day was a new day for me.
And at the end of the dayI was just trying to forget
the previous shift in day.
So, but what we can do is we can pingthem to say, look you've kind of had some.
Some rough calls lately, and wehappen to have noticed that your
sleep has been declining andthat your, your heart rate vari
variability has been a little bit off.

(18:40):
You might wanna consider X, Y, and Zor here's some department resources
that are available to you that youmight be able to take advantage of.
And we'll even go so far as totext them the links, the numbers.
So it's, it's that sort of thingthat has made it to where I think
they see the intent of the system.
They know that again, it's thedata's in our system, so it's not

(19:02):
sitting in the agencies system.
So we have control of it.
And they've been very open with the data.
So all good so far.
All right.
So and then how many agenciesare currently using the system?
Or do you go by a sworn officer?
How, what, what's your metric there?
Yeah.
Hundreds of sworn officers,but we're just getting going.
Mm-hmm.
I mean, we've literally just launched thisyear, so we've got about a dozen agencies

(19:25):
so far, but hopefully when you talk tome next in two or three years, we'll be
up at that a thousand plus or something.
But yeah, we're just getting going and,and as most startups, and also with just
the speed at which I'll even say we usea lot of AI in our, in our development.
So it's of course in the.

(19:45):
Platform itself, but we also leverageit in our, in our day-to-day workloads.
And it's made it much easier for usto be able to produce much faster.
So back in the day, I'd be banging onthe keyboard in my basement on Attack
workstation, trying to get Jason Elderthe newest, you know thing that's gonna
help 'em do crime analysis faster.

(20:06):
Now we're able to even get that out,those sorts of functions and features
and, and capabilities out a lot faster.
And so we're, we're sort of buildingthe plane as we're taking off, but
it's, it, there's some really coolthings that we're gonna be able to
do.
Yeah.
Well, what was your saying?
If you wait until you love yourfirst draft, you waited too long.
Yeah, that's not my saying.

(20:27):
I think that's oh, he was a Netflix guy.
Shoot, I forget his name.
I think it was, it was Neil something.
But anyway, yeah, he said if, if you,if you don't hate your first version,
you've waited too long to release.
Yeah.
And it's true.
And I hate our first version,so that's a good thing, I guess.
Yeah.
Well , this has a lotof potential obviously.

(20:47):
I mean, you, you may be saving livesin the long run lowering costs and
just giving a, bringing awareness.
To a an epidemic as you mentioned.
Yeah.
So kudos to you and, and your team.
Thank you.
And I think we have people who see thatit has potential because we've got some,
some substantial funding to get goingand to get this off the ground and some

(21:11):
really good backers and supporters.
So yeah, I think people, the, the,the country's starting to come
around to supporting and lovinglaw enforcement as they should.
So I'm glad it's kind ofcome back to that now.
Alright, nice.
It was rough there for a couple years.
Yeah.
All right.
Let's, let's move on then.
'cause you also are writing a book on ai.

(21:32):
Yes.
AI and policing is atleast the current title.
As you're getting ready to put thefinishing touches on this book, so
this is again, just how did this bookcome to be and what should we expect
It came out of.
I think the need to demystify a lot of it.
Mm-hmm.
Because I, I think that those have,who have been sort of doing AI or at

(21:55):
least the components of it, they'veseen that, okay, this is not exactly the
newest, this isn't some new invention.
This has been around now.
Compute power has gottenus to where we are.
That's what's really been the the, the,the hockey stick for AI is our ability
to now have the computing power todo what we'd always envisioned to do.

(22:17):
Mm-hmm.
We just couldn't do it because.
It, the, the technology existed froma ma from a mathematical perspective,
from a structural perspective from,from a procedural perspective.
But we didn't have the, the bandwidthto actually manage all of this, and
now we do with with the technology.
So I think what the purpose of the bookis, is to, as I say, dis demystify a

(22:39):
lot of this, to say, look, you've beendoing this we've been doing license plate
recognition, we've been doing facialrecognition, we've been doing things
in Attack Workstation and Arc, GIS andothers that, that have components or.
Capabilities that are, that fallunder the umbrella of artificial
intelligence and to help agenciesunderstand how it can benefit their

(23:01):
agency and where they don't need it.
So it's also a, a, a, a purpose tosay, Hey look, you don't just have to
go get it 'cause it's now availableand everybody's talking about it.
Mm-hmm.
But here is where in fact it mightbe able to help your department
out, whether you're an analyst or adispatcher, or an officer, a chief.
I sort of try and do the bestI can to touch on all of it.

(23:24):
And talk about what the different areasare in AI and which ones are, are really
proven to be useful, and which ones aremaybe still a little not fully baked.
Mm-hmm.
You talked about the computing power.
I sat through a lecture last year on, onai, and the, the presenter described it

(23:47):
as you it's a little bit of a black box orthe dark side of the moon type of thing.
And like if I was to try to calculate,go through and double check all
the calculations and everything, itmight take me like 70 years Yeah.
Exactly.
To go through, to go through this.
And, and so I think that's one ofthe things that brings, especially

(24:10):
in law enforcement, some uneasinessbecause, well, wait a minute.
How do you know the data's rightthat you're getting or how Yeah, how
can we know for sure or how can I becomfortable moving forward with the
information that I'm getting outta this?
If somebody.
Can't double check andprove all the calculations
a hundred percent.
I, I never met too many people in publicsafety who were able to do spherical

(24:34):
calculations of distance, and yet theystill trust arc GIS distance feature.
So did they get thecurvature of the earth?
Right.
So we trust our, our systems and ofcourse you can you can truth them.
There's a lot of that.
So instead of us goingout and tr like trying to.
Try and figure out how chat GPT works.

(24:54):
You yourself can actually do that.
You can take your ownlittle large language model.
You can fill it with some content.
You can begin to see askingquestions, how it turns those
questions into, to vectors numbers.
You can then see visually in threedimensions how the content that you've
loaded in is relatable based on.

(25:17):
The data where the word cat exists,where the word dog exists, where the word
collar exists in relation to both catand dog or how far it is away from pig.
And so you can do these sorts ofthings and see, oh, I, I get it.
I kind of see what it's doing.
But yeah, to in fact, thengo do the vector math.

(25:37):
You're not gonna do that.
But I think you can geta basic understanding.
Nobody, not nobody on this.
I can almost guarantee that nobody onyour podcast today, and this is not
just because of the podcast listeners,this is the state of the world knows
really how an internal combustionengine works anymore and could describe.
You know how the the, the sort ofthe brain, the controller of our cars

(26:01):
makes the, the ignition go and fireat the rate it does and when it does
it and how it balances air and fuel.
They just trust that it's gonna work.
And so yeah.
AI's a little bit of the same stuff too.
Yeah, I,
I think it's also comes downto what you're asking it to do.
I know when I use chat, GPTI, Imight be asking about, help me with

(26:23):
a SQL statement or help me with anExcel function it's usually something
that I'm having problems with.
I'm trying to figure outhow, to manipulate the data.
Yeah.
So whatever it tells me,I'm then testing right away.
I'm like, okay, well what they'reasking me that code or that procedure,
is that, does that, does that work?
Oh, yeah, it does.
Then I'm good to go.

(26:44):
So I'm doing, I'm actually testingwhat it's telling me right away because
it's, I'm asking for suggestions.
Yeah.
Y there's the funny thing about.
Chat, BT and Claw or grok or whateveryour flavor that you like of, of
ai, and they all have their, theirsort of strengths and weaknesses.
I think what a lot of people don't realizethis might be one of the things that

(27:06):
I can contribute today on this call isthey might upload a, a long document.
They're like, I don't wanna read thisstupid book, or I don't wanna do this.
Just tell me what it's about.
And they upload it and they thinkthat the entire book has been
ingested or it's ready to go.
They start asking questions and so theydon't understand the whole token limits.

(27:27):
And that's where it starts to getinteresting because it'll still work.
It doesn't tell you, Hey, look, becauseof the token limits, I wasn't able to
ingest actually this entire document.
All these call for service, allthese police reports it just stopped.
And so when you ask a question, itmay not have the, the amount of sort

(27:48):
of memory and tokens to spend to giveyou the, the most accurate answer.
So that's where it gets interesting.
You kind of do have to know alittle bit about that aspect of it.
But I will tell you the mostfascinating thing, as you mentioned,
like using it as an analyst or asan officer that it's in the book.
And I'll, I'll just, I'll tell you itso you can skip this chapter when you

(28:08):
get to it, that one of the things thatI had done was testing, in testing the
various platforms that are out thereand how they might have applications
and utility and public safety.
One of the, the things that Iactually wrote about in, in a previous
book was this officer Kearney.
And he had noticed being a fellowsouthpaw like myself, a left-hander,

(28:31):
he had noticed a burglary sceneand the window that was the window,
the screen for the window was cut.
He had noticed a C, so when the individualcut the screen, they did it in the
shape of a C, the letter C. Mm-hmm.
And he thought, he stared at it for aminute, he thought, huh, I don't think
a right hander would've cut like that.

(28:53):
That screen cut in the Cform would be natural for me.
And you probably right now.
Mm-hmm.
Jason and others who are listeningthinking, okay, I'm a righty.
Yeah.
I would go from top to theright and then back around.
You would do a backwardC is what you would do.
If you were doing that.
And so he, he surmised that,wow, this must be a left-hander.

(29:13):
And sure enough, they, they eventuallycaught him and the guy was left-handed.
But to be able to reduce yourinvestigative pool from a hundred percent
of the population to, well, only sevento 10% of the population was left-handed.
So that dramatically reducesyour investigative pool.
So that, just, that in itselfwas, I think, a fascinating

(29:35):
idea of just investigativeKung fu Sherlock Holmes ness.
Mm-hmm.
But here's where it got interesting.
I, I took a picture of a screen cut inthe, the shape of, I just described the
c. I gave it to chat PT with his visioncapabilities and cut to , the chase here.
It discovered, or it told me thismost likely is a left-handed burglar,

(30:00):
and I almost fell outta my chair.
I'm like, wait a minute.
How did you see the c?
And make that mental leap thatthis was a south paw that did
this, and I couldn't believe it.
And I thought, well, no.
It's gobbled the internet.
It knows everything about everything.

(30:20):
So somewhere it might have put into, its.
It's, it's, it's it's math.
It's the vectors.
This this passage out of tactical crimeanalysis, research to investigation.
The previous book, it mighthave put something in there and
it was pulling it from that.
And then the more I prompted itmore, I realized there are whole,

(30:40):
of course, psychological studies andbooks and materials that exist about
handedness and how people write and theslanted ness and so on and so forth.
Because then when I prompted it tosay like, where did you get that from?
That's bizarre.
It didn't say, oh, you wrote aboutthis in chapter 12 of this other book.
It actually said, well, the researchshows that handed disc slants and so, and.

(31:05):
It, it, it inferred that.
And I, at that point I'm like,all right, this actually could be
really helpful to public safety.
And there's some other examples thatI've, I've written about in the book
where I've tested it in that way.
But think about that, Jason, thatyou could literally upload the
photos from a crime scene mm-hmm.
And have it potentially providesome insight that you might not have

(31:26):
thought about and maybe unlock somenew direction in the investigation.
That's pretty powerful stuff.
Of course, there'll be people on thecall who say, wait a minute, you're
uploading crime scene photos to chat GBD.
Yeah.
But you can also, I don't know ifpeople are aware of this, but you can
also get your own chat, GBT, , theselanguage models that exist, that have

(31:49):
trillions of tokens in them, they'regigantic that have a lot of data.
You could in fact download those toyour own desktop computer nowadays.
You can get a desktop and even a, an,I think it's an M five chip from apple.
You can get machines that can haveyour own in-house chat, GPT with the
same level of, of data that it's using.

(32:12):
Yeah, I think a lot ofcompanies are doing that.
Yep.
Yeah.
Yep.
Yeah,
so it's not that hard.
I mean, it's not that big a deal.
It's a couple terabytes, but so what, Imean, most people have a couple terabytes
nowadays on their, on their computer.
Yeah.
Yeah.
Floppy disc
back in the bear days.
Here's your yeah, here's your bag.
Yeah.
I'm trying to think of what size that was.

(32:33):
It wasn't the floppy, it wasthe one right after the floppy.
It was the little, likethree by five or something.
Yeah, yeah.
Yeah.
I think that was like, I, Ithink I had one that was two
megs and it was, yeah, those were
the days.
So yeah.
It's, it's cool stuff.
Yeah.
So no, and I think, I thinkyou hit the nail on the head in
terms of some of the dangers.
If, if, if the sourcing of theinformation and , if it doesn't

(32:55):
have all the information.
Right.
Because as all is, if youdon't have all the information.
To do any kind of analysis or forecastingor anything else you lose that capability.
Yeah, for accuracy.
So I, I think it's, again, it's thecombination of what are you asking it to
do and what's the source of information.
'Cause I've always, obviously, if I askit what was the crime trends in Utah?

(33:19):
Yeah.
It's obviously gonna go out thereand use a source and if it uses a
source that defines crime differentlyfrom what I'm expecting, I'm not
gonna get the answer that I want.
Yeah.
And there's ways around that.
I mean certain platforms likeperplexity, they do a good job at
actually providing you the citation.
So that, that's the sort of.
The, when I use it for those purposeswhere I want to validate the response

(33:43):
I'm usually asking for the citationso I could go, I could go validate it.
That's really what I'm doing.
Yeah.
It's, it's just another tool.
Honestly, it's no different.
Back in the day when when I was first ananalyst or even an intern and we had the
idea of putting X's and O's in the, whatwas called HP desk 'cause email hadn't

(34:03):
even been ex invented at that point.
But we had an internal email system.
We were making pin mapswith Aski characters.
We had drawn the city ofTempe and Askie characters.
And then used zeros and xs to putcrime points so that an officer or a
person seeing this, this early emailcould visualize where those were at.

(34:28):
It's no different from taking andgoing from that to Arc GIS and
then Google Earth and then allthe other cool things we have now.
This is just another toolfor analysts to be even more
powerful than they already are.
Hi everyone.
This is Data scientist Andrew Wheeler.

(34:49):
, For my PSA is if, if youever think to yourself.
Can't you use AI to do that?
Just stop right there anddon't say that sentence.
I know that that's more for themanagers than the crime analyst
out there, but every single timesomebody has come to me with, well,
can't you use AI to solve that?
It's never, it's never a goodquestion to actually use any

(35:12):
type of, even not AI solutions.
So you need to think aboutwhat you actually want before
you utter that sentence.
Hi, I'm Kyle Stoker and I'm encouragingyou to vote in the I-A-C-A-A elections.
, You have the opportunity to vote foryour candidate, so make sure you go
to the IACA website and vote becauseour membership has a voice in who

(35:32):
leads the organization, and you wannamake sure that your voice is heard.
Thank you very much.
Well I do want to talk a little bitabout analysis and ai, but before
I do, you are also a professor andyou are adjunct professor with.
And so you've been teaching nowfor a little bit and it got me

(35:54):
thinking as a professor, like whatis your take on students and using
AI in terms of writing assignments.
Yeah.
I'll, I'll answer that the same way thatI just sort of concluded the last section
there, which is, it's a force multiplier.
So when a GIS came out it was a toolthat I could become more efficient,

(36:20):
more effective, I could do deeperanalysis, faster glean newer
insights based on this capability.
And nobody ever accused me of, well,when the good old days, when I used
to push the pin in the map on the wallI knew what was going on and that,
that's how I should, you should do it.
No, , this is just as an analystwould leverage this, this technology.

(36:42):
To be more efficient and effective.
So too should the student.
And so I've been teachingnow in various capacities.
I've been doing it for about 20 years.
I, I've taught classes up at BritishColumbia Institute of Technology, one
of my favorite universities to teachfor amazing forensic science program.
Up there, I got opportunitiesto teach about crime analysis.

(37:02):
There to unbelievably wicked smartprofessionals and done some other things.
But, but Brigham, Brigham YoungUniversity, BYU kus is nice.
A special place.
And they have really smart students there.
And so, like I said, five years, sofive years ago you, they didn't have
access to these, these technologieslike we do right now at our fingertips.

(37:26):
You could code some ofthem, but who's doing that?
Mm-hmm.
And so not, not really a problem.
Or even, of course we mentioned, butfast forward three years ago when chat
BT just first became commercializedand sort of the world could ask it
questions, academia kind of freaked out.
It was like, wait a minute here, thisis you have to push the pen in the map.

(37:49):
If I had to push the pen in the map, thisis how you gotta, you gotta pay your dues.
And so that's not my perspective.
My perspective is they betteruse AI in my classroom.
So I'm looking for opportunities to,and, and, not, not only just introduce,
but in every single lesson I have,I'm trying to get them to see how that

(38:10):
might be applicable or not applicableto what it is they're trying to do.
And I will tell you the lessons thatI taught five years ago, I would teach
'em how to do a net promoter score.
I would walk them through Porter'sfive forces or different strategic
frameworks that business people use tomake sense of the, the, the business
environment, the industry, their company.

(38:31):
Many of those frameworks that we wouldshow them, look, here's how you do this.
You gotta calculate this and do that, andyou gotta grab this and that's gonna come
from here and this is how you do all this.
It can now be done with ai.
So it's, it would be a disservicefor me as a professor trying to set
them up for success in business, tonot show them now how it's done with

(38:53):
the technology and just because itcould be done and I have to revise my.
My curriculum because they can literallyjust drop it in there and have it do.
It is a great thing because , to talkabout academia here and the Bloom's
taxonomy, the hierarchy, the pyramidof where we wanna get them, we wanna
get them to creativity, to synthesis ofof, of sort of ideas and, and insights.

(39:19):
We don't want to have them stay atthe base of the Bloom's Taxonomy,
which is just rote regurgitationof facts and memorization.
We wanna get them analyzing andsynthesizing and creating, and
if you can eliminate a lot of.
The lower tiers in the taxonomy,because now the technology has allowed

(39:40):
us to move through those faster.
I don't have to memorize KaiSquare calculation like I
did when I was an analyst.
I think we talked about this on oneof the podcasts, that this is sort
of a snapshot of where we've come.
When I was an intern trying to becomean analyst and I was applying for the
crime analyst job in Tempe, Arizona,the Tempe Police Department, the test

(40:01):
was, you had to do all these things,one of which was do Kai Square by hand,
that that's not all that easy to do.
You gotta, you gottaknow what you're doing.
Right?
And little old justice studiesdegree, I wasn't a math guy,
so I had to figure that out.
And I learned from Rachel Boba,all right, how do we do this again?
'cause she was a stats PhD. Sowe, she walked me through it and I

(40:23):
understood it and the whole nine yards.
But yeah, so the concept of how it's done.
But is there an analyst in the worldright now who's doing Kai Square by hand?
I hope not.
Yeah.
So those sorts of things are all donethrough these advancements, and so what
I'm trying to do is they, they advancequicker through the business curriculum

(40:44):
to where we can now start really gettingdeep into the analytical, the, the.
The, again, creative, these ideation areaswhere we let the technology do the, the,
the work and the, the, the brunt of it.
But what we can do is use theresults to, to innovate and create.

(41:05):
So that's what it's making.
It's making the classroom, well, first ofall, it's actually making it difficult as
a professor, because truthfully, the lastthing a professor wants to do is every
single semester change their curriculum.
Right.
Change their syllabus.
Because once you get it down, it takes,as 'cause you've talked too, it takes
a lot of work to develop a class andto have it work in 16 weeks and what

(41:28):
they're gonna learn and how they're gonnabe the, the content mastery, all that.
What's interesting is that becauseAI is advancing so quickly,
it is absolutely exponential.
At this point.
We don't know what's next and everysingle semester for the last three years,
I have had to basically redo my class.
To accommodate the newest ways thatthey would learn and, and be taught.

(41:52):
So, yeah, kind of wild.
But when, so here's the thing.
For those who don't know ai, theycan easily be caught out if they're
using it and not learning with it.
So if, if part of it is, look, I wantyou to write a personal strategic
assessment to see where you're at inthe world, what resources you have
available to you, what capabilities youpossess, or knowledge, skills, ability.

(42:15):
And with that, I want you to.
See if there's any opportunities forgrowth, any blind spots you might have,
so that as you augment your so your,your yourself to make yourself more
capable and thus desirable by a companyin hiring, I need you to write that.
Well, that's, that's kind of personal.
And if they were to try and chatCBT, that that's not probably gonna

(42:38):
come out right, because the professorwill be able to see right through
that, that, well, this isn't you.
How did you that's notwhat your skills are.
That's not what you're trying to do.
This isn't, this wasn't thisreally reflective of where
you're at in your understanding.
So I'm trying to, and I think otherprofessors are too, trying to get
it to where you're asking a lot morequestions that might involve their,

(43:01):
their sort of state of where they'reat again, with their knowledge
or their skillset or their, theirtheir activities that they're doing.
That's where you can begin to get at.
The, the call out whenit's being used improperly.
And a lot of people don't knowtoo, that when you just copy and
paste from chat GBT, there's hiddencharacters that are in the text.

(43:22):
Nobody really knows that,but you can use tools to re
recover those hidden characters.
One of the most obvious ones that justbaffles me that people don't remove
this, and I'm sure you're familiarwith this, but it's called the M dash.
It's EM and then Dash and Chacha loves it.
Claude's a big fan of it as well.

(43:43):
So if you just use either of thoseto write for you for whatever reason,
it just loves to be able to putthat double dash in there for you.
And most people are nottaught to write that way.
So anybody who has that, those,and they're writing immediately,
I'm already suspect or suspiciousof, did you really write this?
'cause it's even difficultto get the M dash in in.

(44:06):
Microsoft Word or Google Docs, yougotta kind of know what you're doing.
So that in itself is a deck giveaway, butyeah, there's special hidden characters.
Yeah, I, I did notice itdoes, it is Dash heavy.
I didn't know what particular wascalled, but I have noticed that.
So I guess from, from your perspective,if a student turned in a paper and state,

(44:28):
maybe stated at the end of it, I wrotea draft and this was edited by chat, GPT
mm-hmm.
Would, would that be acceptable?
A
hundred percent for me.
Yeah, 100%.
I hope they use it.
Why wouldn't you use it as, do you useSpellchecker when you write in Yeah.
In Google, right?
I mean, it's, you use the tool toget , the best product you can get.

(44:49):
Mm-hmm.
The, the trick is that youwant to make sure they're.
They're actually learning.
And true mastery of, of a subject orlearning is when you do begin to get
to that creative phase and you getto the synthesis phase, one of the
you know him very well and one ofmy, my favorite people is Dan Helms.
Dan was very smart about that kindof stuff, and you would see him

(45:12):
be able to do this quite a bit.
And one of the things everybodyknows just from Dan Helm's lore,
is that he discovered this becausehe was looking for what, what smart
people do is they find parallels inother industries and disciplines.
So if you're just myopic withyour discipline, how are you gonna
how are you gonna grow beyond it?
So what you do is, like my example ofthe left hander chat, GBT was able to

(45:37):
see outside the discipline of, of, ofcrime analysis, the, the discipline of
investigations and use other researchand knowledge from psychological.
Writings and findings to see that theslant in things that are produced by
one's hand give insight into handedness.

(45:58):
So what Dan was trying to do when he wastrying to use ARC view back in the day to.
Come up with different ways to make arc,do arc view do better at prediction.
He's like, well, whoelse is doing prediction?
And he thought long and hard aboutit and he started doing some research
and he found that the, the biologistsbiology discipline was very good at

(46:20):
using arc view to study animal movements.
So they, as 'cause we were puttingthat thing all over the place back
in the day at Cmap, the arc viewanimal movement extension Yep.
Was a great extension and all it wasis math, but it's looking to other
areas and finding parallels andcorollaries that can help us sort of.

(46:43):
Do those things that, that make us better.
And I guess that's what I'll sayis the hope is that the classroom
is even more robust as a result ofchat gt not dummied down because
the work's being done for them.
They're actually thinking deeper andreasoning more and becoming more creative.
And I ideating and innovating, it's, it'sbecome really good almost with some of

(47:06):
that to, because again, case in point,when I asked it about the screen cut, I
would've never have, I probably shouldhave thought it, but I didn't at the time.
I wasn't thinking the psychological sortof discipline of having that knowledge
base that could have provided me insights.
That was just one of those thingswhere, oh, yeah, officer Kearney
told me that, and I get that.

(47:27):
That makes sense.
Yeah.
You know.
Handedness.
So yeah, that's, that's wherewe're at in academia right now.
And I'm having to redo forthis coming fall semester too.
Keeps you on your toes, right?
Yep.
So I think there's a good segue intoanalysis and AI because I feel, I

(47:49):
always find it funny when you, whensince I can, as long as I can remember,
and even way back in study in thepast of law enforcement analysis,
it's always been this idea that thecomputer's going to replace the analyst.
And AI is not really any differentin that respect, but it's certainly
something that can go hand inhand and not be one or the other.

(48:10):
But I do feel for the analyst,it can take away certain remedial
tasks that may weigh you down andget you away from the analysis.
So it's like I was talking aboutwriting the code, like I'm not spending
time writing that perfect code.
It can help me streamline that,get the code so then I can get the

(48:32):
data to actually do the analysis.
A hundred percent.
I mean, it's like I say earlieron, it's a false force multiplier.
It should be, they should be leveragingit in, in all of their daily tasks.
Even going so far as to help themsort through their email or write a
better memo to the chief's office.

(48:54):
Have it review their workto see that it's accurate.
There's, there's times that I'vemany of the chapters in the book,
I take 'em and I, and I drop'em to Claude and I say, Hey.
Here's the audience, here's what Iwant you to be able to do, and is
there anything that it makes sense?
Or where have I not consistentlyadhered to my theme?
It's, it's been my partner in writing.

(49:15):
It's been my partner increating content for students.
It's been my, my businesscompanion to bounce ideas off of.
Because again, it has thevastness of the internet there.
When I say quote unquote bunny ear,internet it has the vastness of all that
to be able to pull from and to draw fromand regurgitate it back to me so that,

(49:37):
that I can then take that and be like,oh yeah, I didn't think about that.
And it gives me different ways to, toapproach an idea or to go down a path.
So.
I think analysts should embrace itand love it and, and just like they
did agi IS just like they did SaaS orSPSS, it's just gonna be an awesome
tool for them to, to be even better.

(49:57):
Yeah.
It's not gonna replace
them.
It's not gonna replace 'em.
I do feel though that they,it'll do stuff that may, the
analysts may really enjoy doing.
Maybe they enjoy doing the graphicdesign or doing the PowerPoint
or what it, whatever the taskmay be to finish the project.
And this is something that you couldvery easily, oh, I don't have to spend

(50:18):
so much time on graphic design or, layingout the perfect PowerPoint or whatnot.
Like, this is, a lot of thisstuff is, can be done for me.
Yeah.
Although I would say that it's been prettygood . When it's come to creative things
visual things like helping to structure aPowerPoint, and then I do the refinements.

(50:38):
Mm-hmm.
Or it give me the first draft of it,which is usually the hardest thing to
do is just get a your first square onthe piece of paper and then me say,
okay, look, I like this color palette.
Here's a hex code to use finecomplimentary colors here I want this.
And you begin to, you, youalmost become an orchestrator
versus a player in the orchestra.

(50:59):
You're not just strumming the oneinstrument now as a, as a cellist or
a bassoonist or whatever the differentpositions are in an orchestra.
You are the conductor.
You have all of these now, and youare orchestrating it to make beautiful
music is what you're trying to do.
Yeah.
And I, I do, we talkedabout the black box earlier.

(51:22):
Mm-hmm.
I, I also feel that thereis so much data now.
That it is, it's almost impossible for ananalyst to keep up with all the different
data sources and all the different waysthat they can search through data and
try to have a full picture or have allthe information that it's, this is, it

(51:45):
is going to become necessary becausethere is so much data out there that
they can't possibly keep one person, oneperson can't keep up with all the data.
There's no way.
I mean, even the most, you knowsmallish agency, if you start taking
into consideration cat recordsmanagement offense data, you maybe
have additional follow on investigativedata you can get access to.

(52:08):
You got workload data, you'vegot, I mean, if you even think
about, if you wanted to really doa good job at a, at administrative
crime analysis, I'm gonna do good.
Be re redistricting and,and resource allocation.
Well, we would do that withdata that we could manage.
And so we'd probably stick with callfor service and maybe one or two other

(52:30):
variables that we could bring in.
But if you can think about it, if wecould bring in other variables and
allow the computer and the AI to dowhat it can do, we could do better work.
So I could then bring in maybe the,the, the GPS routing from the patrol
cars and idle time and speeds anddistance and, and calculations to

(52:54):
the, from dispatch to arrival routesand all that kind of good stuff.
You could really do some coolthings that might make you even more
accurate with those, those outcomes.
So the hope is that, yeah, you, you just.
Begin to discover other data, or I shouldsay you begin to use it to incorporate

(53:16):
other data, but I will say you couldeven get it to help you discover and
make data based on combining sources and,and, and inputs to give me something new.
Right.
So that could besomething very beneficial.
Yeah.
We talked about earlier aboutbusinesses bringing in their own ai.

(53:40):
Do you know if any policedepartments are doing that?
I do actually.
There are a few in our, in ourour sort of efforts that we've
discovered that they have donetheir own internal large language
model with their own internal chat.
If you think about that if backin, back in my day, back in my
day I used to have with me in my.

(54:00):
My, my beat bag, the, the bag that Iwould take out to my patrol car and it
would sit in my front seat and I wouldobviously be driving, driving it around.
That bag was filled with stuff to makeme be a better cop and have resources.
Two of those items were two,three inch binders, you know?
Mm-hmm.
Those, those binders.
Both those binders, one being the policiesand the other ones being the procedures.

(54:24):
Well, now I don't haveto carry that around.
Back in the day, I would've had tohave thumbed through that to find
out what our policy was for X, Y,and Z or impounding this, or hung
handcuffing, this type of situation.
And so now that's justat their fingertips.
So yes, they're using it to be ableto enable the department officers,

(54:44):
analysts, dispatchers to getinformation quicker with more accuracy.
And, and so that's some ofthe utility of it right now.
More, more retrievalto be honest with you.
Yeah.
Yeah.
It'll be, it'll be interesting.
There might not be anymore silos in our data.
It might not be Yeah.
That,
yeah.

(55:04):
It is wild.
All right.
, I wanna finish up here 'causeit's it is been 10 years
since you sold Bear Analytics.
I, I didn't believe it when I wasprepping today, I was like, when
did, when did that sale go through?
I would've guessed thatit was like six years ago.
I had no idea that it was 10.
It is obviously a differentperspective that you may have now.

(55:26):
So what, looking back at it,what, what are, what are some
thoughts that come to mind?
Yeah.
10 years.
I, I'm with you on that.
It, it does seem like a blink of an eye,but as we get older time shortens it, it
quickens, it was April 2nd, 10 years ago.
So in April 2nd, I just think a funnyday to sell on, because they wanted

(55:46):
to do on the first of the month.
And I said, there's no way I'm sellingmy company on April Fool's Day.
So that's why it's always April 2nd.
So it was April 2nd thatwe sold 10 years ago.
And, and they've done amazing thingswith the, with what we had started.
Mm-hmm.
They've got it tothousands of agencies now.
, My only bummer is that they, theyhave since retired attack workstation.

(56:11):
Mm-hmm.
They just too difficult to maintainthat and the code base million
lines of code and all that.
It was, it got to be quite a bit.
And it just sometimes you gotta go withthe products that are, are in demand.
So that's my only sort of.
Bummer about it.
Having said that, I will tell youthat I use Attack Workstation myself
still for analytics certain things.

(56:32):
In fact, I even had it up last week.
And it's funny because Mark Dobbs, whoworks with Sworn now, mark Dobbs used
to be a bearly as, as and then mm-hmm.
Went on to Walmart, was directorof analytics at Walmart.
He's back now with back with the crew andwe're working together and he's at sworn,
he's the director of analytics here.
He sent me a screenshot saying, yeah,we want to be able to do this in sworn,

(56:54):
and it was a picture out of attackand I thought, oh, that's kind of,
that's good, good job on that one.
But yeah, 10 years the professionis, hasn't really changed.
To be completely blunt.
It's still got the same.
Amazing people in it.
It still has vendors that I know andI built strong relationships with.

(57:14):
There are still people who, when Ileft, are still in patrol or, or now
supervisors or still at the department.
And it's, it's, it's cool.
I, I love being back in trying to providethem technologies that help them do
their jobs better, but also now mighthave an impact on them personally.

(57:35):
And trying to help mybrothers and sisters in blue.
It's also neat too because as we beginto reintroduce ourselves, or I get, I
guess, reacquainted with public safety,which really hadn't, like I said, been
that different when I got back in.
The, one of the people that is a partof Sworn is Andrew McCormick, and
Andrew was an officer in, in Tempe,Arizona with me back in the day.

(57:59):
And he just retired after 27 years.
Got onto SWAT and, and has like reallyinteresting just a, a stellar career.
And so he's now working alongside uswith, in sworn, one of the neat things
is though, that he and I both startedthe academy in the same academy class.
So here we were 27, 28 years ago,starting the academy together and totally

(58:23):
trying to figure that out back then.
And here we are today, bothworking together again, him we
were to go back to the early days.
Him won Lincoln 20 Me won Lincoln 19,and we're trying to do good and so
it's cool to be back together and.
And doing good with these people that Ihad chances to do that with in the past.
So yeah, 10 years.
But it's all good and it is just a number.

(58:46):
Yeah, yeah.
Well I was even thinking it's been fiveyears since we published the Sean Bear's
Guide to hiring a Law Enforcement Analyst.
. It was about five years around thistime that we were recording the session.
So that's
crazy To me, that seems likeit was two years ago, dude.
But yeah.
And so for our listeners, I'll putthe link into the show notes for

(59:08):
all this stuff so you can get moreinformation, including the hiring guide.
Yeah, Sean, I'll just, just wantedto finish up with just maybe
what's next in terms of AI and lawenforcement analysis and some maybe
something that people can be on thelookout for in the next five years.
Study it up.
Now, in terms of ai,
I'll tell you, I wish I could be that guywho would say, here's what you need to do.

(59:30):
Mm-hmm.
And here's what's coming.
And mark my words.
It's gonna do this.
20 years ago we could do that.
There's Moore's Law.
The the, the famous sort of tenantof technology was you could expect
it to double about every 18 months.
And so you could make predictions about,well, where will we be in about two years?

(59:50):
Well, you look at what thetechnology is, the technology
doubles about every 18 months.
So if we're at this kind of capabilityright now, what would it be?
18 months, two years from now?
And you can begin toguess at those things.
Oh, I think we'll have this,or, I think we'll have that.
We can't solve this now becausewe don't have the com, the
computational power, but we will,and then we'll be able to solve it.

(01:00:10):
But where we're at now,Moore's law is no longer valid.
It really isn't because now it'shockey stick, it's exponential.
If you just think about where wewere two to three simple years ago
and where we're at today, it's nuts.
I mean, students and individuals who arewanting to learn coding, let's say, and

(01:00:34):
entering college four years ago, fiveyears ago, AI wasn't even a, a mention,
it wasn't a, a part of the curriculum.
And now they're coming outand they're using it , their
day-to-day coding, and it's.
It's doing a, a large portion of theircoding now when they enter into college,
they don't even know if they're gonna havea job as a computer programmer four years

(01:00:55):
from now because of the advancements.
It's literally going to be ableto do that, I think very soon.
So I think you're gonna see a a lot morecode being written by literally anybody.
Like anybody can use spellchecker.
Anybody's gonna be able to use coding tobuild whatever they wanna build, to build

(01:01:15):
solutions that, that help them be better.
I think that's where we'll be able to gois that I think people that, wanna be able
to have something happen for them, theywon't look first to a commercial solution.
They'll look first to, well,I'll just build it 'cause it's
only gonna take me an hour.
I'll just have my AI.
Bot do it, or the robot who'smowing my lawn right now, I'll have

(01:01:37):
them come in and code it for me.
Yeah, I think we're gonna see Jarvisbefore you know it, I've been trying to
build, and that was part of the thingthat I was doing with, with this writing
this book, is unfortunately because ofthe, the advancements over the last year
and a half, two years that I've triedto write this book, I've had to rewrite
so much because it changes so fast.
I'm at the point now where it'slike I just gotta get this book out

(01:01:58):
because by the time I get it out,it's already gonna be outta date.
So I think we're gonna see that whereit's, you can't predict what's gonna
be three months from now, becausethree months ago you wouldn't have
imagined that we would have videosthat would be created that you
couldn't tell the difference between.
Real and fake.
And here we are.
All right, Sean.
Hey, it was great catching up with you.

(01:02:19):
Thank you so much for yourinsight and catching us up today.
And the best of luck withsworn AI and on on the book.
You're awesome.
I appreciate you and, and all thatyou're doing for the profession.
Thank you for, catching up with me andI think we need to revise that hiring
guide now that you said that becausewe need to incorporate AI into that.
But yeah, thanks for having me on.
I appreciate it.

(01:02:39):
Thank you for making it tothe end of another episode of
Analyst Talk with Jason Elder.
You can show your support by sharingthis in other episodes found on
our website@www.podcasts.com.
If you have a topic you would like usto cover or have a suggestion for our
next guest, please send us an emailat Elliot podcast at time talking.

All Episodes

Analyst Talk - Sean Bair 2.0 - AI, Analysis, and Officer Wellness

Episode Transcript

Popular Podcasts

24/7 News: The Latest

Dateline NBC

The Clay Travis and Buck Sexton Show

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}Analyst Talk - Sean Bair 2.0 - AI, Analysis, and Officer Wellness