America's Criminologist 06 23 26 Guest Aaron Cohen - Founder Gideon AI

Episode Transcript

Speaker 1 (00:05):
Col steal sust wait just trusting him as a faith.

Speaker 2 (00:12):
Shadows secres Lot.

Speaker 3 (00:14):
Doctor currys s I'm Cosquam analogist, co.

Speaker 2 (00:23):
Sales of the Law in the Cholos.

Speaker 1 (00:29):
Les A Crime, Break and Change, Bruce Star a Fame
and here's America's Criminologist, Doctor Kerry Myers.

Speaker 4 (00:41):
Howdy, Howdy, how do you welcome to America's Criminologist. Another episode.
I'm doctor Curry Myers, retired major county sheriff, former special
agent State trooper. Also spent fifteen years in the world
of academia. Don't hold that against me, serving as a
school dean and as a professor teaching criminology. This show
where we talk about crime culture and the foundations to

(01:02):
sustain a civil society. I have a new book coming
out in the spring of twenty twenty seven. It's called
The Rise of the Pharaoh Man and the Return from
Pharaoh Man. As you know, I often talk about the
elements of feraoh Man, a wilding of humans that has
predominantly occurred really within the last decade or so. This
is going to be a two volume examination of America's

(01:24):
moral and cultural crisis. The first volume, Rise of Pharaoh
Man explains how we lost our way. The second volume,
The Return of Pharaoh Man, offers a practical roadmap for
restoring the three important pillars of society, faith, family, and formation.
So stay tuned for more updates as we get closer
to those reliefs. Dates and always. If you want to

(01:47):
check out my work and applied criminology, criminal justice, ethics,
all those things, you can visit my substack page at
doctor Currymeers dot substack dot com. This segment of America's
Criminologists is brought to you by Real Resilience. That's our
eel Resilience. It's a transformative program helping inmates prepare for

(02:08):
successful re entry. It uses inspirational films and guided discussions.
Real Resilient Resilience teaches practical tools to build resilience, strengthen relationships,
and restore hope for the future. It's real learning through
real stories where positive choices are modeled, not just preached.
Real resilience because re entry isn't just released, It's a

(02:30):
second chance done right. Okay, here is my daily riff.
One of the hallmarks of professional policing in the twenty
first century is moving from a reactive response to proactive prevention.
Evidence based policing models, which I am a huge proponent of,
we don't use it enough, uses a combination of data.

(02:53):
It uses combinations of research validated practices to identify what
actually works in reducing crime and victimization. And with today's technology,
we have all this data right at our disposal, and
yet we can't get to it most of the time
in order to actually implement evidence based policing now, predictive applications,

(03:15):
when used responsibly and within the constitutional boundaries of the
United States, can help law enforce law enforcement identify patterns,
recognize emerging threats, and interviewing intervene earlier in the violence cycle.
The goal isn't to predict the future with certainty. It's
to better allocate resources, improve decision making, and potentially prevent

(03:38):
tragedies before lives are lost. That's the promise of evidence
based policing, Smarter strategies, better outcomes, safer communities. In short,
my friends, law enforcement agencies possess enormous amounts of information,
but most of it remains locked away in reports, databases, records,

(04:00):
and often used only after an event has already occurred,
if at all. The future of policing lies in leveraging
advanced AI driven systems capable of rapidly retrieving, connecting, and
analyzing information across multiple sources to identify relationships, patterns, and
potential threats. And humans cannot do this alone. It's called

(04:23):
volume overload. They do not have the capability. We don't
have the manpower to be able to do it. AI
is our friend. We can't be afraid of it, and
we have to be able to use it responsibly. In
the future. These technologies can help transform policing into more
evidence led profession focused on prevention, efficiency and public safety,

(04:45):
while remaining firmly grounded in our very important constitutional principles.
That brings us to the guests today. I am super
excited to have Aaron Cohen. You really can't look at
the new news or have him appear on news because
he is such an important figure on terrorism and counter

(05:08):
terrorism and he speaks the truth. That's what I like
about him, as well as he lays it out the
way it should be done. He's had a trained career
in special operations and intelligence veteran and has served in
the counter terrorism space for over two decades. Again, he's
one of the most recognized people on camera when it

(05:28):
comes to law enforcement issues, counter terrorism issues, reoccurring guests
on Fox News, you probably see him a lot there.
He was on for I remember seeing him back on
Larry King when he was on all the time in
Larry King and as a reoccurring guest. He's also an actor.
He's been in numerous films and he's trained many Hollywood

(05:52):
stars on self defense and close quarter battle techniques. He's
the CEO and founder of Gideon AI. Gideon is an
intelligence platform designed to help law enforcement detect, assess, and
prioritize potential loan actor threats and really as well as
large conspiracy threats because a lot of these loan actors

(06:13):
may be working with others, but the conspiracy, the overall conspiracy,
is not known to them. Even so there may be
a network of fifty people when they think they may
be working by themselves or with one other. And this
is an important concept. It fits into the evidence based
policing model and I am super stoked to have him

(06:33):
on board. So Aaron, Welcome to America's Criminologists. Thanks so
much for taking the time. It's really an honor for
me to have you on the show.

Speaker 2 (06:40):
Appreciate you having me on and letting me speak to
your guests. It's not often I get a law enforcement
professional that I get to speak with outside of the
law enforcement space, So to be able to do it
in the media worlds, it's refreshing, grateful to be here.
Thank you, Thank you so much.

Speaker 4 (06:57):
We try to you know, we have about fifty seven minutes,
so we try to get fairly in depth, which is
I think is important. I do a lot of radio
and television as a guest myself, and you can't talk
with any kind of efficacy on programs like this in
a five or seven minutes.

Speaker 2 (07:13):
We'll try, we'll try and move it quick and hit
these points here.

Speaker 4 (07:16):
Well, yeah, so Aaron, many Americans think terrorism and mass
violence are primarily organized group activities. Yet today we hear
more and more about loan actors and again sometimes layered
conspiracy threats. How has the threat environment changed over the
past twenty years?

Speaker 2 (07:35):
So twenty years ago, the threat environment, in my opinion,
was largely dominated by organizations. So you had a a kaita,
you had isis obviously we're still dealing with them. You
have foreign terrorist networks, street gangs obviously coming from the
sheriff's world with a large or major county where you're
coming from here in Los Angeles where I live, structured

(07:55):
groups with leadership, financing, recruitment pipelines is how I describe it,
and command and control mechanisms. And today the center of gravity,
in my opinion, has shifted dramatically towards the loan actor.
And so the Internet has effectively democratized radicalization, I think

(08:19):
is the most straightforward way to put it. A person
no longer needs to join a terror organization to become dangerous,
doctor they just don't. You can find ideology, grievances, tactical information, validation,
and community online without ever leaving their bedrooms. So I
think what we're seeing today is the rise of the

(08:42):
self directed offender, is what I'm going to call him.
And these are individuals who may never receive orders from anyone.
They radicalize themselves, They build their own narratives, and they
develop their own grievances, and increasingly they move from thought
to action without traditional indicator is that law enforcement was
trained to look for twenty years ago. And so the

(09:04):
challenge isn't that these individuals are invisible. I want to
make that clear it's that they often exist in digital
spaces that are moving faster than traditional investigative models. And
so that's why I often say when I'm speaking to
a major department and we've got dozens and dozens of

(09:24):
them across over twenty states. Now with my new AI platform,
what I often say to them is that the Internet
is now the largest threat intelligence environment in human history.

Speaker 4 (09:40):
Yeah, no question about it. And the thing is it
kind of goes with my feral man concept is we
have this wilding of humans that they have been radicalized
through social media and social contagions. So some of them
have the dark psychological traits because they've gotten into it
so deep and so far that they can't get out
of it, and they get reinforced. You know, it's this

(10:02):
constant reinforcement that's occurring on a positive sense because they're
only you know, it's that cognitive bias where they're constantly
bandwidths effect on each other. From your experience, what behavioral
patterns tend to emerge before an individual commits a terrorist act,
a school shooting, or workplace by in it, there are
certain markers that they may possess and do. And this

(10:25):
is the importance of Gideon, But what are they most demonstrated?

Speaker 2 (10:30):
So it's a good question. I think the most important
thing that the public needs to understand is that the
attacks rarely emerge out of nowhere. The idea of the
lightning bolt attack is absolute nonsense. It's a myth. And
what we typically see is progression, and it starts with
grievance and something happens that the individual perceives is unfair, humiliating,

(10:53):
maybe threatening, or I guess the word would be intolerable,
and so that grievance becomes fixation. The fixation becomes ideation,
and then the ideation becomes research. That's the pipeline. And
then the research becomes what we call the preparation phase,
and the preparation then becomes the leakage, which is what

(11:17):
we're seeing the manifestos. We just saw one the other
day with a shooting in Montreal and quote NJ. And
so eventually that pathway culminates in what we call the mobilization.
The mobilization is the weapon gets loaded, the attack is underway.
Now law enforcement is forced to respond, and so the

(11:40):
specific details vary, but the behavioral pathway remains remarkably consistent
across terrorism doesn't mat School shootings could be workplace violence
and whatever form the targeted violence takes shape in. So
this is why modern threat assessment focuses on behavior and
not the labels. And people can call themselves anything today,

(12:02):
you know, and so what matters is what they're doing.

Speaker 4 (12:07):
I'm I spend a lot of time on crime prevention
through environmental design SEPTED strategies. One of the one of
the failures of SEPTED is that we do not do
enough behavioral analysis within the SEPTED model. I'm going to
make a pretty good assumption that Gideon AI can even
uh do a better job in helping with SEPTED assessments

(12:30):
moving forward on particular particular corporation schools that are unique
to a certain area, Because not all schools the same,
not all businesses are the same, not all football stadiums
are the same. They all have a unique identifying footprint
and activity based on the culture around them, the size
of the city, all those other markers that may be

(12:52):
going on. Is AI more than just law enforcement, but
can be helped in actually the build out and future,
say for spaces.

Speaker 2 (13:02):
I think AI is going to drip into all of it.
So as long as you're focusing on the behavior, and
that's one of the main differences between well, it's not
one of the main differences. It's specifically how I build
Gideon and what I designed it around, which is I'm
not focusing on the ideology. I'm focusing on the behavior.
The behavior doesn't care about the ideology doesn't care whether

(13:23):
you're an insul, whether you're a transhooter, whether you're inn Islamist,
doesn't care. What you could be, a far left extremist,
a far right extremist doesn't The location doesn't matter either.
So as long as you're following the behavioral pathway what
I laid out, you're going to catch the correct leakage.
And we'll get into more of that as we kind
of push through here, and I can sort of expand

(13:43):
on that.

Speaker 4 (13:45):
Right, let's talk about Gideon specifically. So what is it
exactly for the average listener and viewer. What's the best
way to explain Gideon.

Speaker 2 (13:55):
So, Gideon is an AI powered early warning intelligence platform
designed to help blon enforcement identify potential loan actor threats
before violence occurs. That's the main definition of it. The
easiest way to think about it for your viewers think
about it as an early warning radar system. Most policing
tools are built to investigate events after they've happened. Gideon's

(14:19):
designed to identify behavioral indicators before the violence occurs. So,
but I want to be clear, doesn't invest It doesn't
replace investigators or detectives. It also doesn't replace threat assessment teams.
It simply helps agencies identify signals that may otherwise go unseen.
And so the goal here is in prediction. The goal

(14:42):
is earlier detection. That's the That's the basic way to
explain it.

Speaker 4 (14:47):
And how does the platform function operationally? Does it? Is
it on the cloud? Is it has to be unique
to a server at the organization? How do officers and
detectives use it? Is it best served by the intelligence
analysts using this? How does that kind of application.

Speaker 2 (15:02):
Work so operationally Gideon, there's no integration. It's its own
server and it sits on its own cloud based setup.
And what it basically does is it continuously analyzes publicly
available online environments. Okay, so the platform is again looking
for those behavioral indicators, but specifically for targeted violence pathways.

(15:29):
You do not need to be a megacity with a
ten man intel team. You can be a twenty five
to fifty man agency, which is what we spoke about
pre interview here, which is the majority of the size
of agencies. We're a force multiplier and we've got thresholds
to meet every budget and the challenge of these budgets.
And so what Gideon does is it evaluates the patterns

(15:51):
rather than isolated statements. This part is important, and it
evaluates escalation, and so it's looking for coext So once
you're an agency and you onboard and we put you
on a pilot, you're gonna have sixty days there just
to soak in how the platform works. All the alerts
are gonna be delivered via email. You never have to
log in. So Gideon is going to be a twenty

(16:11):
four to seven operator online hunting and evaluating context. There's
no keywords associated here, and when we find behavior that
hits a certain threshold, the platform then generates an alert
for review. Someone on our team, a human analyst, will
then assess the information, will add context if needed, and
then we'll begin to look for attribution. And without attribution,

(16:33):
it's passive. You've got nothing. Intel is only as good
as the location. But the problem with the Internet is
that it's the largest anonymous threat intelligence environment. So ultimately
a law enforcement once we get attribution, will decide whether
first further assessment is warranted. The key point is that

(16:54):
there's a human that remains in the loop at every stage.
Then what we'll do is we'll push that out to
our network of partner agencies. After we've routed it to
the correct agency. It may go to a fusion center,
may go to a real time crime center, may go
directly to that agency, or there's no location or attribution
or not enough exigence for the platform to give us

(17:16):
an emergency disclosure or one of our partner agencies in
emergency disclosure, then we'll go ahead and push it to
the FBI by default let us sit on their lap,
and then they commemit and take a crack at it.
But the point is is that we're able to surface
a tremendous amount of exigen posts with very imminent threat
language and get it to the correct jurisdiction in under

(17:40):
an hour, which is unheard of and so open source intelligence.
Nothing classified on here. We're not going into any private
chat groups. If it's not online publicly, we're not interesting.
By the way, these guys aren't hiding. This isn't isis
you know, two thousand and three or four, And so

(18:02):
that's basically how the platform operates and how the mechanism
functions in a nutshell.

Speaker 4 (18:09):
And you mentioned fusion centers, So for the average person,
they may not know there's fusion centers in regions throughout
the United States to help gain intelligence, identify intelligence, share information,
intelligence back and forth. Quite frankly, a lot of it's
historic and out of date because.

Speaker 2 (18:26):
They I mean, well they're relying on tips. Yeah, yeah,
it's going to rely so without tips, the fusion centers
are dead in the water. In fact, most law enforcement
relies on tips.

Speaker 4 (18:35):
So it's exactly right.

Speaker 2 (18:36):
So my operational goal with the whole platform here is
to move the response time farther upstream. So right now
law enforcement is positioned to be able to respond to
an attack, and the training has gotten better than ever.
I will spend twenty years in the training business. When
I got out of these rarely special forces, it's training
swat teams for fifteen years. So having been on the
reactive side of this thing, it was like, this's got
to be a better way to do this. So by

(18:56):
looking at the AI looking at all the data from
the Secret Service in the FBI Israel. So it's got
to be a way to operationalize all of the data
from all these past loan actor attacks. So the goal
is just to move the response time further up the timeline,
you know, left of boom. It's a little cliche, but
it's the term. But the fact is is that again,
the Internet is the largest battlefield today. If you're not

(19:18):
in that space, you're reacting. But we can do so
much better than that. And so now we are. We
moved the timeline up and we're getting agencies signal for
the first time in the history of public safety. So
I Gues spen hours talking about this. It's just it's
a very exciting time to be in this space, to

(19:39):
really lead with this. Now what I'm calling intelligence based policing.
It's kind of what I'm framing it as.

Speaker 4 (19:45):
Yeah, no, it's it's I it could be something as
new as the advent of DNA. I mean, that's how
important what you're working on can be. I mean, it's
a major threshold. And you mentioned behavior ontology, which is
that you know, it's that relationshide human action. You know, behaviors,

(20:07):
those kinds of things. But law enforcement also has volumes
and volumes of intel inside dead reports that they're not
able to access. Because that does your system can it
go in and cultivate that information out as well as
doing the behavior ontology side.

Speaker 2 (20:26):
So it's an interesting question. We can't take what's what's
in their data. But let me back it up a
little bit. So Israel learned something? Really, I want to
talk about the ontology. It's it's an important piece that
you hit there. Israel learned something very important through decades
of counter terrorism operations. Behavior it matters the most, and

(20:47):
I know we hit this for a bit, but people
become obsessed with the ideology. The ideology alone is a
very poor predictor of violence. I want to be clear
about that. Many people hold extreme views, but they never
can make crimes. And so the more important question, in
my opinion, and what the ontology of the engine is
built on, is is the individual escalating, Are they becoming fixated?

(21:14):
Are they researching attacks? And if you look at the data,
over eighty five percent of these attackers that we've seen
just over the last couple of years have leaked something.
I think it's over eighty something percent are they mobilizing.
A behavioral ontology is simply a structured way of evaluating

(21:35):
those observable behaviors, if that makes sense. So it basically
just gives a framework for understanding how people move towards violence,
if that makes sense. So that's really one of the
biggest pieces here, and one of the challenges moving forward
will be and I know the platforms are getting better
at policing themselves for that imminent behavioral leakage is moving

(21:59):
away from leakage that isn't just focused on imminent threats,
but on escalatory language, which will be the challenge because
we have laws that we want to follow here, so
we're surfacing both. Anyone can route or find a red alert.
I want to shoot up to school. I wanted this.
I wanted that we get tons of those. But the

(22:21):
real intel, in my opinion, lies in the escalation language,
which is farther upstream, farther away from the mobilization cycle.
And so that's my goal. My goal is to get
as many agencies, thousands of them, to understand what that
escalation language looks like. Because the goal of the platform

(22:45):
is early contact, whether it's a wellness check, it doesn't matter.
It's early contact. Early contact creates friction, Friction disrupts pathway,
and the sooner you can get it or farther away
from that imminent last post where I'm going to go
shoot up the school, which may be too late, how
do we get them earlier in the cycle, as they're

(23:05):
escalating before that mobilization language. So that's going to be
the future as we kind of move this thing out
and continue to build it out and as the AI
gets smarter, it's exciting to see it getting better, which
it is.

Speaker 4 (23:17):
And you mentioned force multiplier. We have diminished ranks, especially
in larger blue cities where it may take us decades
to make up for the loss of officers that have
occurred either through retirements, early retirements, just quitting, or purposeful
reduction and force that's been done by leftist mayors and

(23:37):
people like that. This is an important component as a
force multiplier. You mentioned it earlier. Just how important of
a role does it play in adding to the manpower
If you're if I'm a police chief of a fifty
person department in a city of about thirty thousand, how

(23:59):
can this play a role and force multiplication for my team?

Speaker 2 (24:03):
So what you're saying, So what you're asking is agencies
that are struggling with the staffing shortages right now, did
all the political fallout correct? It's a good question. I
think that the reality is that most agencies are being
asked to do more with fewer people. I think we
both agree, and at the same time, the problem is
that the threat environment is expanding. So ten years ago

(24:26):
an investigator wasn't expected to monitor dozens of social media
platforms and forums and online communities, and today that's part
of the actual environment. So the challenge is scale, right,
So no agency can hire enough analysts to manually review
the entire Internet. So technology becomes a core force multiplier

(24:46):
in my opinion, And so not only will we be
able to will our platforms like Gideon able to help
fill those gaps, but I think the goal and what
we've been able to do here, and we've got several
real universities, we've got several small colleges, the goal is
helping that small intelligence team or small agency with no

(25:12):
intelligence team who's never even submitted an emergency disclosure to
a platform based on an exgen post that was surfaced
and that we may afforded to them. It gives them
the ability for that team to function like a much
larger one. So the best analogy I think is radar.
Pilots still fly planes, Radar simply helps them see farther.

(25:34):
I built getting in to serve a similar purpose, if
that makes sense. So we're able to act as a
force multiplier for agencies who may have fifteen sworn tiny jurisdiction,
maybe one or two deputies or officers on duty. And
so just having access to the platform is putting these
tiny departments upstream, and we're able to build them accordingly,

(25:56):
so it doesn't strain their budget. But the fact is
is if you look at the at the at the
statistical data, these loan actors are popping off everywhere. This
is not a large city problem, is what I'm trying
to say.

Speaker 4 (26:10):
Exactly off everywhere. Look at all the ones that just
got arrested on the FIFA everywhere everywhere, Nebraska, Oklahoma, I
mean they were you know, it's smaller and they're going
to hide the interesting part that you have as we
were talking a lot about the terrorism component, but criminals
are known for leakage as well. Uh so this could

(26:33):
play a role not only in terrorism threats, but understanding
transnational crime, other crime that's occurring, could Gideon could be
used for those purposes as well to find out criminality correct,
not just terrorism.

Speaker 2 (26:49):
Yeah, well, yeah, it's it's it's a good point that
the model of the AI can certainly be oriented everywhere,
but at the moment, Getting is purely a counter to
errors and platform built and design for loan act or interdiction.
The reason why is because that's my background. But I
can assure you that we're seeing things involving Internet crimes

(27:10):
against children just connected to the larger transnational criminals syndicates
out there, So there's definitely room to be able to
expand the operational scope, but for now we're focusing on
that loan actor terrorism because that is that, in my opinion,
is the imminent threat right now that just seems out
of control that agencies can't put their finger on because

(27:30):
they don't know that GETTING exists yet. But they're learning fast,
and we're getting and we're expanding quickly, so we're probably
one of the fastest growing law enforcement tech verticals in
the world right now. As far as getting to agencies,
the word is spreading. But you know, cops, they talk.

Speaker 4 (27:49):
So the word is spreading. Once they like something, they
like it.

Speaker 2 (27:52):
You know, what do they say? The only thing that
cops hate more than change is what something new?

Speaker 4 (27:57):
Something new?

Speaker 2 (27:58):
Yeah, so you got to really make this simple for them,
and we've done that. So it's spreading quickly and we're
able to really they're very impressed. Once they're on the
platform and seeing what's upstream. It's just a completely new world.
They're seeing that they had no idea existed. But I
cack and the rest of it. It's coming soon.

Speaker 4 (28:14):
Yeah, Aaron, stay for our second segment coming up. Sure,
all right, we'll be right back with Aaron Cohen. Very
interesting topic on the use of Gideon. You're listening to
America's Criminologists. We'll continue right after this commercial break.

Speaker 5 (28:44):
Hi, folks, Doctor Curry Myers here to let you know
that my new show is called America's Criminologist with doctor
Curry Myers every Tuesday at one pm. Is a former
State Trooper special agent share for a major county. I
will offer sharp insights into the pressing issues shaping American society.
I'll have guests, news and my insights as an applied
criminologist throughout the one hour show so criminals and the

(29:05):
progressive politicians that allow them to fester. Beware because this
show is directed at you, America's criminologist, every Tuesday at
one pm on km e T.

Speaker 2 (29:16):
Hello, listeners, this is Christopher from the Christopher Show.

Speaker 1 (29:19):
Hey, if you miss one of our shows you're at
km ET.

Speaker 4 (29:22):
Don't worry about it.

Speaker 2 (29:22):
You can go to our web page.

Speaker 1 (29:24):
And that's kme e T fourteen ninety am dot com.
Go to the homepage, click on the SoundCloud tab and
hear any show anytime you want.

Speaker 3 (29:37):
Take KMET Radio with you everywhere you go by downloading
the new, updated, free iPhone, Android, phone and tablet remote
app found in the Apple App Store or Google play Store.
Listen live to your favorite shows on all of your
remote devices. Information can be found on the KMET Radio
website at km T fourteen ninety am dot com. That

(30:00):
weapon dress again km e T fourteen ninety am dot com.
Download your free new KMET Remote app today. Go to
the Apple App Store or Google play Store for your
free new KMET Radio Remote mobile app.

Speaker 2 (30:17):
Today.

Speaker 1 (30:21):
We're chatting with our guest always safety first eight one
to one, megaphone.

Speaker 2 (30:25):
Man only one one Before you dig.

Speaker 1 (30:29):
You never straight from your mission of letting homeowners, landscapers
and excavators know. You should always call eight one.

Speaker 2 (30:34):
One before you dig.

Speaker 1 (30:35):
Only one one call for both small and big jobs,
and two working days in advance, A one one. Any
parting words, and I'm guessing they will be only one
one before you dig.

Speaker 2 (30:48):
Brought to you by Southwest Gas A one one.

Speaker 6 (30:52):
As you know, I have been talking about standing up
for our country, doing something. These big box stores skate
open little guy's clothes.

Speaker 2 (31:01):
Will our voices be heard?

Speaker 6 (31:03):
Does voting with our dollars work? Patriotswitch dot com gives
patriots the power to walk away from the big box
stores forever. Stop supporting the enemies of freedom. Now that's
why patriotswitch dot com was created for those who support America.
Patriotswitch dot com.

Speaker 7 (31:22):
Hey, my friends, is doctor Curry Meyers, America's criminologist. I
have a new sponsor for my show. I mean Real Resilience.
That's our a e l transformative program helping inmates prepare
for successful re entry and uses inspirational films and guided discussions.
Real Resilience teaches practical tools to build resilience. It's really

(31:43):
an important tool. Real Resilience because reentry isn't just released,
it's a second chance done right.

Speaker 4 (31:57):
All right, Welcome back to America's Criminologist. Forget to check
out my work on Substack at doctor Currymeyers dot substack
dot com. We're joined by Aaron Cohen discussing Gideon, a
threat detection platform designed to help law enforcement identify potential
loan act or threats before violence occurs. Erin is the
CEO and founder of the organization Okay Erin, Let's talk

(32:20):
a little bit about some critics who like to label
this stuff pre crime, that kind of that word gets out.
They don't, they don't they they I think the pre
crime the Minority Report reports stands out for for civil libertarians.
How does one address those kinds of critics and let

(32:40):
them know that, like anything, boundaries are important than this
will have boundaries as well.

Speaker 2 (32:47):
So it's an important question. I think the biggest misunderstanding
that people here when they hear the word predictive and
assume we're trying to predict the future.

Speaker 5 (32:55):
We're not.

Speaker 2 (32:56):
We're detecting behavior, and there's a very big difference. We're
not identify find people who might someday commit a crime.
There's no halos here. This isn't Steven Spielberg, It's not
Tom Cruise that was a film. This is in nineteen
eighty four. We're identifying I want to make it clear
publicly observable behaviors that research has repeatedly associated with violent pathways.

(33:17):
We've taken over thirty five years of data from the
Secret Service and the FBI, and all we've done is
operationalize it to look for the same exact behavior. In fact,
getting is comprised and built the core of the wet
engine of over one hundred threat behaviors specifically associated with violence.
And so to be clear, law enforcement does this every day.

(33:39):
You would know this. Threat assessment teams already do this
every day. School threat assessment teams already do this every day.
We're simply helping law enforcement identify information earlier and at scale.
And I tell people this isn't minority report. This is
behavioral threat assessment supported by technology. Very different things.

Speaker 4 (34:00):
Let's look at something like the nine to eleven attacks.
There was leakage being done by those attackers. There was leakage,
was law enforcement stop some of them for violations of law?
They were taking some classes in how to fly airplanes,
not land them, but fly fly airplanes. Would your system

(34:24):
have been able to help identify that kind of leakage,
not able to prevent that kind.

Speaker 2 (34:31):
Of not back in two thousand and one, because the
internet was still was still I.

Speaker 4 (34:36):
Mean in today's I mean in today's If in today's system,
where they would be kind of doing the same thing,
would this be a process that would be caught easier
with your system?

Speaker 2 (34:50):
I agree that any terrorism that happens today, there's some
type of leakage online. So to directly answer your question,
if there was a nine to eleven style attack, which
I don't think is what today's generation of terrorism is
going to look like. Again, all of the radicalization is
happening digitally. Most of the radicalization is not ideal ideological,

(35:13):
So this isn't pre nine to eleven Islamis with their
grievances towards the White House and the old you know, uh,
get America out of the Middle East. But regardless of
the ideology, there is leakage. There's always leakage. The DC

(35:34):
shooter posted posted a manifesto. The Minneapolis shooter posted a
ten minute tactical video online to all the magazines with
Trump written on them and on the tactical gear the
shooter that we just saw on Cotonege in Montreal. One
hundred page manifesto. So the reason why all of this

(35:55):
material is catchable, and the reason why soft a platform
like getting is able to grab it is because today
due to the Internet and the thousands of bot farms
being put out there by China and Iran and Turkey,
which is designed to tear Americans apart, Terra America's relations

(36:15):
from Israel apart, infuse us with all of these other
radical ideologies on both sides of the of the political spectrum.
Think about that. Our enemies and I'm not going to
get conspiracy here, have already fired shots into our country
without having stepped foot here. They're able to spike the
comments section of the of these posts and cause them

(36:37):
to shoot up to the top of those algorithms, which
creates the rage bait, and then the media platforms and
the social media platforms reward rage not facts, and that's
what's pulling this country apart. And so all of that
material is being posted a lot, most of it's being
posted long before the manifesto, and it's being posted in
the form of grievance. It's being posted in the form

(36:58):
of ideation. Giving example, suicide ideation is one of the
major early warning radar blips that Gideon is hunting. The
reason why is because today's loan actor the goal is
not homicide. If you think about what they statistically do
to themselves after they've opened fire and the bodies are
dead on the grounds, what ends up happening the moment

(37:21):
that tag team or law enforcement unit makes entry. By
the time they get to them, they end up doing
what They end up killing themselves. They end up killing themselves.
And the reason why is because homicide was never the goal.
Suicide was always the goal. Homicide was just a way
to give meaning to their own self destruction. So suicide

(37:42):
ideation is one of the critical early warning intel nuggets,
and that's what Gideon is hunting systematically, and we're trying
to get it before it pivots outwards, and that's where
that early escalation warning comes up. And that's why it matters.
There's a major, major ment health crisis that's percolating and
we're seeing it unfold in the wilds on the internet.

(38:05):
So we're hunting it, we're monitoring it, and the moment
that and it's being posted out in the public, it's
not being hidden in these chat rooms and signal groups.
Those signal groups start to merge. Those private groups start
to operationalize once the online radicalization has cemented, Once enough
people who are willing to die because they've been radicalized

(38:27):
have come together, and as they start to pivot outwards,
that's when they start forming these little groups like we
saw with the UFC two to fifty attack that was
thwarted brilliantly by the FBI. I might add in the
agencies and the Secret Service that was involved, and so
the signal is there, Doc, and I'm gonna say that, yes,
we're we will find it. We're gonna hunt down every
single bit of it, and we're gonna route it as

(38:47):
quickly as possible. We may not get to it sooner
than later, but we're going to do our best to
try and get to as much of it as we can.
The more agencies we onboard on the platform, the quicker
we can move that intel. Because now we're creating this uh,
this National Early Actor Interdiction Network. It'll be the largest
intelligence network of law enforcement professionals ever. And that's really

(39:10):
the product that's the moat when you've got thousands of
agencies who are professionals, small, big, medium, doesn't matter, fusion centers,
real time crime centers, mega city agencies, small town police departments.
Some of the best investigators, as you know, are some
of these small towns. They're talented, you know, it doesn't
matter where you are. And so I believe that we
can get to them. We're going to get to and

(39:30):
we're going to try and get to every single one
of them. And the goal here is just to keep
building this thing out, getting the network larger, getting the
AI even tighter, in the language learning model even tighter,
and then eventually starting to bring on federal partnerships to
UH to help us continue to expand it. I think
it's all preventable, Doc.

Speaker 4 (39:48):
Yeah, and the loan these loan actors that are out
there also get radicalized by someone either through the Internet,
social contagions, or we have cells operating in this country
where their job is to try to radicalize. We see
this on the college campuses. It seems like there's always
someone always on the college campus has some connection to

(40:10):
either the Chinese or Islamis in some way back and forth,
or perhaps there's some communist idealization. So we have these
cells that are operating as well. Some of them are
a lot quieter and their their main job is to
try to recruit operators and foot soldiers to be able

(40:32):
to go out and do this loan actor work.

Speaker 2 (40:35):
So well, they're already they're already doing it. They don't
even have to be here. So if you've got a
bot farm with a thousand cell phones spiking anti American
rhetoric into the comments section and spiking anti American content,
than what you've got when you've got a nineteen or
twenty year old student in college like Mangione for instance,
who's all, who's what we would classify as an anti

(40:58):
capitalist actor. Or you got the recent shooter up in
Montreal just yesterday who initially was labeled as an inso
before the full manifesto was released, but now they're saying
he was a far left extremist, anti Israel, anti American,
anti capitalist, anti Trump. Again, regardless of the ideology, the

(41:19):
fact is is that something something hit his feed. This
guy did not meet an al Qaeda recruiter in some alley.
You know what he met, doc, He met an algorithm.
So nobody has to be here. All of the comments
can get spiked by thousands of BoNT farms outside of

(41:40):
our shores and get right into the mobile phones of
these kids. And guess what happens. He keep staring at
that phone long enough and the parents aren't going to
take them off the phone. Keep staring at that phone
long enough, and what's going to happen. You're gonna have
a kid who eventually crosses over, who was already suicidal
now starts to pivot into homicidal idea. Once you've got

(42:01):
that convergence of that outwards pivot, now you've got yourself
violent thoughts getting posted, homicidal ideation being posted, a history
of suicide ideation being posted. I want those posts, and
I want to know where that actor is, and I
want to get that information to that agency as quickly
as possible so they can make early contact. Because eighty

(42:24):
percent of these actors, in my opinion, don't even know
why they're being radicalized. In fact, if you watch what
they're saying, they have no idea what they're talking about.
Nobody understands the history of anything that's going on today.
It's absolutely nuts. So let's get them. Let's get early contact,
get those law enforcement officers to those doors, and let's
get some wellness checks going, and you'd be surprised when

(42:45):
you look at the history here, got a lot of
mental health and suicide ideation long before the gun was
ever loaded. So again, critical intel nugget, we're gonna hunt
it all. We're gonna get it to these agencies as
quickly as possible.

Speaker 4 (43:00):
You know, you mentioned the importance of this could be
to really change the law enforcement world when it comes
to intelligence gathering and being able to assess better and
more quickly. But you have to have still some boots
on the ground to be able to go do these interviews,
knock on the doors, that's right, do you do you foresee,
at some point there be a team that is dedicated

(43:23):
to doing network, specifically at a department where they're their
main job is to go, uh do those initial.

Speaker 2 (43:33):
Knocking talks knock, knock and talks. Yeah, it's a good question.
So I think given the scale and the aggressive growth
of radicalization online, on the fact that America's enemies are
able to get into these algorithms and spin these kids
up to the point where they want to load up
a weapon and open fire My contention is that local

(43:56):
municipal counties, not just federal now most of them have
criminal intelligence capabilities. They've got nodes, they've got investigator, they've
got analysts. Now it's getting better. It's been getting better
for the last decade, and Intel's been seeping into that
to the space more and more on the law enforcement side.
But I can tell you that at my agency where

(44:16):
I serve in the Southeast, I've been a reservist there
coming up on twenty years. Start off on the training side,
we just stood up the first ever loan actor interdiction unit.
The goal of the unit is to be able to
operationalize the types of behavior that my system is already using,

(44:37):
to be able to hunt down that signal, to determine
and make sure that those actors are not in our jurisdiction.
If that makes sense, that's just basic detective deduction or
deductible reasoning if I'm not mistaken from your world. So
we've already stood up probably the first ever local loan

(44:58):
actor intelligence and counter terrorism specific unit on the local
level to be able to hunt down that signal, make
sure that it's not in our region of I'll say
half a million citizens but then move it to the
right jurisdiction as quickly as possible. And I look at
this as digital mutual aid. You know, if you need

(45:21):
something from a county, sure if you know this, if
you need something from a county, you guys don't have
access to aerial assets, but you've got a hiker who's down.
You've got mutual aid agreements with those surrounding jurisdictions, right right.
We need to create that capability and that culture for

(45:42):
the digital threat space as well, which means my contention
is that local and county state agencies already have these
because they're working with diffusion centers. But really this is
a local problem because these attackers are going to pop
off in those local backyards. I think that loan actor, interdiction,
intel driven units, whether it's one investigator or five or

(46:03):
or just one, if that's all you can muster to
get to the Internet and use tools like Gideon, and
there's other great ones. There's data Minor, there's a lot
of different tools now that real time crime centers are
using to be able to get to that signal early
enough if and make sure that it's not in your
area of responsibility, and then learn how to move it quickly.

(46:24):
That to me is going to be the future of
intel based policing. I don't know of any other platforms
sure if that do specifically what Gideon was built to do.
Most of the platforms will tell you either what happened
in the past, like Pallunteers, you can put together a
mosaic of how to get Ben Loden in the future.
Some of them will tell you what's going on right now.
They'll monitor Twitter feeds so that you can get response

(46:48):
to that specific area if it's near you. I believe
that we're the only platform that's actively trying to tell
you who someone is becoming, and that's what makes us unique.
We're only focused on what will potentially transpire, and if
it doesn't turn into anything, I say great. But the

(47:11):
good news is that somebody who's spiraling in suicide ideation,
who's pivoting into homicide ideation through obsessive, intrusive thoughts and
has a history of mental illness, let's make contact. What's
the problem, get them the help they need. But the
problem is is that if we can't get the post

(47:32):
to the correct agency, there's no case sheriff, there's no
case number, so how do we start the case. We've
got to get the posts. We need thousands of these
agencies in this country helping us move them to the
correct place. And we do that through emergency disclosures, request
to the platforms, and getting these kids to help they need.

(47:55):
They just they still don't understand what's happening to them.
It's okay, we do uh, And I say as much early,
the morally contacted better.

Speaker 4 (48:04):
Yeah, and it's that breakdown of those three important tiers
that I talked about. Faith, information, they don't have those,
that's right. They haven't been formed correctly. They haven't don't
have a family life, and they don't have a faith life.

Speaker 2 (48:15):
So and their networks are online. Honestly, you've got a
whole generation of kids. It's different than you and I.
I mean, their entire community is happening on Discord now,
it's happening on Reddit. You know, it's it's a different time.
They're not using the Internet to share their wins.

Speaker 4 (48:28):
Uh.

Speaker 2 (48:28):
They're using the Internet to make connections. And if you've
got nihilistic, violent extremist groups infiltrating these environments, creating self
harmed communities and and and riling up the suicide ideation. Uh,
you're creating an entire generation of killers. I don't know
how else to put that without being totally honest. You

(48:49):
won't mind. That's my Israeli hat. You're creating killers online
right now, and we've got to get to them and
talk to these kids first and pull them out of
that environment as quickly as possible.

Speaker 4 (48:58):
Yeah. I know you recently spoke to the new Major
County Sheriff's Association. I was a member of that as well.
It's within the National Sheriff's Association. It's a very important organization.
I think that sheriffs are in a unique position because
of expediency. They have the ability to implement things a
lot quicker than a police chief who may have to

(49:18):
build a budgeture.

Speaker 2 (49:20):
You know, they just have the autonomy they can and
they can form that policy.

Speaker 4 (49:23):
They canform things quickly, and that's what I liked about
being a sheriff. I encourage sheriffs to put this to
the top of the agenda because working in connection with
each other as major county sheriffs filtered down to other
sheriffs across the country, is vital and key to being
successful in being able to have contact information that you

(49:47):
can share. Again you talked about it could be a
small county, but the ability to say I know sheriffs
such and such in this small county as it's going
to him or her and be able to be actionable
because it's not in my jurisdiction.

Speaker 2 (50:02):
And sheriff any any agency has the same uh incredible
capability on my platform. And I know, uh, you know I.
I'll give you an example. I when I I we
talked when I spoke at the I've got a great
relationship with Sheriff Mike Bouchard from Oakland County.

Speaker 4 (50:16):
Yeah, he was a president when I was.

Speaker 2 (50:19):
Yeah, he's fantastic. Mike is a good buddy of mine.
I'm honored to have him in my orbit.

Speaker 6 (50:24):
UH.

Speaker 2 (50:24):
He's been instrumental in in helping to move information, helping
to shape opening doors for us. He was the one
who brought me out to DC in February to speak
to the UH UH the Intelligence Commander's UH sub Committee,
which is where I spoke. And I can tell you

(50:45):
that being in that room and being amongst the hundred
you know of the major county sheriffs and their intel
commanders was eye opening. They all want to know how
to do this. Everybody's starving for the solution. If you're
a sheriff out there, real quick, sheriff. If you're a
sheriff out there, get on my platform. I'll put you.
I'll put you up here for sixty days, no obligation.

(51:07):
Just get upstream and see what we're seeing. And if
it makes sense for you operationally, which I promise you
it will, then we'll get you guys on an affordable
license to be able to stay on the platform, because
without it, you're going to be dead in the water.
So if you're out there and you're listening, large major county,
if you've got five sworn and you're in I don't

(51:29):
know where, I don't care, call me. You'll have the
same capability as a large agency with a ten men
intel team, by the way, a ten or twenty man
or fifty men intel team. And we've got some of
the largest agency sheriff on the platform. Now, I still
ask them the same question when I'm briefing them on
the platform and onboarding them. How scraping the entire internet

(51:51):
working out for you? If not, it's impossible. You can
have one hundred analysts sitting on an entire floor in
a high rise. You're still not going to be able
to hit them million posts per month that are out there,
So you've got to use the technology. But we've got
to have the agency's partnered so that they can get
into the network and create the first ever loan actor

(52:13):
national Early Warning network, which is really my vision of
the platform. The MOAT, the sale to acts On or
Motorola in seventy two months from now is the moat.
The ability to have twenty five hundred to five thousand
of the finest intelligence teams, major city, tiny agency, don't care.

(52:33):
They're all talented, the ability to be able to move
intel quickly. And I'll tell you the longer you're on
the platform, Sheriff, the better these guys are getting at
understanding behavioral threats science, because that is a big difference
between seeing ten alerts and seeing a thousand alerts at
the end of twelve months, Man, do you learn a lot?

(52:55):
And I've had sheriffs call me under sheriffs call me
to Aaron. The signal is incredible. Man, Am I learning
a lot? It's all just kind of clicking and making sense.
And so Sheriff, I've got to educate five thousand agencies
here and I don't have a lot of time to
do it. I've got to move quick. It's a very
it's a very dangerous time right now. So the key

(53:17):
here is speed.

Speaker 4 (53:18):
Yeah. Absolutely, And it's really been foundationally important that the
sheriff that you county, that you reside in and you're
a part of that group and you've been able to
almost do a beta you know, with them, to be
able to show with time efficacy, so that's you're ahead
of that.

Speaker 2 (53:36):
I couldn't be more grateful for my sheriff FBI National
Academy grad forward thinking, investigator, minded, absolute rock star. He's
been great to me for twenty years since the first
training course I came out and gave to them to
onboarding the agency and I'm out there several times a

(53:56):
year and now to be able to stand up the
first Intelligence and counter Terrorism unit for the agency has
been tremendous, as you mentioned, to be able to really
wet test getting in to make sure that we're moving
through false positives. I mean, look, you know, uh uh,
if you don't have that analysis and the human element

(54:18):
and the ability to be able to test it in
a real environment, you're not gonna You're just gonna get
a ton of noise. So I mean, you can have
the word gun surfacing from from music lyrics, you can
have the word bomb resonating from uh from slang, you
know what I mean. So you really got to like
flush all this stuff out. So very grateful to have
my agency on board, Grateful for my sheriff and and

(54:40):
and my leadership and my captain and all of them.
They've been awesome.

Speaker 4 (54:44):
Arin, what's the biggest threat facing America today?

Speaker 2 (54:54):
For decades, we've invested heavily in systems designed to respond
after violence. Uh, it's import It'll always be. But the
Internet has changed the threat landscape, and today attackers leave
a behavioral trail before they mobilize, and so the opportunity
for public safety is learning how to identify that trail.
The biggest threat today is the Internet. It's the largest

(55:15):
threat intelligence environment in the world. If you're law enforcement,
you've got to start orienting towards that threat space. But
it's doable. Coming from Israel, I can tell you creativeness
is the is the key to solve all problems, and
that's what we're doing. Join my network and let's stop

(55:38):
the next one together.

Speaker 4 (55:41):
So important. An Is there any way people can follow
your work.

Speaker 2 (55:45):
Erin, Yeah, you can if you guys are interested. If
there's any law enforcement out there listening to me, you
want to get on a pilot, just get me on LinkedIn,
Just DM me directly and I'll have the head of
my agency partnerships reach out to you guys, will get
you on a platform super quick. Uh LinkedIn, Aaron Cohen,
get me up there and I'll communicate with you directly,

(56:08):
no where EGO, just ping me and uh and we'll
get you guys set up and on the platform and upstream.
Uh immediately. And I look forward to you guys reaching out.

Speaker 4 (56:17):
Aaron, this is one of the most important concepts in
law enforcement and recreciare I thank you very much for
coming on the show and being a part of it
and sharing your mission with us. So thank you very
much for coming.

Speaker 2 (56:28):
Here, Jariff, I appreciate you. Thank you for having me,
and I'm sure we'll talk again soon. Sounds good, but
we're gonna share some case studies soon.

Speaker 4 (56:36):
Yeah, it'd be great. So, folks, thanks so much for
watching the show. Check out Aaron's work if you're law
enforcement following, and check out my work at doctor Curiemeiers
dot substack dot com. Stay safe and we will see
you next week.

Speaker 2 (56:52):
Shadows see this amer and co Squam analogist calling to
the Hill.