December 12, 2024 • 29 mins
In this episode of Trading Tomorrow – Navigating Trends in Capital Markets, host Jim Jockle sits down with Jennifer Arnold, co-founder and CEO of Minerva, to explore the future of compliance and the evolving role of technology in combating financial crime. Jennifer shares her journey from anti-money laundering expert to RegTech innovator and unveils how Minerva uses technology and automation to revolutionize financial crime detection. This episode delves into the intersection of innovation, regulatory challenges, and the human role in technology-driven compliance.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:06):
Welcome to Trading Tomorrow Navigating Trends in
Capital Markets the podcastwhere we deep dive into
technologies reshaping the worldof capital markets.
I'm your host, jim Jockle, aveteran of the finance industry
with a passion for thecomplexities of financial
technologies and market trends.
In each episode, we'll explorethe cutting-edge trends, tools
and strategies driving today'sfinancial landscapes and paving
(00:29):
the way for the future.
With the finance industry at apivotal point, influenced by
groundbreaking innovations, it'smore crucial than ever to
understand how thesetechnological advancements
interact with market dynamics.
Today, we're excited to haveJennifer Arnold, co-founder and
(00:54):
CEO of Minerva, joining us.
Jennifer is a visionary in theregtech space with a passion for
leveraging advanced technologyto combat financial crime.
Her career began incommunications before
transitioning to anti-moneylaundering, risk and compliance.
As an expert in AML, she hasled large-scale transformations
at major financial institutions.
She is now at the forefront ofusing AI and automation to drive
(01:17):
innovation in financial crimedetection and compliance.
Minerva, her brainchild, issetting a new standard for
efficiency and effectiveness inAML by leveraging deep learning
and predictive intelligence tostay ahead of bad actors.
In today's episode, we'lldiscuss how Jennifer is changing
the landscape of financialcrime prevention and exploring
the broader trends in regtechand compliance.
(01:39):
Jennifer, welcome to the show.
Speaker 2 (01:41):
Thank you for having me.
Speaker 1 (01:42):
Perhaps you can share with Spark the idea behind
Minerva and how your backgroundin AML influenced its creation.
Speaker 2 (01:48):
Oh, my goodness, yeah , so really it started.
The idea for Minerva startedpercolating probably back in
2014, 2013.
I was working on animplementation of a product
called Oracle Mantis for capitalmarkets teams, so this is a
very large transactionmonitoring case management tool,
(02:09):
and I was spending a lot oftime with the investigators
working on mapping, you know,current process into new process
, et cetera, et cetera, and theinvestigators were reluctant to
really engage in the project ofthe design of this new tool that
they were going to be using andas I, you know, kept digging
(02:30):
with them to figure out what itwas.
Why were they holding back onme?
It was really this notion thatwe were spending, you know,
millions and millions andmillions of dollars to buy this
tool that was going to createalerts and cases more quickly
than they've ever been createdbefore for this particular team,
but nothing else in their worldhad changed.
So they still consulted, youknow, like five to seven
(02:55):
internal systems and vendorsolutions.
They still executed multiplesupon multiples of Google
searches.
They still spent most of theirday copying, pasting data from
various sources back into a casedocument, and so, as they sort
of looked forward into what wascoming for them, it was way more
work being produced much morequickly, but nothing that they
(03:19):
were using to actually do.
The investigation and the riskassessment piece was evolving at
the same rate as the tool thatwas coming in, and so they were
very anxious about beingoverwhelmed.
For me, I was like, oh, this isthe right right, this is an
incredibly painful problem.
And then, as I spent more timein the space and moved to my
(03:42):
next bank and spent more timewith investigators, the problem
seemed fairly universal.
There was this truckload ofmanual and fairly menial work
that needed to be done incompleting an investigation,
including everything from therisk assessment to the
documentation and ensuring thatyou've complied with your
(04:03):
internal policies and you've metyour regulatory requirements,
but most of it didn't have muchto do with the risk assessment
risk analysis part of their job,and I just started thinking
about ways that we might thinkof giving them more time to be
risk professionals and do riskanalysis and do higher order
work than do all of this busywork that comes with the
(04:24):
investigation process.
Speaker 1 (04:26):
So now you've mentioned in the past that
compliance teams havehistorically had to choose
between efficiency andeffectiveness in their tools, so
how does Minerva address thischallenge?
Speaker 2 (04:38):
Yeah, so I think we think about it a couple of
different ways, right?
So if we can move the acceleratethe investigation process for
that analyst, they will getthrough more work much more
quickly, with the emphasis beingon risk assessment versus data
gathering and copy and paste.
(04:59):
So that's already a win in termsof the effectiveness of the
program, because their time isbeing spent on the effectiveness
of the analysis that they'redoing.
Are they making the rightdecision about the customer in
that moment, given theinformation that they have in
front of them?
And then I think when we canmove them closer to working in a
near real-time type paradigm,we have a much better chance of
(05:23):
affecting real change Instead of.
You know, as many of us do inthe financial services, we're
sitting on several months worthof backlogs, of alerts and cases
that won't get adjudicated foranother few months.
So you know, an incident couldhave happened but it might not
actually make it anywhere intothe investigator's hands from
(05:46):
six months to a year afterward.
So being able to help them gofaster, being able to take away
the busy work and have themfocus on risk assessment,
tackles both the productivityand the effectiveness piece in
our view.
Speaker 1 (06:00):
The systems are now producing so many more alerts,
right, that require follow-upand you know, and all of these
other manual processes have notnecessarily caught up.
You know how are investigatorsprioritizing and you know to
what extent are.
You know, are there any falsepositives or are there?
(06:21):
You know what's theprioritization escalation look
like.
Speaker 2 (06:26):
Yeah, that's a great question and there's a lot of
threads in there to pull on.
So if we take the first thread,being false positives, of
course that is a challenge, likeindustry-wide.
That applies, you know, to namescreening, to transaction
monitoring, et cetera, et cetera.
For us, when we do riskscreening on the entity or
(06:47):
individual, the client itself,right, and taking a look at
their risk, we are using a very,I would say, complex matrix of
data to better and moreaccurately identify the client
in the first place.
So we can avoid some of thosefalse positives just avoid
producing them by actuallygetting better at identifying
(07:09):
the actual client in playinstead of everyone with a name
who sounds alike, spells alike,etc.
Speaker 1 (07:16):
So, jennifer, you know, one question I have is you
know we're dealing withreal-time data.
You know one could assume vastamount of data, but within those
things that are getting flaggedand elevated and the volume
that you're speaking about, youknow, to what extent are you
potentially seeing falsepositives or how are
investigators potentiallyprioritizing with such an
(07:39):
increased volume?
Speaker 2 (07:42):
Yeah.
So there's quite a few threadsin there and I'll just tug on a
few of them.
So the question around falsepositive every transaction
monitoring system, every namescreening engine, is spitting
out false positives and that hasa lot to do with the parameters
and thresholds that are set,you know, low enough to let the
to to gather up, likestatistical material, data and
(08:05):
and not leave anybody out whoyou might want to look at.
I think we just startedthinking about that problem in a
different way, which is how dowe get better at identifying the
actual customer so that whenthe analyst is looking at their
information on the screen, theyactually know that's their
customer and that this is thework worth doing versus a false
(08:26):
positive and it's somebody whohas the same name but is not the
same DOB, is not the sameaddress, has no other
identifiers.
So we really think about theaccuracy and the frankly, the
cogency of the profile thatwe're providing back to our
users to help them betteridentify their own clients so
they can get through the workmore quickly.
(08:46):
False positives are a challenge,I think, everywhere.
You know.
In some legacy systems thoseare based on keyword and name
matching, which of course youknow would naturally have to
generate a lot of falsepositives if they're doing their
job.
But you know it's problematicbecause the volume work that
gets created, you knowtransaction monitoring is the
(09:07):
forever job.
Right, you observe customerpatterns and transactional
behavior and you tune your rules, but you have to keep going
back to the data and seeing ifthe parameters are moving to
make sure that you're actuallylooking at the behavior that is
material for whatever theproduct or service it is that
you're monitoring.
So that's, I mean it's a wholelot of work.
(09:30):
We really tackle it by tryingnot to create false positives,
through better identification ofthe target and really, you know
, acting as a co-pilot to thatinvestigator or analyst so they
can get their work done.
And then you talk aboutprioritization.
Again, it depends on theorganization and the complexity
(09:51):
of the systems that they'reusing.
In my perfect sort of nirvanaworld, an alert arrives fully
born into the hands of ananalyst and they see the
transactional behavior thatcaused the alert to be triggered
.
And then they see all thecontextual data around the
client that tells me who theyare, how long they've been a
client, who are they connectedto, where does their money come
(10:15):
from, and then I'm able to,either manually or in an
automated fashion, risk rankthose alerts and tackle them
that way.
That's what I would like to seeis to see those data sets come
together in a really materialway.
Right now it's incrediblydifficult For some organizations
.
A manager will go in and tryand triage some of those and
(10:35):
they'll use transactional datato try and identify if a
transaction is higher risk thananother.
So it would take priority inthe queue.
But there isn't a greatsolution out there.
I think our answer is gettingthe whole picture of the client,
transactional behavior,contextual information and their
(10:57):
KYC data in one place so theanalyst can go wow, does this
make sense for what I know aboutthis client, or is this utter
nonsense?
Speaker 1 (11:05):
So you know, Jennifer , you spoke about your start and
almost a resistance.
Right, you know, now you'regetting transactional data, new
speeds, new velocity as itrelates to data information yet
the industry is still dealingwith multiple systems, manual
processes.
Fast forward to today, 2024,2025, where is the industry now?
Speaker 2 (11:27):
Oh well, that's actually an excellent question.
Where is the industry now?
Oh well, that's actually anexcellent question.
So I think it is.
The industry is moving forward,but I would say that, like the
industry from an AMLprofessionals and financial
services provider, exists on aspectrum right Some who are
still fully 100% legacyproviders, some who are really
(11:48):
looking ahead and trying tofigure out how to future-proof
their businesses by employingmore advanced technologies like
an applied AI to help themunderstand their data more
quickly so they can make betterdecisions more quickly about
their clients.
Speaker 1 (12:02):
So, Jennifer, can you share a real-world scenario in
which your software has actuallycombated crime?
Speaker 2 (12:08):
Yeah, actually, I think we're just coming out with
a case study and I think I'mallowed to talk about it.
We have a client that we sharewith our partners at Equifax and
they're one of the largest sortof like luxury leasing
automobile folks out there andyou know their challenge was
(12:28):
trying to avoid all the manualwork that was slowing down their
ability to sell, and part ofthat was meeting these
compliance requirements thatthey had to respond to prevent
(12:52):
losses from happening in thefirst place by simply looking at
adverse media for some of theleasing applicants that were
coming in to get some very highvalue vehicles.
Another one which I think islike near and dear to me is we
work with an organization, ananti-human trafficking
organization in Canada andanother one here in the States,
but the one in Canada we ran anoperation because when we have
(13:14):
big events like Super Bowl, filmfestival et cetera, it attracts
a lot of people from a lot ofdifferent places.
Unfortunately, it also attractsa lot of traffickers and
through the use of our data andsome really great volunteers, we
were able to help extract twoyoung women, two girls, who had
been trafficked during the filmfestival over the border.
Speaker 1 (13:37):
Wow, that's amazing.
It is Probably feels good to goto work every day, knowing that
you're changing lives.
Speaker 2 (13:44):
You know what?
It's really really cool.
It's really overwhelmingsometimes because we often talk
in the abstract about thisindustry and what a pain it is
and the regulators, and it's acheckbox exercise and blah, blah
blah.
But the purpose of the work.
The purpose of the work is toprotect our financial
infrastructure and the purposeof that work is to protect
(14:06):
everyone who lives in thesecountries.
So you know why not do thatwork?
Well, Amazing.
Speaker 1 (14:11):
Thank you for sharing those stories.
So you know I'd be remiss inthis podcast if I didn't mention
AI it's.
You know our listeners areprobably like wow, he made it 12
minutes without mentioning it.
But AI is becoming moreprevalent in compliance programs
.
You know what are some of theunique ways that Minerva is
using AI in deep learning.
Speaker 2 (14:32):
Yeah, so Minerva is an AI-native platform.
When we built her, she wasbuilt as an AI to solve exactly
this problem.
So Minerva's AI really comesinto play in three places, and
we call ourselves an applied AI,which is our customers, their
regulators, et cetera can take alook at Minerva.
(14:54):
They can see all the data thatgoes in, they can see the data
transformation activities andthey can see the outputs.
And we provide data lineage forevery single piece of
information that we tap intowhen we're doing a risk
assessment, and that's becausethat's what regulators need to
be comfortable and, therefore,that's what our customers need
to have in order to be able touse AI inside their four walls.
(15:17):
So we use different types of AI.
We use, obviously, a lot ofnatural language processing to
help us understand things likecontext and sentiment at risk.
When we're looking at profileinformation, we use what Damien,
our CTO.
We have an insane entityresolution engines that really
(15:43):
help Minerva understand whichpieces of data belong to which.
Jennifer Arnold, right, like,if you Google Jennifer Arnold,
you'll find out that there'smillions of us, but if I'm only
looking for one, how do I dothat?
And that's what Damien hasbuilt.
That is the big, I think kindof the big win on the Minerva
side is, instead of having ananalyst go through a fraction of
(16:03):
the amount of informationtrying to figure out which,
jennifer Arnold, the databelongs to, we can do it in a
few seconds for an analyst andget them on the right track.
Speaker 1 (16:14):
So what would you say are some of the biggest trends
that you're seeing right now inrec tech?
Speaker 2 (16:19):
Okay, I'm going to sound like an old grump, but I'm
just going to say it.
I know everyone is very excitedabout Gen AI and I'll be honest
, when Gen AI blew up last year,we removed AI from our name
because people were gettingreally excited, but possibly not
for the right reason.
So Gen AI can be reallypowerful, but I have two
(16:39):
concerns One, which is ourregulators are not yet
comfortable with most Gen AIbecause it's fairly black box.
They can't see the data goingin, they can't see the
transformation, they can'trelate it to the output.
That's a problem from aregulatory perspective and so it
will be a problem for ourclients.
The other part is there's somereally cool stuff going on where
(17:02):
the Gen AI is being used tohelp create the narrative part
of a regulatory filing called aSAR or an STR, and this is
really where the analyst has totell a story.
I am filing a SAR today becauseand they talk about the
activity and they talk about theprofile, et cetera, et cetera,
(17:23):
and sort of put together all theinformation in narrative form
for the regulator, and so Jennyand I are being deployed into
doing this piece of work.
Why?
Because it can be timeintensive, I think where I'm
challenged is it's at thismoment, where the investigator
sits down to think, to tell thatstory, where we need their
(17:44):
critical thinking skills themost, and so bypassing them here
instead of accelerating, youknow, the kind of grunt work
that needs to happen up front.
I'm not sure.
I'm not sure it's the rightmove.
It makes me a little nervousand, you know, like, the reason
there's a human in the loop isbecause we need the critical
(18:06):
thinking skills there to say,well, actually, that doesn't
make any sense and it's notright.
So that's how I, that's how Ifeel about that, and we see a
lot of that going on in theindustry.
And then there's just the, youknow, data privacy and security
concerns.
Like, if you're using an opensource model, how are you
(18:26):
protecting your customer's dataas it gets sent back to
something you know outside ofyour four walls as an
organization, being processed bya machine that, in theory, many
other people, hundreds ofmillions of people, have access
to?
Right, so how do we protectagainst that?
And private LLMs are a greatsolution for that, and there's
(18:47):
some amazing private LLMorganizations out there.
But that's how I think aboutthat, right, like, I love the
idea of AI, accelerating theprocess, identifying real risk,
being able to differentiatebetween low and medium and high,
et cetera.
There's a lot of value in doingthat, and that's the part that
I'm most excited about.
Speaker 1 (19:08):
You know, one thing that always fascinates me is the
people component on this, andthere's some people out there
that AI is going to take awayeverybody's job and automation
and robots and great buildfactories in the US, but you
don't need humans to work inthem, but you raise the critical
(19:30):
thinking of humans.
You know how is the job of aninvestigator evolved or the
skill set required to do thatjob with.
You know the data andinformation, that's coming at
them now.
Speaker 2 (19:47):
Yeah, like, the best investigators will be those with
the data science background,right, like that's.
I think that's the, that's themagic sauce there.
You know, the the human in theloop for the investigation
process is really important.
You know, one, for some optictype reasons, like it makes
regulators more comfortable ifthey know that a human has
(20:08):
looked at it.
A human has looked at it.
And two, I just there's.
You know, ai, in my view, is anaugmentation to human capacity.
It is not a replacement ofhuman capacity.
So, if you know, minerva canprocess four and a half billion
disparate data points into asingle profile in under 20
(20:29):
seconds.
A human brain can't do that.
What a human brain can do islook at that profile that
Minerva has assembled and say,yo, that's nonsense, this
doesn't make sense, this doesn'tmake sense.
Oh, but this is the gem righthere, and be able to start their
investigation that way.
That's meaningful.
Speaker 1 (20:47):
So how do you balance embracing innovation right now?
I mean, obviously there's somany new technologies.
The pace of evolution is rapid.
The changes in AI seem, youknow, exponential month over
month in terms of what's comingto market.
(21:10):
You know the speed of whichtransactional data is being
processed.
You know how are you balancinginnovation within your own
solution, but also, you knowdealing with complex regulations
.
You know organizationalmovement, you know, and demand
(21:30):
in terms of, you know, keepingpace with that innovation.
How do you manage the balance?
Speaker 2 (21:37):
Yeah, this is going to be.
It might be disappointing and abit pedantic of a response, but
I kind of go back to standardprogram management protocols
when I think about this, whichis innovation for the sake of
innovation, fun, not superuseful.
So, in our space specifically,what is the sandbox that we're
allowed to play in?
What are the guardrails that wehave to stay in to not get
(22:00):
ourselves into trouble with theregulator or create any
unnecessary risks for ourcustomers?
And then what is the primaryuse case for that innovation and
what is the value add?
Is it 10x, is it 100x?
Then let's talk about what thatactually means and what that
actually looks like.
And then let's start talkingabout not just the comparison to
(22:22):
the before and after of the useof whatever the innovation is
going to be, but then what doesit do for future state?
What are the other knock-oneffects good and bad for that
organization or for us if weproceed down that road?
I kind of run it through thatlittle matrix in my brain of
what's its purpose, what problemis it solving, and is it big
(22:44):
enough?
Is the problem big enough to besolved this way?
Am I taking a bazooka to goafter a mosquito, because then
that's just comedy, right Likethat's hubris, and we'll see
some of that.
Speaker 1 (22:57):
So you know.
The one thing I do wonder isyou know so?
Every fall a new iPhone comesout right and you know there's
people who will be online at theApple store day of release.
And then there's people who areon generation eight and they're
perfectly fine, and and they'reeither fine because the tech is
good and it works and and Imake phones and I get texts, or
(23:20):
there's a barrier to cost whichis preventing upgrades.
You know where?
Where's the client base rightnow?
Are they?
Are they?
Are they standing outside theapplesauce store or do they got
flip phones and saying yeah,yeah, I'm good, so interesting.
Speaker 2 (23:35):
It really depends on the client.
So if I look at our, you knowthe folks we spend the most time
with.
So let's think, like mid-marketfintechs, neobanks they are a
mix of the people standingoutside of the iPhone store and
then their slightly jealousfriend who's looking over the
shoulder thinking maybe theyshould make a move.
(23:56):
When we talk to certainly olderand more established financial
services providers financialinstitutions you know there's a
lot of knowledge and experiencethere and they are much slower
to move because the cost of thechange can be, or is perceived
(24:17):
to be, prohibitive.
It's not that the new solutionwould cost more.
It's that changing from thesystem that they have is a
high-risk move.
When you're moving data youknow that old from one system to
another it can be quite, quitechallenging.
Speaker 1 (24:35):
Got it and you know.
One thing I do have to ask isyou know, obviously financial
regulations, you know, arecontinuing to evolve.
Some would say rapidly, somewould say too slowly, but
there's different camps on allof that and we'll see, with you
know, new regime changes inpresidential politics, what
happens there.
But we know technology ischanging.
So you know where do you seethe compliance landscape heading
(24:59):
over the next five years?
Speaker 2 (25:01):
Well, you make a really excellent point about the
US, specifically because youknow, if I think, if Trump is to
be believed, trump, I think, ifTrump is to be believed, the US
might be moving in the oppositedirection of many of its peer
nations no-transcript skills tobuild the kinds of regulatory
(25:46):
frameworks, et cetera, that theyneed.
My primary, I guess, focus orconcern right now is I hear a
lot of organizations saying,well, we're just going to wait
to see what the regulator does,and I get that Like it's a solid
business choice, because whywould you go and make an
investment in something if youdon't know you're actually going
to be forced to do it.
And then I think about it fromthe other side, which is if we
(26:11):
agree that money laundering isactually attack on the nation's
sovereignty, which it is rightIf we look at Russia, china,
iran and the cyber activity thatwe see there.
So, if we agree that moneylaundering is an attack on our
sovereignty, why can't theindustry, why can't tier one
(26:35):
banks, why can't credit unions,why can't you know fintechs,
crypto, defi, as their owncommunities, move in lockstep on
some very simple, very low costimprovements that would
strengthen the integrity of thefinancial system overall
improvements that wouldstrengthen the integrity of the
financial system overall.
They don't need to be told by aregulator to do it, but if they
(26:55):
agree among themselves thatthere are some things they could
be doing better, for example,if everyone agreed that, yes,
they would do adverse media atonboarding, then it's not a
competitive situation.
Right, then it's not.
We're not.
It costs us more to onboardthan it costs you to onboard.
We don't end up in that kind ofdiscussion, though.
(27:16):
No one ever really talks aboutthe expense of offboarding a
client, especially after they'vebeen found to be laundering
money through your organization.
Speaker 1 (27:27):
So you know, sadly, jennifer, we've come to the last
question of the podcast.
We call it the trend drop.
It's like a desert islandquestion.
So you know, if you could onlywatch or track one trend in reg
tech and AI, you know what wouldit be.
Speaker 2 (27:43):
Oh man, the regulatory thinking around how
data is being used to performAML risk assessment and how it
may or may not be in conflictwith privacy law in some
jurisdictions, and how will weresolve that?
Speaker 1 (28:04):
Well as someone who's in marketing.
We talk about privacy lawsevery day.
I bet you do?
Speaker 2 (28:09):
Yeah, because some of that bank data like if you're
looking for a bank and you're inmarketing, they've got tons of
information.
They can share a sliver of itwith the marketing team, right,
and that's appropriate.
Speaker 1 (28:20):
And even, as you were saying before, in terms of how
certain transaction data isgetting processed.
I'm like oh my God, this is PII.
Speaker 2 (28:31):
Yeah, sorry.
And often organizations willsay, oh, we can't do that, it's
a privacy issue.
Oh, we can't share thatinformation as a privacy issue.
They really need to go talk totheir legal team, because often
it isn't a privacy issue, itjust feels like one.
And so you've got to check yourfacts.
Speaker 1 (28:46):
Well, that's good advice to any listener right,
Bringing in an angle we've neverdiscussed before.
So, Jennifer, I want to thankyou so much for your time, your
insight, really enjoyed ourconversation.
Speaker 2 (28:57):
Thank you so much for having me and thank you for
letting me ramble on about AML.
Speaker 1 (29:09):
Thank you so much.
Thanks so much for listening totoday's episode, and if you're
enjoying Trading Tomorrow,navigating trends and capital
markets, be sure to like,subscribe and share, and we'll
see you next time.
Welcome to Trading Tomorrow Navigating Trends in
Capital Markets the podcastwhere we deep dive into
technologies reshaping the worldof capital markets.
I'm your host, jim Jockle, aveteran of the finance industry
with a passion for thecomplexities of financial
technologies and market trends.
In each episode, we'll explorethe cutting-edge trends, tools
and strategies driving today'sfinancial landscapes and paving
(00:29):
the way for the future.
With the finance industry at apivotal point, influenced by
groundbreaking innovations, it'smore crucial than ever to
understand how thesetechnological advancements
interact with market dynamics.
Today, we're excited to haveJennifer Arnold, co-founder and
(00:54):
CEO of Minerva, joining us.
Jennifer is a visionary in theregtech space with a passion for
leveraging advanced technologyto combat financial crime.
Her career began incommunications before
transitioning to anti-moneylaundering, risk and compliance.
As an expert in AML, she hasled large-scale transformations
at major financial institutions.
She is now at the forefront ofusing AI and automation to drive
(01:17):
innovation in financial crimedetection and compliance.
Minerva, her brainchild, issetting a new standard for
efficiency and effectiveness inAML by leveraging deep learning
and predictive intelligence tostay ahead of bad actors.
In today's episode, we'lldiscuss how Jennifer is changing
the landscape of financialcrime prevention and exploring
the broader trends in regtechand compliance.
(01:39):
Jennifer, welcome to the show.
Speaker 2 (01:41):
Thank you for having me.
Speaker 1 (01:42):
Perhaps you can share with Spark the idea behind
Minerva and how your backgroundin AML influenced its creation.
Speaker 2 (01:48):
Oh, my goodness, yeah , so really it started.
The idea for Minerva startedpercolating probably back in
2014, 2013.
I was working on animplementation of a product
called Oracle Mantis for capitalmarkets teams, so this is a
very large transactionmonitoring case management tool,
(02:09):
and I was spending a lot oftime with the investigators
working on mapping, you know,current process into new process
, et cetera, et cetera, and theinvestigators were reluctant to
really engage in the project ofthe design of this new tool that
they were going to be using andas I, you know, kept digging
(02:30):
with them to figure out what itwas.
Why were they holding back onme?
It was really this notion thatwe were spending, you know,
millions and millions andmillions of dollars to buy this
tool that was going to createalerts and cases more quickly
than they've ever been createdbefore for this particular team,
but nothing else in their worldhad changed.
So they still consulted, youknow, like five to seven
(02:55):
internal systems and vendorsolutions.
They still executed multiplesupon multiples of Google
searches.
They still spent most of theirday copying, pasting data from
various sources back into a casedocument, and so, as they sort
of looked forward into what wascoming for them, it was way more
work being produced much morequickly, but nothing that they
(03:19):
were using to actually do.
The investigation and the riskassessment piece was evolving at
the same rate as the tool thatwas coming in, and so they were
very anxious about beingoverwhelmed.
For me, I was like, oh, this isthe right right, this is an
incredibly painful problem.
And then, as I spent more timein the space and moved to my
(03:42):
next bank and spent more timewith investigators, the problem
seemed fairly universal.
There was this truckload ofmanual and fairly menial work
that needed to be done incompleting an investigation,
including everything from therisk assessment to the
documentation and ensuring thatyou've complied with your
(04:03):
internal policies and you've metyour regulatory requirements,
but most of it didn't have muchto do with the risk assessment
risk analysis part of their job,and I just started thinking
about ways that we might thinkof giving them more time to be
risk professionals and do riskanalysis and do higher order
work than do all of this busywork that comes with the
(04:24):
investigation process.
Speaker 1 (04:26):
So now you've mentioned in the past that
compliance teams havehistorically had to choose
between efficiency andeffectiveness in their tools, so
how does Minerva address thischallenge?
Speaker 2 (04:38):
Yeah, so I think we think about it a couple of
different ways, right?
So if we can move the acceleratethe investigation process for
that analyst, they will getthrough more work much more
quickly, with the emphasis beingon risk assessment versus data
gathering and copy and paste.
(04:59):
So that's already a win in termsof the effectiveness of the
program, because their time isbeing spent on the effectiveness
of the analysis that they'redoing.
Are they making the rightdecision about the customer in
that moment, given theinformation that they have in
front of them?
And then I think when we canmove them closer to working in a
near real-time type paradigm,we have a much better chance of
(05:23):
affecting real change Instead of.
You know, as many of us do inthe financial services, we're
sitting on several months worthof backlogs, of alerts and cases
that won't get adjudicated foranother few months.
So you know, an incident couldhave happened but it might not
actually make it anywhere intothe investigator's hands from
(05:46):
six months to a year afterward.
So being able to help them gofaster, being able to take away
the busy work and have themfocus on risk assessment,
tackles both the productivityand the effectiveness piece in
our view.
Speaker 1 (06:00):
The systems are now producing so many more alerts,
right, that require follow-upand you know, and all of these
other manual processes have notnecessarily caught up.
You know how are investigatorsprioritizing and you know to
what extent are.
You know, are there any falsepositives or are there?
(06:21):
You know what's theprioritization escalation look
like.
Speaker 2 (06:26):
Yeah, that's a great question and there's a lot of
threads in there to pull on.
So if we take the first thread,being false positives, of
course that is a challenge, likeindustry-wide.
That applies, you know, to namescreening, to transaction
monitoring, et cetera, et cetera.
For us, when we do riskscreening on the entity or
(06:47):
individual, the client itself,right, and taking a look at
their risk, we are using a very,I would say, complex matrix of
data to better and moreaccurately identify the client
in the first place.
So we can avoid some of thosefalse positives just avoid
producing them by actuallygetting better at identifying
(07:09):
the actual client in playinstead of everyone with a name
who sounds alike, spells alike,etc.
Speaker 1 (07:16):
So, jennifer, you know, one question I have is you
know we're dealing withreal-time data.
You know one could assume vastamount of data, but within those
things that are getting flaggedand elevated and the volume
that you're speaking about, youknow, to what extent are you
potentially seeing falsepositives or how are
investigators potentiallyprioritizing with such an
(07:39):
increased volume?
Speaker 2 (07:42):
Yeah.
So there's quite a few threadsin there and I'll just tug on a
few of them.
So the question around falsepositive every transaction
monitoring system, every namescreening engine, is spitting
out false positives and that hasa lot to do with the parameters
and thresholds that are set,you know, low enough to let the
to to gather up, likestatistical material, data and
(08:05):
and not leave anybody out whoyou might want to look at.
I think we just startedthinking about that problem in a
different way, which is how dowe get better at identifying the
actual customer so that whenthe analyst is looking at their
information on the screen, theyactually know that's their
customer and that this is thework worth doing versus a false
(08:26):
positive and it's somebody whohas the same name but is not the
same DOB, is not the sameaddress, has no other
identifiers.
So we really think about theaccuracy and the frankly, the
cogency of the profile thatwe're providing back to our
users to help them betteridentify their own clients so
they can get through the workmore quickly.
(08:46):
False positives are a challenge,I think, everywhere.
You know.
In some legacy systems thoseare based on keyword and name
matching, which of course youknow would naturally have to
generate a lot of falsepositives if they're doing their
job.
But you know it's problematicbecause the volume work that
gets created, you knowtransaction monitoring is the
(09:07):
forever job.
Right, you observe customerpatterns and transactional
behavior and you tune your rules, but you have to keep going
back to the data and seeing ifthe parameters are moving to
make sure that you're actuallylooking at the behavior that is
material for whatever theproduct or service it is that
you're monitoring.
So that's, I mean it's a wholelot of work.
(09:30):
We really tackle it by tryingnot to create false positives,
through better identification ofthe target and really, you know
, acting as a co-pilot to thatinvestigator or analyst so they
can get their work done.
And then you talk aboutprioritization.
Again, it depends on theorganization and the complexity
(09:51):
of the systems that they'reusing.
In my perfect sort of nirvanaworld, an alert arrives fully
born into the hands of ananalyst and they see the
transactional behavior thatcaused the alert to be triggered
.
And then they see all thecontextual data around the
client that tells me who theyare, how long they've been a
client, who are they connectedto, where does their money come
(10:15):
from, and then I'm able to,either manually or in an
automated fashion, risk rankthose alerts and tackle them
that way.
That's what I would like to seeis to see those data sets come
together in a really materialway.
Right now it's incrediblydifficult For some organizations
.
A manager will go in and tryand triage some of those and
(10:35):
they'll use transactional datato try and identify if a
transaction is higher risk thananother.
So it would take priority inthe queue.
But there isn't a greatsolution out there.
I think our answer is gettingthe whole picture of the client,
transactional behavior,contextual information and their
(10:57):
KYC data in one place so theanalyst can go wow, does this
make sense for what I know aboutthis client, or is this utter
nonsense?
Speaker 1 (11:05):
So you know, Jennifer , you spoke about your start and
almost a resistance.
Right, you know, now you'regetting transactional data, new
speeds, new velocity as itrelates to data information yet
the industry is still dealingwith multiple systems, manual
processes.
Fast forward to today, 2024,2025, where is the industry now?
Speaker 2 (11:27):
Oh well, that's actually an excellent question.
Where is the industry now?
Oh well, that's actually anexcellent question.
So I think it is.
The industry is moving forward,but I would say that, like the
industry from an AMLprofessionals and financial
services provider, exists on aspectrum right Some who are
still fully 100% legacyproviders, some who are really
(11:48):
looking ahead and trying tofigure out how to future-proof
their businesses by employingmore advanced technologies like
an applied AI to help themunderstand their data more
quickly so they can make betterdecisions more quickly about
their clients.
Speaker 1 (12:02):
So, Jennifer, can you share a real-world scenario in
which your software has actuallycombated crime?
Speaker 2 (12:08):
Yeah, actually, I think we're just coming out with
a case study and I think I'mallowed to talk about it.
We have a client that we sharewith our partners at Equifax and
they're one of the largest sortof like luxury leasing
automobile folks out there andyou know their challenge was
(12:28):
trying to avoid all the manualwork that was slowing down their
ability to sell, and part ofthat was meeting these
compliance requirements thatthey had to respond to prevent
(12:52):
losses from happening in thefirst place by simply looking at
adverse media for some of theleasing applicants that were
coming in to get some very highvalue vehicles.
Another one which I think islike near and dear to me is we
work with an organization, ananti-human trafficking
organization in Canada andanother one here in the States,
but the one in Canada we ran anoperation because when we have
(13:14):
big events like Super Bowl, filmfestival et cetera, it attracts
a lot of people from a lot ofdifferent places.
Unfortunately, it also attractsa lot of traffickers and
through the use of our data andsome really great volunteers, we
were able to help extract twoyoung women, two girls, who had
been trafficked during the filmfestival over the border.
Speaker 1 (13:37):
Wow, that's amazing.
It is Probably feels good to goto work every day, knowing that
you're changing lives.
Speaker 2 (13:44):
You know what?
It's really really cool.
It's really overwhelmingsometimes because we often talk
in the abstract about thisindustry and what a pain it is
and the regulators, and it's acheckbox exercise and blah, blah
blah.
But the purpose of the work.
The purpose of the work is toprotect our financial
infrastructure and the purposeof that work is to protect
(14:06):
everyone who lives in thesecountries.
So you know why not do thatwork?
Well, Amazing.
Speaker 1 (14:11):
Thank you for sharing those stories.
So you know I'd be remiss inthis podcast if I didn't mention
AI it's.
You know our listeners areprobably like wow, he made it 12
minutes without mentioning it.
But AI is becoming moreprevalent in compliance programs
.
You know what are some of theunique ways that Minerva is
using AI in deep learning.
Speaker 2 (14:32):
Yeah, so Minerva is an AI-native platform.
When we built her, she wasbuilt as an AI to solve exactly
this problem.
So Minerva's AI really comesinto play in three places, and
we call ourselves an applied AI,which is our customers, their
regulators, et cetera can take alook at Minerva.
(14:54):
They can see all the data thatgoes in, they can see the data
transformation activities andthey can see the outputs.
And we provide data lineage forevery single piece of
information that we tap intowhen we're doing a risk
assessment, and that's becausethat's what regulators need to
be comfortable and, therefore,that's what our customers need
to have in order to be able touse AI inside their four walls.
(15:17):
So we use different types of AI.
We use, obviously, a lot ofnatural language processing to
help us understand things likecontext and sentiment at risk.
When we're looking at profileinformation, we use what Damien,
our CTO.
We have an insane entityresolution engines that really
(15:43):
help Minerva understand whichpieces of data belong to which.
Jennifer Arnold, right, like,if you Google Jennifer Arnold,
you'll find out that there'smillions of us, but if I'm only
looking for one, how do I dothat?
And that's what Damien hasbuilt.
That is the big, I think kindof the big win on the Minerva
side is, instead of having ananalyst go through a fraction of
(16:03):
the amount of informationtrying to figure out which,
jennifer Arnold, the databelongs to, we can do it in a
few seconds for an analyst andget them on the right track.
Speaker 1 (16:14):
So what would you say are some of the biggest trends
that you're seeing right now inrec tech?
Speaker 2 (16:19):
Okay, I'm going to sound like an old grump, but I'm
just going to say it.
I know everyone is very excitedabout Gen AI and I'll be honest
, when Gen AI blew up last year,we removed AI from our name
because people were gettingreally excited, but possibly not
for the right reason.
So Gen AI can be reallypowerful, but I have two
(16:39):
concerns One, which is ourregulators are not yet
comfortable with most Gen AIbecause it's fairly black box.
They can't see the data goingin, they can't see the
transformation, they can'trelate it to the output.
That's a problem from aregulatory perspective and so it
will be a problem for ourclients.
The other part is there's somereally cool stuff going on where
(17:02):
the Gen AI is being used tohelp create the narrative part
of a regulatory filing called aSAR or an STR, and this is
really where the analyst has totell a story.
I am filing a SAR today becauseand they talk about the
activity and they talk about theprofile, et cetera, et cetera,
(17:23):
and sort of put together all theinformation in narrative form
for the regulator, and so Jennyand I are being deployed into
doing this piece of work.
Why?
Because it can be timeintensive, I think where I'm
challenged is it's at thismoment, where the investigator
sits down to think, to tell thatstory, where we need their
(17:44):
critical thinking skills themost, and so bypassing them here
instead of accelerating, youknow, the kind of grunt work
that needs to happen up front.
I'm not sure.
I'm not sure it's the rightmove.
It makes me a little nervousand, you know, like, the reason
there's a human in the loop isbecause we need the critical
(18:06):
thinking skills there to say,well, actually, that doesn't
make any sense and it's notright.
So that's how I, that's how Ifeel about that, and we see a
lot of that going on in theindustry.
And then there's just the, youknow, data privacy and security
concerns.
Like, if you're using an opensource model, how are you
(18:26):
protecting your customer's dataas it gets sent back to
something you know outside ofyour four walls as an
organization, being processed bya machine that, in theory, many
other people, hundreds ofmillions of people, have access
to?
Right, so how do we protectagainst that?
And private LLMs are a greatsolution for that, and there's
(18:47):
some amazing private LLMorganizations out there.
But that's how I think aboutthat, right, like, I love the
idea of AI, accelerating theprocess, identifying real risk,
being able to differentiatebetween low and medium and high,
et cetera.
There's a lot of value in doingthat, and that's the part that
I'm most excited about.
Speaker 1 (19:08):
You know, one thing that always fascinates me is the
people component on this, andthere's some people out there
that AI is going to take awayeverybody's job and automation
and robots and great buildfactories in the US, but you
don't need humans to work inthem, but you raise the critical
(19:30):
thinking of humans.
You know how is the job of aninvestigator evolved or the
skill set required to do thatjob with.
You know the data andinformation, that's coming at
them now.
Speaker 2 (19:47):
Yeah, like, the best investigators will be those with
the data science background,right, like that's.
I think that's the, that's themagic sauce there.
You know, the the human in theloop for the investigation
process is really important.
You know, one, for some optictype reasons, like it makes
regulators more comfortable ifthey know that a human has
(20:08):
looked at it.
A human has looked at it.
And two, I just there's.
You know, ai, in my view, is anaugmentation to human capacity.
It is not a replacement ofhuman capacity.
So, if you know, minerva canprocess four and a half billion
disparate data points into asingle profile in under 20
(20:29):
seconds.
A human brain can't do that.
What a human brain can do islook at that profile that
Minerva has assembled and say,yo, that's nonsense, this
doesn't make sense, this doesn'tmake sense.
Oh, but this is the gem righthere, and be able to start their
investigation that way.
That's meaningful.
Speaker 1 (20:47):
So how do you balance embracing innovation right now?
I mean, obviously there's somany new technologies.
The pace of evolution is rapid.
The changes in AI seem, youknow, exponential month over
month in terms of what's comingto market.
(21:10):
You know the speed of whichtransactional data is being
processed.
You know how are you balancinginnovation within your own
solution, but also, you knowdealing with complex regulations
.
You know organizationalmovement, you know, and demand
(21:30):
in terms of, you know, keepingpace with that innovation.
How do you manage the balance?
Speaker 2 (21:37):
Yeah, this is going to be.
It might be disappointing and abit pedantic of a response, but
I kind of go back to standardprogram management protocols
when I think about this, whichis innovation for the sake of
innovation, fun, not superuseful.
So, in our space specifically,what is the sandbox that we're
allowed to play in?
What are the guardrails that wehave to stay in to not get
(22:00):
ourselves into trouble with theregulator or create any
unnecessary risks for ourcustomers?
And then what is the primaryuse case for that innovation and
what is the value add?
Is it 10x, is it 100x?
Then let's talk about what thatactually means and what that
actually looks like.
And then let's start talkingabout not just the comparison to
(22:22):
the before and after of the useof whatever the innovation is
going to be, but then what doesit do for future state?
What are the other knock-oneffects good and bad for that
organization or for us if weproceed down that road?
I kind of run it through thatlittle matrix in my brain of
what's its purpose, what problemis it solving, and is it big
(22:44):
enough?
Is the problem big enough to besolved this way?
Am I taking a bazooka to goafter a mosquito, because then
that's just comedy, right Likethat's hubris, and we'll see
some of that.
Speaker 1 (22:57):
So you know.
The one thing I do wonder isyou know so?
Every fall a new iPhone comesout right and you know there's
people who will be online at theApple store day of release.
And then there's people who areon generation eight and they're
perfectly fine, and and they'reeither fine because the tech is
good and it works and and Imake phones and I get texts, or
(23:20):
there's a barrier to cost whichis preventing upgrades.
You know where?
Where's the client base rightnow?
Are they?
Are they?
Are they standing outside theapplesauce store or do they got
flip phones and saying yeah,yeah, I'm good, so interesting.
Speaker 2 (23:35):
It really depends on the client.
So if I look at our, you knowthe folks we spend the most time
with.
So let's think, like mid-marketfintechs, neobanks they are a
mix of the people standingoutside of the iPhone store and
then their slightly jealousfriend who's looking over the
shoulder thinking maybe theyshould make a move.
(23:56):
When we talk to certainly olderand more established financial
services providers financialinstitutions you know there's a
lot of knowledge and experiencethere and they are much slower
to move because the cost of thechange can be, or is perceived
(24:17):
to be, prohibitive.
It's not that the new solutionwould cost more.
It's that changing from thesystem that they have is a
high-risk move.
When you're moving data youknow that old from one system to
another it can be quite, quitechallenging.
Speaker 1 (24:35):
Got it and you know.
One thing I do have to ask isyou know, obviously financial
regulations, you know, arecontinuing to evolve.
Some would say rapidly, somewould say too slowly, but
there's different camps on allof that and we'll see, with you
know, new regime changes inpresidential politics, what
happens there.
But we know technology ischanging.
So you know where do you seethe compliance landscape heading
(24:59):
over the next five years?
Speaker 2 (25:01):
Well, you make a really excellent point about the
US, specifically because youknow, if I think, if Trump is to
be believed, trump, I think, ifTrump is to be believed, the US
might be moving in the oppositedirection of many of its peer
nations no-transcript skills tobuild the kinds of regulatory
(25:46):
frameworks, et cetera, that theyneed.
My primary, I guess, focus orconcern right now is I hear a
lot of organizations saying,well, we're just going to wait
to see what the regulator does,and I get that Like it's a solid
business choice, because whywould you go and make an
investment in something if youdon't know you're actually going
to be forced to do it.
And then I think about it fromthe other side, which is if we
(26:11):
agree that money laundering isactually attack on the nation's
sovereignty, which it is rightIf we look at Russia, china,
iran and the cyber activity thatwe see there.
So, if we agree that moneylaundering is an attack on our
sovereignty, why can't theindustry, why can't tier one
(26:35):
banks, why can't credit unions,why can't you know fintechs,
crypto, defi, as their owncommunities, move in lockstep on
some very simple, very low costimprovements that would
strengthen the integrity of thefinancial system overall
improvements that wouldstrengthen the integrity of the
financial system overall.
They don't need to be told by aregulator to do it, but if they
(26:55):
agree among themselves thatthere are some things they could
be doing better, for example,if everyone agreed that, yes,
they would do adverse media atonboarding, then it's not a
competitive situation.
Right, then it's not.
We're not.
It costs us more to onboardthan it costs you to onboard.
We don't end up in that kind ofdiscussion, though.
(27:16):
No one ever really talks aboutthe expense of offboarding a
client, especially after they'vebeen found to be laundering
money through your organization.
Speaker 1 (27:27):
So you know, sadly, jennifer, we've come to the last
question of the podcast.
We call it the trend drop.
It's like a desert islandquestion.
So you know, if you could onlywatch or track one trend in reg
tech and AI, you know what wouldit be.
Speaker 2 (27:43):
Oh man, the regulatory thinking around how
data is being used to performAML risk assessment and how it
may or may not be in conflictwith privacy law in some
jurisdictions, and how will weresolve that?
Speaker 1 (28:04):
Well as someone who's in marketing.
We talk about privacy lawsevery day.
I bet you do?
Speaker 2 (28:09):
Yeah, because some of that bank data like if you're
looking for a bank and you're inmarketing, they've got tons of
information.
They can share a sliver of itwith the marketing team, right,
and that's appropriate.
Speaker 1 (28:20):
And even, as you were saying before, in terms of how
certain transaction data isgetting processed.
I'm like oh my God, this is PII.
Speaker 2 (28:31):
Yeah, sorry.
And often organizations willsay, oh, we can't do that, it's
a privacy issue.
Oh, we can't share thatinformation as a privacy issue.
They really need to go talk totheir legal team, because often
it isn't a privacy issue, itjust feels like one.
And so you've got to check yourfacts.
Speaker 1 (28:46):
Well, that's good advice to any listener right,
Bringing in an angle we've neverdiscussed before.
So, Jennifer, I want to thankyou so much for your time, your
insight, really enjoyed ourconversation.
Speaker 2 (28:57):
Thank you so much for having me and thank you for
letting me ramble on about AML.
Speaker 1 (29:09):
Thank you so much.
Thanks so much for listening totoday's episode, and if you're
enjoying Trading Tomorrow,navigating trends and capital
markets, be sure to like,subscribe and share, and we'll
see you next time.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.