Financial advisors are beginning to leverage AI for efficiency and advanced applications, but many untapped opportunities remain. While some advisors are using AI for strategic business planning, compliance documentation, and multilingual capabilities, others are still exploring where it fits into their practice.
Hear Sean Bailey and Devin Kropp from Horsesmouth share insights on:
- How advisors are using AI for strategic business planning and what initiatives they are developing
- The workflows being built around AI-powered data extraction from documents
- The opportunities and risks of using AI for compliance documentation
- The potential of AI-driven multilingual capabilities and personalized financial planning
- The lowest adoption areas, such as investment policy development, and the untapped opportunities advisors might be missing
Get an in-depth look at how AI can drive greater efficiency, reduce manual workload, and create new possibilities for your practice.
Resources mentioned:
The AI-Powered Financial Advisor Program Virtual Workshop March 6, 13, 20, and 27 https://www.horsesmouth.com/ai
AI Marketing for Advisors Workshop
Virtual Workshop: April 3, 10, 17, 24 and May 1
https://www.horsesmouth.com/aimarketing
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
Welcome to today'sshow on AI for Practice Efficiency,
how advisors are streamlining workflowsand unlocking new opportunities.
I am is now seven cropand I'm joined by Sean Bailey, Horse's
Mouth editor in chief.
Today, Sean and I will be sharingsome early results from a survey
we have out in the fieldabout how advisors are using AI
(00:20):
to skyrocket their practices, efficiencyand pave the way for future growth.
So, Sean, thanks for joining me today.
We can jump right into it here.
The survey shows that about 12%
of advisors are using AI for strategicbusiness planning.
What specific initiativesdo you think an advisor
could kind of developusing AI in this way?
(00:42):
Yeah, it's a good question.And it's really fascinating.
And of course, you know,we're about to go through, some results
from our survey of use casesthat are on the lower end.
Right.
We've talked previously about,you know, using AI for content
development, for marketing,for preparing client meeting notes.
But this is a general purpose technology,meaning that it can be,
(01:04):
you know, applied, you know, across
nearly all domains and,
and within each domain,across nearly all types of job tasks.
And so I just wanted to put that out therefor background.
And we're going to be we're plumbingsome of the moment at this moment in time.
Some of the usage is that,have a lot of potential.
(01:27):
And in some advisors are doing thembased on our survey.
But, they're not widespread by,
by any means at this point in time.
Now, your question about,using AI for strategic
planning, is really,a good, it's a favorite topic of mine.
And I was listening to a podcast last weekwith, Satya Nadella,
(01:49):
who the CEO of Microsoft and I mentioned,I, I mangled his quote last week.
But it actually applies very muchto this, this question of strategy.
Right? Because we're not talking tacticsand things.
I was mentioning before about, you know,content development, client meeting notes,
these are all tactical kinds of thingsyou do within your business, right?
(02:09):
But now we're talking about strategy,which is kind of the bigger,
the bigger picture.
Looking ahead, trying to anticipate,the direction of your business,
what's going on in the market,what opportunities you have.
And I can be, you know, hugely.
Helpful in this way.
And Satya Nadella said, in this podcast,
(02:30):
he said the new workflow for me
is that I think with AI,
and then I share it with my colleagues.
Okay.
And what that means is like, hey, he's,you know, the CEO of,
you know, one of the top five businessesin the world, right?
And billions and billions and revenue,hundreds of thousands of employees.
(02:51):
And so he's a guy who definitely thinksabout strategy all the time, right?
He's not in there,noodling around at the tactical level
of, you know,what's going on, within his business.
And so when it comes to strategic planning
for financial advisors, well, you know,what does that even mean?
(03:11):
And, it could mean things like,I want to grow my business.
I feel like I'm extremelywell known in this marketplace, here.
But, what if I wantedto do something different?
And you could have, you know, a deep
and meaningful conversation,you know, with your eye about.
(03:36):
Okay, you know, maybe there's, you know,maybe you're in a semi, semi-rural,
or some suburban areaand you know that there's another,
small metropolitan area 30 miles awayand you're thinking, wondering if,
if we could put, put an office over there,plant a flag, so to speak.
(03:56):
There aren't many financial advisorsover there that are in this area.
And there's some new thingshappening there.
Maybe there'sa new manufacturing facility.
Whatever the case is, you can sit downand you can invoke,
invoke the role of strategic planning
consultant who specializesin working with financial advisors.
We talk about, how in our programs,we teach people, the proper framework
(04:20):
for communicating with AI modelsto get the best results.
And so when it comes to using AI
for strategic business planning,you would want to be,
invoking some level of expertiseto kick off your conversation.
Right?
So you can say act as that, you know,strategic business
(04:41):
planning consultantwho works with financial advisors on
how to identifyand grow new markets in their business.
I want to have right.
That's the role right now on the task.
I want us to have a sort of freewheelingbrainstorming conversation about,
how I might think about, planting the flagand opening a new office.
(05:04):
Yeah, there's maybe 2 or 3 potential areaswithin 30
to 45 minutes ofof where you're living geographically.
And you would say, you know where that is.
And then you'd say,let's have a conversation,
about, about that.
Ask me questions.
And let's just see where the conversationgoes.
(05:25):
I have done this sort of thing.
I've interviewed peoplewho've done this sort of thing.
You know,it is totally mind blowing experience.
Because these AI models are so smart.
They've been trained on so much data.
And the AI will start asking yousome very, insightful questions
about what it is you want to achieve,going into this market.
(05:47):
Then it might, have it producean analysis,
a demographic analysis of, of of the areasthat you're looking at.
What are the, you know, what are the,you know, what's the net worth of,
families in these regions,who are the main employers?
How, what type of, client
(06:09):
that exists in your businessnow where you're successful?
How many of those types of clientsmight we find?
In that area?
And then the conversation can switchsometimes from strategic planning,
a strategic business planningconsultant to a sort of, data analyst,
who is familiar with Census Bureau
data and small BusinessAdministration data.
(06:31):
And you can have a, you know, meaningful,
conversation about the types
of professionals who need financialplanning services or investment management
consulting services in that area,and go from there.
Right.
So that would be,yeah, that's that's an example of,
having strategic conversations,
(06:53):
with, with these AI models.
And now,you know, more specifically, again,
you know, hey, you couldthat's one approach, right?
But you could just simply be wantingto talk about your business now,
and doing a better job in your marketwhere you're at
and you could, you know, have a strategic,
business planning conversationaround your existing client base.
(07:15):
Right.
You could take, your existingclient segmentation
and take that dataand share that with the, AI model.
And again, and ask, you know, ask
it, you know,what is it seeing, in your data,
how does that speakto ask me some questions
about what you're seeing there, howcan we do a better job of segmenting this?
(07:38):
My, my client base so that,
we can either be more efficientor we can have better insights.
That will then lead to other conversationsand tactics
that would allow us to, growthat particular client segment.
Right.
So that's that'sanother kind of thing you could do.
You could even, have the AI model examineyour, existing client segmentation.
(08:03):
And from there, start to build out,you know, client personas, ideal client
personas of your existing,book of business, which many advisors,
I don't think really have done.
They've had an a sense of whattheir ideal client is like.
And it's kind of work to some degree.
But they haven't had the time becausefrankly, before I came up with them,
it would be expensiveand and time consuming to really sit down
(08:27):
and look at your, your list of clientsand all the information you have
about them, and try and tease outsome more finely tuned segments
that could give you actionable knowledgethat you could,
you know, startimplementing your business.
I used to be very hard.
Now it's it's a lot, a lot,
easier, to do that sort of thing.
(08:50):
And it's, it's easierto do a sort of thing that most advisors
avoid, avoided doingbecause it took too much time.
It took too much timewhen they had so many other things to do.
So that's another example of,those are just a few examples
of of how you can useAI for strategic business planning.
And there is, you know, moreand more happening in this space.
(09:10):
And, you know, Devin will just mention,of course, we're having our introduction
to, you know, toAI for financial advisors.
That program kicks off, this Thursday.
It's called the AI PoweredFinancial Advisor.
And, we'll be,you know, teaching advisors, the basic
the foundational knowledge that they needto have these effective, these kinds
(09:32):
of effective conversationswith, the AI models to, you know,
start to use them, more effectively,you know, in the, in their business.
Yeah. I think that's great.
Those are great examples.
And another,kind of way I saw this being used is,
you know, I meet with a lot of advisorsthrough our advisor client
marketing bootcamp.
(09:52):
About their marketing plan for the year.
And, at the end of last year,I had a couple conversations with advisors
who specifically were looking,at strategies of how to bring in,
some younger clients, maybe most oftheir clients are entering retirement.
They're going to be kind ofin this distribution phase of life.
And they were thinking about
how can I bring in, clientswho are working,
(10:13):
who are accumulating money right now,which are kind of,
you know, balance that out.
And we talked a lot about using,
AI to think about howthey could expand their services.
So if they weren't alreadyoffering things like,
you know, college planningand using AI to look at their client lists
and see if that would be somethingthat would be relevant
to people and worth, you know, exploring,or if that's something they're interested
(10:35):
in, then using AI to kind of likeyour suggestion, build out those personas
and think about the kind of peoplethat they need to,
you know, target and get in front ofand strategies to do that.
So I think there's so many waysthat you can really use,
AI part that's kind of strategybuilding tools for sure.
Yeah.
And that example that you just offeredcould be, you know,
fall under the umbrella in essence of,you know, personalized financial planning.
(10:58):
Right? Right.
I mean, you're
you you're definitely have your normalblocking and tackling that you're doing
for your clientson the financial planning front.
But you were just suggestingnow look more deeply,
through the lens of,okay, which, clients, in the next
five years are going to be strugglingwith the issue of, oh my God,
how do we send, you know, how do we payfor sending the kids to college?
(11:18):
So that would that would be like,
yeah, using it as personalizedfinancial planning for sure.
Right.
You know, another and another topictoo is like, you know, suppose you're
an advisor who's like a lot of these preretiree, retiree baby boomers.
You're getting to the point where, you'reready to, sell the business, right.
Or think about selling the business.
And so again, you could havea very fruitful strategic conversation
(11:42):
about succession planning,what you have or don't have.
You could, invoke the role,the expert role
of, you know, you are a consultant
that works on successionplanning with financial advisors.
You've helped a, more than 300 advisors inthe last 20 years sell their businesses.
And, I want to have,you know, opening conversation about,
(12:06):
what I need to be thinking aboutas it relates to a developing,
developing my own succession plan.
Ask me questions one at a time.
I want to spend half an hourwith you on this, and then at the end,
we'll, develop a bullet point summaryof the key issues that we've discussed.
And, you can record you meeting you.
The I model, the one who were sayingis an expert in succession planning.
(12:30):
You can then, recommend to me, you know,
you know, 2 or 3 next steps, for us to,you know, get more into,
a full blown, discussion of successionplanning.
So, you know, that's, you know.
Yeah.
Yet another way to use itfor strategic planning.
There are so many, so many,so many examples.
(12:50):
And the survey also found that about 13%of advisors
are using AI for data extractionfrom documents.
And I think some of this is probably,you know,
I that's baked into software that they'reusing to to help in this process.
I know some of the more popular softwareshave this now.
But even after, you know,
the data has been extractedand all of that, what kinds of workflows
do you think advisors could kind of createaround this capability,
(13:13):
and how could I help themstreamline their efficiency?
Yeah, you know, I don't know.
We, you know,
we've been this whole thing is lessthan is just barely two years old.
This whole ChatGPT,generative artificial intelligence.
And, we know
one of the things is we're looking at thisand you're looking at AI's abilities.
Is to tease out the difference between,things that generative AI can help you
(13:36):
with, that there's not an existingexcellent tool set, already doing that.
Okay.
And, and so the at this point in time,you know, only 13% when I saw that
the data extraction from documents, I'mreminded of an advisor in,
our October class, which was an AI poweredfinancial advisor program.
(13:58):
That's the introduction to AI.
And, he came into one of our sessions,like, on fire
and totally enthused about, his use case.
I think he had, like, 10 or 12 clients
that, were holding some sort of,
investment positionthat was expiring, and,
(14:19):
they needed to,you know, their, their, their, you know,
that portion of their investmentportfolio needed, needed to be,
you know, reallocated,or re read redeploy and reinvested.
Right.
And I, they all were in some particulartype of, investment vehicle
and he had like a, you know, 120 page PDFabout that investment vehicle.
(14:44):
And there was somethere is a particularly really good, table
that was inside, inside this documentthat he said in years past,
he tries to copy and paste it intosomething that he can give to the clients.
That's useful.
And, it's like it was a nightmare,and trying to do it.
And because he'd been in our class,and realized that, oh,
(15:08):
he could actually take that 120 pagePDF document that he received from,
you know, from the asset manager who's,you know,
running the moneyfor that particular investment.
He can upload that 120 page, report,
and he can have a conversation,through the AI model
(15:29):
where the AI is being directedto look specifically
into this PDF,for the answers to my questions.
And he was able to, query
the AI model looking at this,you know, huge PDF and get it to,
produce back to the screenthe very specific table or chart, that
(15:49):
or a specific type of table or chartthat was just too cumbersome
and difficult to do in the past.
And instead this thing just,you know, produced it like that.
And so, in a very effective way,he was very happy with it.
And so, you know, in in some cases,
we in our your previous question, Devin,we were talking
(16:10):
globally macro strategy.
Right.
And now we're talking about the absoluteopposite where the AI model
has acted like a very good office managerwho knows how to,
you know, create neat tables and charts,and extract them from, a big cumbersome
file, you know, that many financialadvisors themselves aren't very good at.
(16:32):
And so we, go from,
you know, broad strategic questionsall the way down to super tactical things.
Hey, hey, ChatGPT,can you help me reproduce this table
in a way that I can use it,
in this report that I'm creating for,you know, 13 of my clients?
Yeah, I think that's a huge use case,even though it seems silly.
(16:54):
You know, things like formatting,but that is stuff that can take,
you know, someone.You are someone on your team.
A lot of time to go through.
And especially when you're doing it againfor a small number of clients,
you can kind of why is it worth itto be spending this time?
But when you can have I come in and formatthings the way that you want them
to be formatted in a waythat you can easily copy and paste.
I think that is actually a huge use casethat is sometimes overlooked,
(17:18):
because it's not this grand, outputthat we're getting,
you know, a whole marketing planor something like that.
But from a, you know, everyday
kind of use case, that's somethingthat, you know, advisors can be using.
And to your point, to, you know,
beyond just data extraction,but being able to simplify,
these more complex documentsthat you maybe
don't want to share the whole documentwith the client, it's too much.
(17:40):
You're not going to be able to understand.
Maybe being able to summarize withI come up with bullet points from,
you know, this larger fileand then have that to share
with clients as you're talking pointsas an email cover letter,
is something that I think is really simpleto do on an everyday basis
and can really make an impactin your communication, going forward.
(18:00):
Yeah.
Because, you know, also we, you know,we live in this world where like, like,
you know, for any variety of reasonsrelating to this last example,
we were referring to, the, the companythat's running the money for those 13
clients and the advisor who's,you know, managing the relationship there,
you know, they're producing120 page PDF report on whatever.
(18:21):
Now, who the heck has time to read120 page report?
Now, if you look at that report, you know,who knows why.
It's 120 pages.
And sure,it probably has an executive summary,
but sure, there's a ton of stuffthat's loaded into that report that's
probably being required,you know, for compliance reasons.
At the, at the, the asset manager or,
(18:42):
or they're including itbecause they know that the information
is going to be shared
with the advisors who are their clientsor the advisors clients.
And so there's so much, dataand information we get, these days.
And it's larded with things,that are, there for a variety of reasons,
but aren't there to communicate
(19:03):
the ground truthof what exactly the document is all about.
And that's where the, you know,super powerful summarizing features of,
generative AI, you know, come into playlike you were like
you were just pointing out.
So, yeah, I'm being able to add, you know,you have a meeting with a client
and you have notes from thatbeing able to use
(19:23):
AI to pull out the important stuffso you can add it to your CRM.
It's a simple task like thatwhere, you know,
when you add it up, how much timethat would take you to do for each client.
It ends up being a lot of time,that can you can get saved with using AI.
So, yeah, I think there'sa lot of opportunity in this area as well,
but moving on.
(19:44):
We just kind of mentioned complianceand how, you know,
sometimes you need to document thingsfor compliance purposes.
We're seeing about 9% of advisorsusing AI for compliance documentation.
I know in our AI for marketing program,we kind of introduced this,
GPT to maybe help with some compliance,
(20:04):
you know, review before you got there.
This is obviouslystill kind of a gray area.
We don't really know how compliancedepartments are going to react to this.
When we spoke last weekabout how the CFP board just released
some very broad documentationon using generative AI,
but do you see any kind of opportunitiesor risks in this area
around compliance documentationand using AI in that process?
(20:27):
Yeah.
You know, listen, I'll be honest,
you know, our our exposure here at Horse'sMouth as it relates to compliance
and advisors is is almost exclusivelyin the marketing space, right?
Whereas the advisors themselvesgot a lot of other things, that,
compliance requirements, you know, that,that we're not involved with.
(20:47):
Right.
But, one key thing
of course, is that, when there are new,
new rules or existing rulesthat are being brought to your attention
from your own compliance departmentor whatever the case may be, again,
they are they can be, very,
(21:08):
difficultto read and summarize or understand.
And so again,you know, one simple thing for anybody
who is looking at a particular,compliance regulation that may be,
their compliance,people have said something about,
and you might not understandwhat your compliance person is saying.
(21:29):
Oh, my God, that's not ever happenedbefore.
Right.
You can go to, you know,you can go to the SEC, Greg
or a Finra, Greg, that relates to,you know, your particular firm
and you can find the,you know, relevant, site location,
you know, which these things
have all been written by lawyersand they've been written by lawyers.
(21:49):
Even worse, written by lawyersspecializing in compliance,
meaning they can be absolutely difficultto understand.
So you can flip out,you know, a page and a half, two
and a half pages, one paragraph,whatever the case is of of,
barely, decipherable,
English that is trying to explainsomething about compliance.
(22:11):
And you can go back to the AI modeland you can say,
I want you to act, you know, as a,
you know, skilled, compliance professional
who has worked withwith financial advisors for 30 years.
I want to have a small discussionabout this regulation in the
compliance code that my compliance officerhas said dadada about.
(22:34):
Right.
And then you say,I want you to explain this to me
like I'm a 12 year old, okay?
Boom. It is.
The AI model is able to go back,you know, read,
the jargon filled, compliance.
Rule and, can help you,try and, understand,
(22:57):
what is being said by taking the,you know, jargon filled, compliance
laden language and boiling it down to,you know, readable English that you can.
Oh, okay.
Now I understand, you know,what's going on,
and, and or I can understandit better than I understood before.
And now I can go backto my compliance officer
(23:18):
and ask a good follow up questionthat will clarify this thing
that we're going back and forth about,that there seems to be some confusion.
Okay.
So that is, you know, that is one example.
Now, you also mentioned that in ourAI market,
AI marketing for advisors program,which starts in April.
That's a five week program.
(23:39):
And we just finished it,
back in the beginning of February,the first time we did it.
And we do have a comply with compliance,
I think we call it a compliance pre-check.
Tool or assistant compliance,pre-check assistant.
We're all in that program.
It's all aboutcreating marketing content. Right.
And so we have created an AI tool that,
(24:03):
isn't giving youthen a review of your material.
Do you still need to go to Finraor whomever is handling your compliance?
But it is trained on all of the relevant,
complianceand Finra rules around marketing.
And, the tool can,
(24:23):
you can take material that you're aboutto share with your compliance department
and run it through this pre-check,this compliance pre-check.
I compliance pre-check tool,
that we have createdand it's made available,
so that it can spot some thingsbefore you send it off.
The complianceand bring it to your attention and cite,
(24:45):
you know, the relevant,the relevant marketing Finra rules
that it thinks there mightyou might have an issue with.
And so it's flagging itbefore you send it off the compliance
so you yourself can look at itand double check.
And if you agree with the AI modelthat yeah, you know what,
maybe that statement is a little bittoo strong.
(25:05):
It's pushing the envelope, so to speak.
Let's pull in our hornsand rewrite it in a good way that,
we would, past complianceand still do a strong job of communicating
whatever it is we're trying to communicatein our marketing
to prospects and clients.
And why is that so important?
Because potentially, without that, youwould have shipped it off the compliance.
(25:28):
Who knows how long it would takefor compliance to look at it.
They've got this thing,they'd send it back to you.
And so you get this back and forth thinggoing on between the advisor
and the compliance department.
And this tool is simply designed to, flagsome of those things,
give you the opportunity to address themso that we can reduce the amount
of timegoing back and forth on marketing content.
(25:48):
And your compliance department.
So those are a couple of examples.
And, I suspect an advisorwho is, you know, more schooled
in the compliance stuff, when he orshe is trained on how to use AI.
And they'll be, you know,they'll be able to use this, you know,
they'll be able to find other effective,
(26:11):
uses of AI in their broader,
issue of, you know, compliance tasks,
that they have, including also trainingyour team, on compliance stuff, too.
So, it is an interesting field.
We're going to be looking more at it.
But again, you know, 9% are doing it now.
And that's 9% of of peoplewho are, represent 20% of all advisors
(26:35):
who are probably, you know,who are using AI on a regular basis.
So, that's yeah, I think that's the keything to keep in mind with that one.
Definitely.
Yeah.
I, I was chucklingduring your first example because,
you know, within our materialsthat we submit for, for fin Riverview.
I handle a lot of that internally for us.
And recently I did exactly what you said.
(26:56):
We got something back from Finra.
Finra where they quoted some, you know,rule that was just in gibberish to me.
I could literally could not understand.
And if anyone submitted anything to Finra,you know, that they just tell you
the rule.
They don't tell you where in youryour document that exist.
You just have to kind offigure it out for yourself.
And so I actually,you know, pulled up the rule.
(27:16):
I want to perplexity.
I and asked for a little bit of researchon it, to understand what that rule meant.
I brought it over to ChatGPT.
I asked her to explain it to me, like.
Like basically I was in fifth grade.
And then I even took it as far to then,you know, upload the document in question
and ask it to analyze that documentand try to figure out where
(27:36):
the the overall,you know, what they were referring to.
This was a very tricky rolethat I didn't have a ton of success
with them being able to analyze it,but when I actually spoke to our kind of
compliance partner and this, they alsocouldn't figure out exactly where it was.
And so it ended upbeing one of those wacky Finra things.
But by just simply askingfor an explanation,
(27:57):
it kind of got me to a pointwhere I could figure out, okay,
I don't think that this actually applies.
I think some of themjust being really picky with, this rule,
and I could actually understandit enough to have it to your point,
have a conversation where I felt confidentthat, you know,
we could push back and say,I think this needs to be re reviewed
(28:18):
because this doesn't make any sense.
And so it cut down, you know, normallybefore this, I've done this in the past
where you have to go find the Finraright here all through the document.
Try to figure it out on your own,have internal conversations.
And I was able to cut down on all of thattime.
And, you know, so at the time I processeddealing with Finra or compliance,
but at least again, I was going into itwith some kind of educated,
(28:42):
talking points about, you know, whyI thought that it should be reviewed
and we could leave it as a soI think that is certainly a helpful point.
And I think people just need to kind of,use common sense
around, you know, complianceand AI and how to make sure that you're
not going to get yourselfinto any issues there with compliance.
But I think it actually has a huge,
(29:03):
benefit, especially if you're someonewho's, you know,
taking meeting notesand you need to submit things like that
to compliance, havingI go through those notes and, you know,
format in a way that compliance wants it,there can be a lot of, you know, pros
for the documentation side of,of how to use these tools,
to help with your compliance departmentand make it
(29:24):
less of a headache for, you know,anyone who has to deal with that.
And also, as an aside,and you know, presumably this will be,
an example ofAI being used to everyone's benefit
except for maybe the compliance officerswho could very well,
I mean, compliance could become, one ofthose areas where there is job loss,
(29:46):
to some degree,because, the kinds of things they do
are very, in alignment withthe kinds of things that AI model can do.
And so,you know, it's either 1 or 2 things.
One, everybody will have their ownvirtual compliance officer,
kind of like what we're doingwith our compliance pre-check tool.
And they'll be just they'll simply be lessfriction in the compliance process
(30:09):
because, the advisor will have their ownvirtual compliance officer
and everything will be kind of caughtand addressed before it's ever submitted.
But then on the other,the other end in Finra, the person
reading the compliance submissions,you know, could very well, you know, be a
AI agentthat is being overseen by a human being.
(30:30):
But the point being thatwe also could see,
you know, massive speed upin, responses and answers from compliance.
So therefore, you know,there could be, tremendous
new efficiency gains in the complianceworld, in the near future.
And it would be my predictionthat the compliance world
is pretty differentfive years from now, for sure.
Yeah, I think that's definitely true.
(30:52):
And so, you know,kind of continuing on this trend here,
this is a very kind of untapped areaamong advisors.
This, kind of advanced AI applicationsthat we've been seeing
in the survey results.
But reports that only about 4% are usingAI for kind of personalized
financial planning, which we talkedabout a little bit earlier.
But, I mean, I think there'sa lot of potential growth in this area.
(31:13):
Would you agree?
And do you have any kind of insightson where we might see AI being built
into that personalizedfinancial planning aspect of the business?
And, you know, by the way,just to remind you, the question was about
what was related to using AIfor multilingual
capabilities and personalizedfinancial planning. Yes.
Thank you.
(31:33):
Yes. And, you know, it's that multilingualthing that really catches my attention.
You know, I live here in new Jersey,and I think I heard recently
that there are,you know, that there are 50 different
languages spoken and New Jersey's,you know,
how many people we had 7or 8 million or something like that.
We're one of the most denselypopulated, states in the union.
We're close to New York City andPhiladelphia, major port of entry here.
(31:54):
So there are a lot of peoplein this state,
whose native language is not English.
They're here, their families are working,they're making money.
They have their American citizens.
They have financial planning issues now.
And, thethe question that comes up here is
being able is for an advisorwho is, you know,
(32:18):
may have a client who, you know, maybeyou have a client who speaks English.
And but they have, a parent, with themand maybe they're from Afghanistan and,
the parent is involvedin, it's their money.
But the younger relative who speaks
English is the one who interfaceswith the financial advisor.
(32:38):
And, in the past, it could be
that, you know,the son is translating some
financial concepts to the parentsor whatever the case may be.
But these AI models can speakall of these languages, right?
So now you have the ability to, take,
your existing tools and your existingadvice that are written in English and,
(33:03):
translate them, instantly, into,
you know, usable bits of informationthat can
then be shared with the,speaker of a non-English language.
And it can, you know, be very helpful
in those kinds of difficult situations,the difficult meeting situations.
But that's just one example, right?
(33:23):
Now we know and have heard over the years,for instance, you know, a horse's mouth
savvy Social Security planning program is,you know, many advisors have built,
you know, massive businessesaround bleeding with Social Security
planning, in their communities,as a way of meeting people.
And, you know, the Social Security keyaspect of retirement.
(33:43):
And we've had some people over the yearswho, you know, spoke,
Spanish and, had Spanish speaking clients.
And, you know,they would kind of translate on their own
some of the, you know, information,
the market and financial education,materials in Social Security.
But now, you know, it it franklywould not be very difficult to like,
(34:05):
you know, take a,you know, take a presentation for social,
Social Security, written in Englishand have it translated into Spanish.
And I know it sounds crazy, but,
you know, it could probably just be donein just a couple of minutes.
And, and, you know, as long as you havesomebody who does speak the language,
who can then kind of go through and say,yes, this makes sense.
(34:26):
And it's properly translated.
But so that's, you know, that's,that's another example of, using it.
But, you know, the question also is aboutpersonalized financial planning.
And, you know,one of the things that you're able to do
is, take very specific financial planning
or investment management challengesthat you're having with your clients.
And you can take your scenario to theAI model.
(34:50):
When but when but when I say AI model,I mean ChatGPT or Claude or Gemini.
You know, most in our in our programs,we have everybody use ChatGPT
for a variety of reasons.
But there are other models out thereas well.
So, to be semi neutral, I say AI models.
But anyway, the point is that, you can go,
to these models and with, you know, say,hey, look, I'm meeting with this client
(35:14):
or this particular typeof financial planning challenge.
Here's the situation.
You're not sharingpersonal identifiable information.
You don't need to beat that granular level to get very, very,
useful and effective, advice, guidance or,brainstorming,
in terms of trying to see what,you know, how
(35:35):
we can, not this particularfinancial planning challenge.
So in that respect, we can say, yeah,there are ways
it's only 4%that are doing it or realizing it.
And so we're going to be doinga lot more on this going forward as well,
because, you can in your eyes,I've done a number of times
in our conversation this morning,I've invoked various, various expertise.
Right.
And so you can have a conversation,with the AI model,
(35:58):
and you can invoke the role of a certifiedfinancial planner with 30 years
experience working with,you know, whatever, whatever
subtopic within financial planningthat you're, you're trying to address.
You can cite that as part of your prompt.
Then, you know, includethat as part of your, part of the task of,
like, we've got this problem,this challenge.
(36:20):
We've got to make some,we want to come up with some ideas of,
what might be the best solution for thisparticular financial planning challenge.
And, like, we were talking about strategy,broad strategy
for your company at the top of the, hour.
You know, now, you could be havinga similar type of conversation
in the realm of financial planning,when you've given the AI model,
(36:43):
you know, the facts of the situation,and that can be, very effective as well.
Yeah, I think that's great.
And your first example kind of reminded meto have, you know, we have advisors
who, also, you know, have Americanand Canadian clients.
And sometimes, you know, we're asked to,you know, does this stuff apply to Canada?
And I think, you know,
(37:03):
if you're in that areawhere maybe you're unsure
about some of the rulesyou have for clients
who are moving overseasand you want to help them with that,
even if they're going to an Englishspeaking country.
You can kind of use AI to help,
you know,get schooled up on some of the differences
that they might be facing, you know,as they move to a different country
or reside in a different countryas well. Right?
I mean, those cultural,you know, those cultural differences,
(37:25):
cultural personalization, you know.
Absolutely.
And, you know, also,just simply using it to like, okay,
you'vegot some clients coming in there from,
you know,they're successful professionals.
But they're from,you know, Iran or Poland or,
you know, or Chile or whateverthe case may be.
And, you know,you don't know anything about
(37:47):
Iran or Poland or Chile other than thatthey're not the United States.
But in advance of the meeting,you know, they're going to come in,
maybe it's the first meeting and,you know, you could again,
say to the AI, hey, I'm about to havethis, meeting with this couple.
You know, they're a
successful, tech entrepreneur, from Chile.
They want a financial advisor,to help them with their investments.
(38:10):
And, you know,I don't know anything about Chile.
Kind of give me a little briefing on it.
And, give me some, you know,light hearted without making jokes
or anything like that. So just help me
learn a little bit about themand the history of their country.
And all that sort of thing,maybe as a way of sort of enhancing,
or, you know, enhancing the relationshipor making some,
(38:31):
you know, making some smallbut meaningful connection.
In the, yeah, in the meeting.
Right. Yeah. Yeah, I think that's great.
So as we kind of close out this sessionhere for today for advisors
who haven't started kind of experimentingaround this area of AI in their business,
but one simple way that they could kind ofget started this week,
we're thinking about,these more advanced use cases for AI.
(38:55):
Good question.
Well, you know,
I think the most important thing forpeople, is to start using it, right?
And, that's,
you know, Ethan Malik, the professorof innovation at the Wharton School,
the University of Pennsylvania says,that we need to be thinking
about these AI's as a co intelligencethat can collaborate with us.
(39:16):
We're in charge, though.
And so but he says it takes about 8to 10 hours of really working with them.
And so and which of course iswhy we've created these training programs
because we're translatingfor financial advisors
100% all our programs,our use cases for advisors.
So we're not we're not talkingabout broad, uses or general uses of AI.
(39:39):
We're talking about very,very specific kinds of things.
So you know, I would just tellsomebody to, think about a challenge
or think about a challenge or problemyou're having in your business right now.
Think aboutwho would be a really smart person
who could offer you some insightabout your challenge.
Explain that situation in a promptand then,
have a conversation about itand just see, see where it goes.
(40:02):
Right.
But again, I mentioned, you know, thereis this learning.
This is super powerful.
New, tool.
We have seen, you know, we're seeing
big growth and adoption by advisors.
We did this cert.
We did our first survey back inMay and June of last year.
We had like 12% of advisorswho were using AI on a daily basis.
(40:25):
Now our our current surveythat's in the field,
we're looking at, in the 24 to 26% range.
So basically a doubling in six months.
These are advisorswho are recognizing, wow,
this thing is really powerfuland can be very, very helpful with me.
It can help me save time.
It can, produce high quality work.
(40:47):
And it can also, serve to,
inform me, and brainstorm with me
across all sorts of different fronts.
But again, you know, 8 to 10 hoursis what you really need, but for what?
To start to answer your question, Devin.
Yeah, taketake one of your problems to it.
Invoke some expertise.
And, remember, you're not
(41:09):
this is not the chat box looks.
The chatbox looks too much like a search box.
We know, we see search.
We see advisorssearching on, horse's mouth all the time.
And we know people search on 1 or 2 wordsand basically.
Right.You can't do that with the AI models.
You have to have a robust conversationand back and forth
and you'll be amazed at what you learn.
So anyway, Devin, I'll I'll leave itat that if you just want to,
(41:33):
you know, pointpeople to where they can go to be, or you.
Yeah.
If they still want to sign up
for our program, which starts Thursdayand runs for 30 days in a row,
why don't you tell thema little bit about that?
And so that thank you, everyonefor being here with us today.
And thanks for joining me for this chat.
As Sean mentioned,
our AI powered financial advisor program,which is kind of our intro to AI.
(41:53):
It's a four weekprogram begins on Thursday, March 6th.
You can learn moreby going to your horse's mouth.com/ai.
We've had over
100 advisors go through this program nowat this point, and three different groups,
with great success.
So we're really looking forwardto kicking off another group,
this Thursday, March 6thand then starting in April, on
April 3rd, we are doing ourAI marketing for advisors workshop.
(42:16):
This is a five week programthat really gets into the nitty gritty
of how to use AI in your marketing.
We teach you how to kind of come awaywith a full marketing plan.
You know, that you can then implementthroughout the year, and we talk about
how you can use AI in all different partsof that plan to actually,
make your marketing come to life.
So if you want to learn more about that,you can go to horse's mouth.com/i
(42:38):
marketing. It's on the screen right nowfor you there.
We hope to see you at one of our events.
And thanks for being, part of thislive stream today.
Sean, thanks for joining me,and we hope to see you again soon.
Hey, Kenya.
Welcome to today'sshow on AI for Practice Efficiency,
how advisors are streamlining workflowsand unlocking new opportunities.
I am is now seven cropand I'm joined by Sean Bailey, Horse's
Mouth editor in chief.
Today, Sean and I will be sharingsome early results from a survey
we have out in the fieldabout how advisors are using AI
(00:20):
to skyrocket their practices, efficiencyand pave the way for future growth.
So, Sean, thanks for joining me today.
We can jump right into it here.
The survey shows that about 12%
of advisors are using AI for strategicbusiness planning.
What specific initiativesdo you think an advisor
could kind of developusing AI in this way?
(00:42):
Yeah, it's a good question.And it's really fascinating.
And of course, you know,we're about to go through, some results
from our survey of use casesthat are on the lower end.
Right.
We've talked previously about,you know, using AI for content
development, for marketing,for preparing client meeting notes.
But this is a general purpose technology,meaning that it can be,
(01:04):
you know, applied, you know, across
nearly all domains and,
and within each domain,across nearly all types of job tasks.
And so I just wanted to put that out therefor background.
And we're going to be we're plumbingsome of the moment at this moment in time.
Some of the usage is that,have a lot of potential.
(01:27):
And in some advisors are doing thembased on our survey.
But, they're not widespread by,
by any means at this point in time.
Now, your question about,using AI for strategic
planning, is really,a good, it's a favorite topic of mine.
And I was listening to a podcast last weekwith, Satya Nadella,
(01:49):
who the CEO of Microsoft and I mentioned,I, I mangled his quote last week.
But it actually applies very muchto this, this question of strategy.
Right? Because we're not talking tacticsand things.
I was mentioning before about, you know,content development, client meeting notes,
these are all tactical kinds of thingsyou do within your business, right?
(02:09):
But now we're talking about strategy,which is kind of the bigger,
the bigger picture.
Looking ahead, trying to anticipate,the direction of your business,
what's going on in the market,what opportunities you have.
And I can be, you know, hugely.
Helpful in this way.
And Satya Nadella said, in this podcast,
(02:30):
he said the new workflow for me
is that I think with AI,
and then I share it with my colleagues.
Okay.
And what that means is like, hey, he's,you know, the CEO of,
you know, one of the top five businessesin the world, right?
And billions and billions and revenue,hundreds of thousands of employees.
(02:51):
And so he's a guy who definitely thinksabout strategy all the time, right?
He's not in there,noodling around at the tactical level
of, you know,what's going on, within his business.
And so when it comes to strategic planning
for financial advisors, well, you know,what does that even mean?
(03:11):
And, it could mean things like,I want to grow my business.
I feel like I'm extremelywell known in this marketplace, here.
But, what if I wantedto do something different?
And you could have, you know, a deep
and meaningful conversation,you know, with your eye about.
(03:36):
Okay, you know, maybe there's, you know,maybe you're in a semi, semi-rural,
or some suburban areaand you know that there's another,
small metropolitan area 30 miles awayand you're thinking, wondering if,
if we could put, put an office over there,plant a flag, so to speak.
(03:56):
There aren't many financial advisorsover there that are in this area.
And there's some new thingshappening there.
Maybe there'sa new manufacturing facility.
Whatever the case is, you can sit downand you can invoke,
invoke the role of strategic planning
consultant who specializesin working with financial advisors.
We talk about, how in our programs,we teach people, the proper framework
(04:20):
for communicating with AI modelsto get the best results.
And so when it comes to using AI
for strategic business planning,you would want to be,
invoking some level of expertiseto kick off your conversation.
Right?
So you can say act as that, you know,strategic business
(04:41):
planning consultantwho works with financial advisors on
how to identifyand grow new markets in their business.
I want to have right.
That's the role right now on the task.
I want us to have a sort of freewheelingbrainstorming conversation about,
how I might think about, planting the flagand opening a new office.
(05:04):
Yeah, there's maybe 2 or 3 potential areaswithin 30
to 45 minutes ofof where you're living geographically.
And you would say, you know where that is.
And then you'd say,let's have a conversation,
about, about that.
Ask me questions.
And let's just see where the conversationgoes.
(05:25):
I have done this sort of thing.
I've interviewed peoplewho've done this sort of thing.
You know,it is totally mind blowing experience.
Because these AI models are so smart.
They've been trained on so much data.
And the AI will start asking yousome very, insightful questions
about what it is you want to achieve,going into this market.
(05:47):
Then it might, have it producean analysis,
a demographic analysis of, of of the areasthat you're looking at.
What are the, you know, what are the,you know, what's the net worth of,
families in these regions,who are the main employers?
How, what type of, client
(06:09):
that exists in your businessnow where you're successful?
How many of those types of clientsmight we find?
In that area?
And then the conversation can switchsometimes from strategic planning,
a strategic business planningconsultant to a sort of, data analyst,
who is familiar with Census Bureau
data and small BusinessAdministration data.
(06:31):
And you can have a, you know, meaningful,
conversation about the types
of professionals who need financialplanning services or investment management
consulting services in that area,and go from there.
Right.
So that would be,yeah, that's that's an example of,
having strategic conversations,
(06:53):
with, with these AI models.
And now,you know, more specifically, again,
you know, hey, you couldthat's one approach, right?
But you could just simply be wantingto talk about your business now,
and doing a better job in your marketwhere you're at
and you could, you know, have a strategic,
business planning conversationaround your existing client base.
(07:15):
Right.
You could take, your existingclient segmentation
and take that dataand share that with the, AI model.
And again, and ask, you know, ask
it, you know,what is it seeing, in your data,
how does that speakto ask me some questions
about what you're seeing there, howcan we do a better job of segmenting this?
(07:38):
My, my client base so that,
we can either be more efficientor we can have better insights.
That will then lead to other conversationsand tactics
that would allow us to, growthat particular client segment.
Right.
So that's that'sanother kind of thing you could do.
You could even, have the AI model examineyour, existing client segmentation.
(08:03):
And from there, start to build out,you know, client personas, ideal client
personas of your existing,book of business, which many advisors,
I don't think really have done.
They've had an a sense of whattheir ideal client is like.
And it's kind of work to some degree.
But they haven't had the time becausefrankly, before I came up with them,
it would be expensiveand and time consuming to really sit down
(08:27):
and look at your, your list of clientsand all the information you have
about them, and try and tease outsome more finely tuned segments
that could give you actionable knowledgethat you could,
you know, startimplementing your business.
I used to be very hard.
Now it's it's a lot, a lot,
easier, to do that sort of thing.
(08:50):
And it's, it's easierto do a sort of thing that most advisors
avoid, avoided doingbecause it took too much time.
It took too much timewhen they had so many other things to do.
So that's another example of,those are just a few examples
of of how you can useAI for strategic business planning.
And there is, you know, moreand more happening in this space.
(09:10):
And, you know, Devin will just mention,of course, we're having our introduction
to, you know, toAI for financial advisors.
That program kicks off, this Thursday.
It's called the AI PoweredFinancial Advisor.
And, we'll be,you know, teaching advisors, the basic
the foundational knowledge that they needto have these effective, these kinds
(09:32):
of effective conversationswith, the AI models to, you know,
start to use them, more effectively,you know, in the, in their business.
Yeah. I think that's great.
Those are great examples.
And another,kind of way I saw this being used is,
you know, I meet with a lot of advisorsthrough our advisor client
marketing bootcamp.
(09:52):
About their marketing plan for the year.
And, at the end of last year,I had a couple conversations with advisors
who specifically were looking,at strategies of how to bring in,
some younger clients, maybe most oftheir clients are entering retirement.
They're going to be kind ofin this distribution phase of life.
And they were thinking about
how can I bring in, clientswho are working,
(10:13):
who are accumulating money right now,which are kind of,
you know, balance that out.
And we talked a lot about using,
AI to think about howthey could expand their services.
So if they weren't alreadyoffering things like,
you know, college planningand using AI to look at their client lists
and see if that would be somethingthat would be relevant
to people and worth, you know, exploring,or if that's something they're interested
(10:35):
in, then using AI to kind of likeyour suggestion, build out those personas
and think about the kind of peoplethat they need to,
you know, target and get in front ofand strategies to do that.
So I think there's so many waysthat you can really use,
AI part that's kind of strategybuilding tools for sure.
Yeah.
And that example that you just offeredcould be, you know,
fall under the umbrella in essence of,you know, personalized financial planning.
(10:58):
Right? Right.
I mean, you're
you you're definitely have your normalblocking and tackling that you're doing
for your clientson the financial planning front.
But you were just suggestingnow look more deeply,
through the lens of,okay, which, clients, in the next
five years are going to be strugglingwith the issue of, oh my God,
how do we send, you know, how do we payfor sending the kids to college?
(11:18):
So that would that would be like,
yeah, using it as personalizedfinancial planning for sure.
Right.
You know, another and another topictoo is like, you know, suppose you're
an advisor who's like a lot of these preretiree, retiree baby boomers.
You're getting to the point where, you'reready to, sell the business, right.
Or think about selling the business.
And so again, you could havea very fruitful strategic conversation
(11:42):
about succession planning,what you have or don't have.
You could, invoke the role,the expert role
of, you know, you are a consultant
that works on successionplanning with financial advisors.
You've helped a, more than 300 advisors inthe last 20 years sell their businesses.
And, I want to have,you know, opening conversation about,
(12:06):
what I need to be thinking aboutas it relates to a developing,
developing my own succession plan.
Ask me questions one at a time.
I want to spend half an hourwith you on this, and then at the end,
we'll, develop a bullet point summaryof the key issues that we've discussed.
And, you can record you meeting you.
The I model, the one who were sayingis an expert in succession planning.
(12:30):
You can then, recommend to me, you know,
you know, 2 or 3 next steps, for us to,you know, get more into,
a full blown, discussion of successionplanning.
So, you know, that's, you know.
Yeah.
Yet another way to use itfor strategic planning.
There are so many, so many,so many examples.
(12:50):
And the survey also found that about 13%of advisors
are using AI for data extractionfrom documents.
And I think some of this is probably,you know,
I that's baked into software that they'reusing to to help in this process.
I know some of the more popular softwareshave this now.
But even after, you know,
the data has been extractedand all of that, what kinds of workflows
do you think advisors could kind of createaround this capability,
(13:13):
and how could I help themstreamline their efficiency?
Yeah, you know, I don't know.
We, you know,
we've been this whole thing is lessthan is just barely two years old.
This whole ChatGPT,generative artificial intelligence.
And, we know
one of the things is we're looking at thisand you're looking at AI's abilities.
Is to tease out the difference between,things that generative AI can help you
(13:36):
with, that there's not an existingexcellent tool set, already doing that.
Okay.
And, and so the at this point in time,you know, only 13% when I saw that
the data extraction from documents, I'mreminded of an advisor in,
our October class, which was an AI poweredfinancial advisor program.
(13:58):
That's the introduction to AI.
And, he came into one of our sessions,like, on fire
and totally enthused about, his use case.
I think he had, like, 10 or 12 clients
that, were holding some sort of,
investment positionthat was expiring, and,
(14:19):
they needed to,you know, their, their, their, you know,
that portion of their investmentportfolio needed, needed to be,
you know, reallocated,or re read redeploy and reinvested.
Right.
And I, they all were in some particulartype of, investment vehicle
and he had like a, you know, 120 page PDFabout that investment vehicle.
(14:44):
And there was somethere is a particularly really good, table
that was inside, inside this documentthat he said in years past,
he tries to copy and paste it intosomething that he can give to the clients.
That's useful.
And, it's like it was a nightmare,and trying to do it.
And because he'd been in our class,and realized that, oh,
(15:08):
he could actually take that 120 pagePDF document that he received from,
you know, from the asset manager who's,you know,
running the moneyfor that particular investment.
He can upload that 120 page, report,
and he can have a conversation,through the AI model
(15:29):
where the AI is being directedto look specifically
into this PDF,for the answers to my questions.
And he was able to, query
the AI model looking at this,you know, huge PDF and get it to,
produce back to the screenthe very specific table or chart, that
(15:49):
or a specific type of table or chartthat was just too cumbersome
and difficult to do in the past.
And instead this thing just,you know, produced it like that.
And so, in a very effective way,he was very happy with it.
And so, you know, in in some cases,
we in our your previous question, Devin,we were talking
(16:10):
globally macro strategy.
Right.
And now we're talking about the absoluteopposite where the AI model
has acted like a very good office managerwho knows how to,
you know, create neat tables and charts,and extract them from, a big cumbersome
file, you know, that many financialadvisors themselves aren't very good at.
(16:32):
And so we, go from,
you know, broad strategic questionsall the way down to super tactical things.
Hey, hey, ChatGPT,can you help me reproduce this table
in a way that I can use it,
in this report that I'm creating for,you know, 13 of my clients?
Yeah, I think that's a huge use case,even though it seems silly.
(16:54):
You know, things like formatting,but that is stuff that can take,
you know, someone.You are someone on your team.
A lot of time to go through.
And especially when you're doing it againfor a small number of clients,
you can kind of why is it worth itto be spending this time?
But when you can have I come in and formatthings the way that you want them
to be formatted in a waythat you can easily copy and paste.
I think that is actually a huge use casethat is sometimes overlooked,
(17:18):
because it's not this grand, outputthat we're getting,
you know, a whole marketing planor something like that.
But from a, you know, everyday
kind of use case, that's somethingthat, you know, advisors can be using.
And to your point, to, you know,
beyond just data extraction,but being able to simplify,
these more complex documentsthat you maybe
don't want to share the whole documentwith the client, it's too much.
(17:40):
You're not going to be able to understand.
Maybe being able to summarize withI come up with bullet points from,
you know, this larger fileand then have that to share
with clients as you're talking pointsas an email cover letter,
is something that I think is really simpleto do on an everyday basis
and can really make an impactin your communication, going forward.
(18:00):
Yeah.
Because, you know, also we, you know,we live in this world where like, like,
you know, for any variety of reasonsrelating to this last example,
we were referring to, the, the companythat's running the money for those 13
clients and the advisor who's,you know, managing the relationship there,
you know, they're producing120 page PDF report on whatever.
(18:21):
Now, who the heck has time to read120 page report?
Now, if you look at that report, you know,who knows why.
It's 120 pages.
And sure,it probably has an executive summary,
but sure, there's a ton of stuffthat's loaded into that report that's
probably being required,you know, for compliance reasons.
At the, at the, the asset manager or,
(18:42):
or they're including itbecause they know that the information
is going to be shared
with the advisors who are their clientsor the advisors clients.
And so there's so much, dataand information we get, these days.
And it's larded with things,that are, there for a variety of reasons,
but aren't there to communicate
(19:03):
the ground truthof what exactly the document is all about.
And that's where the, you know,super powerful summarizing features of,
generative AI, you know, come into playlike you were like
you were just pointing out.
So, yeah, I'm being able to add, you know,you have a meeting with a client
and you have notes from thatbeing able to use
(19:23):
AI to pull out the important stuffso you can add it to your CRM.
It's a simple task like thatwhere, you know,
when you add it up, how much timethat would take you to do for each client.
It ends up being a lot of time,that can you can get saved with using AI.
So, yeah, I think there'sa lot of opportunity in this area as well,
but moving on.
(19:44):
We just kind of mentioned complianceand how, you know,
sometimes you need to document thingsfor compliance purposes.
We're seeing about 9% of advisorsusing AI for compliance documentation.
I know in our AI for marketing program,we kind of introduced this,
GPT to maybe help with some compliance,
(20:04):
you know, review before you got there.
This is obviouslystill kind of a gray area.
We don't really know how compliancedepartments are going to react to this.
When we spoke last weekabout how the CFP board just released
some very broad documentationon using generative AI,
but do you see any kind of opportunitiesor risks in this area
around compliance documentationand using AI in that process?
(20:27):
Yeah.
You know, listen, I'll be honest,
you know, our our exposure here at Horse'sMouth as it relates to compliance
and advisors is is almost exclusivelyin the marketing space, right?
Whereas the advisors themselvesgot a lot of other things, that,
compliance requirements, you know, that,that we're not involved with.
(20:47):
Right.
But, one key thing
of course, is that, when there are new,
new rules or existing rulesthat are being brought to your attention
from your own compliance departmentor whatever the case may be, again,
they are they can be, very,
(21:08):
difficultto read and summarize or understand.
And so again,you know, one simple thing for anybody
who is looking at a particular,compliance regulation that may be,
their compliance,people have said something about,
and you might not understandwhat your compliance person is saying.
(21:29):
Oh, my God, that's not ever happenedbefore.
Right.
You can go to, you know,you can go to the SEC, Greg
or a Finra, Greg, that relates to,you know, your particular firm
and you can find the,you know, relevant, site location,
you know, which these things
have all been written by lawyersand they've been written by lawyers.
(21:49):
Even worse, written by lawyersspecializing in compliance,
meaning they can be absolutely difficultto understand.
So you can flip out,you know, a page and a half, two
and a half pages, one paragraph,whatever the case is of of,
barely, decipherable,
English that is trying to explainsomething about compliance.
(22:11):
And you can go back to the AI modeland you can say,
I want you to act, you know, as a,
you know, skilled, compliance professional
who has worked withwith financial advisors for 30 years.
I want to have a small discussionabout this regulation in the
compliance code that my compliance officerhas said dadada about.
(22:34):
Right.
And then you say,I want you to explain this to me
like I'm a 12 year old, okay?
Boom. It is.
The AI model is able to go back,you know, read,
the jargon filled, compliance.
Rule and, can help you,try and, understand,
(22:57):
what is being said by taking the,you know, jargon filled, compliance
laden language and boiling it down to,you know, readable English that you can.
Oh, okay.
Now I understand, you know,what's going on,
and, and or I can understandit better than I understood before.
And now I can go backto my compliance officer
(23:18):
and ask a good follow up questionthat will clarify this thing
that we're going back and forth about,that there seems to be some confusion.
Okay.
So that is, you know, that is one example.
Now, you also mentioned that in ourAI market,
AI marketing for advisors program,which starts in April.
That's a five week program.
(23:39):
And we just finished it,
back in the beginning of February,the first time we did it.
And we do have a comply with compliance,
I think we call it a compliance pre-check.
Tool or assistant compliance,pre-check assistant.
We're all in that program.
It's all aboutcreating marketing content. Right.
And so we have created an AI tool that,
(24:03):
isn't giving youthen a review of your material.
Do you still need to go to Finraor whomever is handling your compliance?
But it is trained on all of the relevant,
complianceand Finra rules around marketing.
And, the tool can,
(24:23):
you can take material that you're aboutto share with your compliance department
and run it through this pre-check,this compliance pre-check.
I compliance pre-check tool,
that we have createdand it's made available,
so that it can spot some thingsbefore you send it off.
The complianceand bring it to your attention and cite,
(24:45):
you know, the relevant,the relevant marketing Finra rules
that it thinks there mightyou might have an issue with.
And so it's flagging itbefore you send it off the compliance
so you yourself can look at itand double check.
And if you agree with the AI modelthat yeah, you know what,
maybe that statement is a little bittoo strong.
(25:05):
It's pushing the envelope, so to speak.
Let's pull in our hornsand rewrite it in a good way that,
we would, past complianceand still do a strong job of communicating
whatever it is we're trying to communicatein our marketing
to prospects and clients.
And why is that so important?
Because potentially, without that, youwould have shipped it off the compliance.
(25:28):
Who knows how long it would takefor compliance to look at it.
They've got this thing,they'd send it back to you.
And so you get this back and forth thinggoing on between the advisor
and the compliance department.
And this tool is simply designed to, flagsome of those things,
give you the opportunity to address themso that we can reduce the amount
of timegoing back and forth on marketing content.
(25:48):
And your compliance department.
So those are a couple of examples.
And, I suspect an advisorwho is, you know, more schooled
in the compliance stuff, when he orshe is trained on how to use AI.
And they'll be, you know,they'll be able to use this, you know,
they'll be able to find other effective,
(26:11):
uses of AI in their broader,
issue of, you know, compliance tasks,
that they have, including also trainingyour team, on compliance stuff, too.
So, it is an interesting field.
We're going to be looking more at it.
But again, you know, 9% are doing it now.
And that's 9% of of peoplewho are, represent 20% of all advisors
(26:35):
who are probably, you know,who are using AI on a regular basis.
So, that's yeah, I think that's the keything to keep in mind with that one.
Definitely.
Yeah.
I, I was chucklingduring your first example because,
you know, within our materialsthat we submit for, for fin Riverview.
I handle a lot of that internally for us.
And recently I did exactly what you said.
(26:56):
We got something back from Finra.
Finra where they quoted some, you know,rule that was just in gibberish to me.
I could literally could not understand.
And if anyone submitted anything to Finra,you know, that they just tell you
the rule.
They don't tell you where in youryour document that exist.
You just have to kind offigure it out for yourself.
And so I actually,you know, pulled up the rule.
(27:16):
I want to perplexity.
I and asked for a little bit of researchon it, to understand what that rule meant.
I brought it over to ChatGPT.
I asked her to explain it to me, like.
Like basically I was in fifth grade.
And then I even took it as far to then,you know, upload the document in question
and ask it to analyze that documentand try to figure out where
(27:36):
the the overall,you know, what they were referring to.
This was a very tricky rolethat I didn't have a ton of success
with them being able to analyze it,but when I actually spoke to our kind of
compliance partner and this, they alsocouldn't figure out exactly where it was.
And so it ended upbeing one of those wacky Finra things.
But by just simply askingfor an explanation,
(27:57):
it kind of got me to a pointwhere I could figure out, okay,
I don't think that this actually applies.
I think some of themjust being really picky with, this rule,
and I could actually understandit enough to have it to your point,
have a conversation where I felt confidentthat, you know,
we could push back and say,I think this needs to be re reviewed
(28:18):
because this doesn't make any sense.
And so it cut down, you know, normallybefore this, I've done this in the past
where you have to go find the Finraright here all through the document.
Try to figure it out on your own,have internal conversations.
And I was able to cut down on all of thattime.
And, you know, so at the time I processeddealing with Finra or compliance,
but at least again, I was going into itwith some kind of educated,
(28:42):
talking points about, you know, whyI thought that it should be reviewed
and we could leave it as a soI think that is certainly a helpful point.
And I think people just need to kind of,use common sense
around, you know, complianceand AI and how to make sure that you're
not going to get yourselfinto any issues there with compliance.
But I think it actually has a huge,
(29:03):
benefit, especially if you're someonewho's, you know,
taking meeting notesand you need to submit things like that
to compliance, havingI go through those notes and, you know,
format in a way that compliance wants it,there can be a lot of, you know, pros
for the documentation side of,of how to use these tools,
to help with your compliance departmentand make it
(29:24):
less of a headache for, you know,anyone who has to deal with that.
And also, as an aside,and you know, presumably this will be,
an example ofAI being used to everyone's benefit
except for maybe the compliance officerswho could very well,
I mean, compliance could become, one ofthose areas where there is job loss,
(29:46):
to some degree,because, the kinds of things they do
are very, in alignment withthe kinds of things that AI model can do.
And so,you know, it's either 1 or 2 things.
One, everybody will have their ownvirtual compliance officer,
kind of like what we're doingwith our compliance pre-check tool.
And they'll be just they'll simply be lessfriction in the compliance process
(30:09):
because, the advisor will have their ownvirtual compliance officer
and everything will be kind of caughtand addressed before it's ever submitted.
But then on the other,the other end in Finra, the person
reading the compliance submissions,you know, could very well, you know, be a
AI agentthat is being overseen by a human being.
(30:30):
But the point being thatwe also could see,
you know, massive speed upin, responses and answers from compliance.
So therefore, you know,there could be, tremendous
new efficiency gains in the complianceworld, in the near future.
And it would be my predictionthat the compliance world
is pretty differentfive years from now, for sure.
Yeah, I think that's definitely true.
(30:52):
And so, you know,kind of continuing on this trend here,
this is a very kind of untapped areaamong advisors.
This, kind of advanced AI applicationsthat we've been seeing
in the survey results.
But reports that only about 4% are usingAI for kind of personalized
financial planning, which we talkedabout a little bit earlier.
But, I mean, I think there'sa lot of potential growth in this area.
(31:13):
Would you agree?
And do you have any kind of insightson where we might see AI being built
into that personalizedfinancial planning aspect of the business?
And, you know, by the way,just to remind you, the question was about
what was related to using AIfor multilingual
capabilities and personalizedfinancial planning. Yes.
Thank you.
(31:33):
Yes. And, you know, it's that multilingualthing that really catches my attention.
You know, I live here in new Jersey,and I think I heard recently
that there are,you know, that there are 50 different
languages spoken and New Jersey's,you know,
how many people we had 7or 8 million or something like that.
We're one of the most denselypopulated, states in the union.
We're close to New York City andPhiladelphia, major port of entry here.
(31:54):
So there are a lot of peoplein this state,
whose native language is not English.
They're here, their families are working,they're making money.
They have their American citizens.
They have financial planning issues now.
And, thethe question that comes up here is
being able is for an advisorwho is, you know,
(32:18):
may have a client who, you know, maybeyou have a client who speaks English.
And but they have, a parent, with themand maybe they're from Afghanistan and,
the parent is involvedin, it's their money.
But the younger relative who speaks
English is the one who interfaceswith the financial advisor.
(32:38):
And, in the past, it could be
that, you know,the son is translating some
financial concepts to the parentsor whatever the case may be.
But these AI models can speakall of these languages, right?
So now you have the ability to, take,
your existing tools and your existingadvice that are written in English and,
(33:03):
translate them, instantly, into,
you know, usable bits of informationthat can
then be shared with the,speaker of a non-English language.
And it can, you know, be very helpful
in those kinds of difficult situations,the difficult meeting situations.
But that's just one example, right?
(33:23):
Now we know and have heard over the years,for instance, you know, a horse's mouth
savvy Social Security planning program is,you know, many advisors have built,
you know, massive businessesaround bleeding with Social Security
planning, in their communities,as a way of meeting people.
And, you know, the Social Security keyaspect of retirement.
(33:43):
And we've had some people over the yearswho, you know, spoke,
Spanish and, had Spanish speaking clients.
And, you know,they would kind of translate on their own
some of the, you know, information,
the market and financial education,materials in Social Security.
But now, you know, it it franklywould not be very difficult to like,
(34:05):
you know, take a,you know, take a presentation for social,
Social Security, written in Englishand have it translated into Spanish.
And I know it sounds crazy, but,
you know, it could probably just be donein just a couple of minutes.
And, and, you know, as long as you havesomebody who does speak the language,
who can then kind of go through and say,yes, this makes sense.
(34:26):
And it's properly translated.
But so that's, you know, that's,that's another example of, using it.
But, you know, the question also is aboutpersonalized financial planning.
And, you know,one of the things that you're able to do
is, take very specific financial planning
or investment management challengesthat you're having with your clients.
And you can take your scenario to theAI model.
(34:50):
When but when but when I say AI model,I mean ChatGPT or Claude or Gemini.
You know, most in our in our programs,we have everybody use ChatGPT
for a variety of reasons.
But there are other models out thereas well.
So, to be semi neutral, I say AI models.
But anyway, the point is that, you can go,
to these models and with, you know, say,hey, look, I'm meeting with this client
(35:14):
or this particular typeof financial planning challenge.
Here's the situation.
You're not sharingpersonal identifiable information.
You don't need to beat that granular level to get very, very,
useful and effective, advice, guidance or,brainstorming,
in terms of trying to see what,you know, how
(35:35):
we can, not this particularfinancial planning challenge.
So in that respect, we can say, yeah,there are ways
it's only 4%that are doing it or realizing it.
And so we're going to be doinga lot more on this going forward as well,
because, you can in your eyes,I've done a number of times
in our conversation this morning,I've invoked various, various expertise.
Right.
And so you can have a conversation,with the AI model,
(35:58):
and you can invoke the role of a certifiedfinancial planner with 30 years
experience working with,you know, whatever, whatever
subtopic within financial planningthat you're, you're trying to address.
You can cite that as part of your prompt.
Then, you know, includethat as part of your, part of the task of,
like, we've got this problem,this challenge.
(36:20):
We've got to make some,we want to come up with some ideas of,
what might be the best solution for thisparticular financial planning challenge.
And, like, we were talking about strategy,broad strategy
for your company at the top of the, hour.
You know, now, you could be havinga similar type of conversation
in the realm of financial planning,when you've given the AI model,
(36:43):
you know, the facts of the situation,and that can be, very effective as well.
Yeah, I think that's great.
And your first example kind of reminded meto have, you know, we have advisors
who, also, you know, have Americanand Canadian clients.
And sometimes, you know, we're asked to,you know, does this stuff apply to Canada?
And I think, you know,
(37:03):
if you're in that areawhere maybe you're unsure
about some of the rulesyou have for clients
who are moving overseasand you want to help them with that,
even if they're going to an Englishspeaking country.
You can kind of use AI to help,
you know,get schooled up on some of the differences
that they might be facing, you know,as they move to a different country
or reside in a different countryas well. Right?
I mean, those cultural,you know, those cultural differences,
(37:25):
cultural personalization, you know.
Absolutely.
And, you know, also,just simply using it to like, okay,
you'vegot some clients coming in there from,
you know,they're successful professionals.
But they're from,you know, Iran or Poland or,
you know, or Chile or whateverthe case may be.
And, you know,you don't know anything about
(37:47):
Iran or Poland or Chile other than thatthey're not the United States.
But in advance of the meeting,you know, they're going to come in,
maybe it's the first meeting and,you know, you could again,
say to the AI, hey, I'm about to havethis, meeting with this couple.
You know, they're a
successful, tech entrepreneur, from Chile.
They want a financial advisor,to help them with their investments.
(38:10):
And, you know,I don't know anything about Chile.
Kind of give me a little briefing on it.
And, give me some, you know,light hearted without making jokes
or anything like that. So just help me
learn a little bit about themand the history of their country.
And all that sort of thing,maybe as a way of sort of enhancing,
or, you know, enhancing the relationshipor making some,
(38:31):
you know, making some smallbut meaningful connection.
In the, yeah, in the meeting.
Right. Yeah. Yeah, I think that's great.
So as we kind of close out this sessionhere for today for advisors
who haven't started kind of experimentingaround this area of AI in their business,
but one simple way that they could kind ofget started this week,
we're thinking about,these more advanced use cases for AI.
(38:55):
Good question.
Well, you know,
I think the most important thing forpeople, is to start using it, right?
And, that's,
you know, Ethan Malik, the professorof innovation at the Wharton School,
the University of Pennsylvania says,that we need to be thinking
about these AI's as a co intelligencethat can collaborate with us.
(39:16):
We're in charge, though.
And so but he says it takes about 8to 10 hours of really working with them.
And so and which of course iswhy we've created these training programs
because we're translatingfor financial advisors
100% all our programs,our use cases for advisors.
So we're not we're not talkingabout broad, uses or general uses of AI.
(39:39):
We're talking about very,very specific kinds of things.
So you know, I would just tellsomebody to, think about a challenge
or think about a challenge or problemyou're having in your business right now.
Think aboutwho would be a really smart person
who could offer you some insightabout your challenge.
Explain that situation in a promptand then,
have a conversation about itand just see, see where it goes.
(40:02):
Right.
But again, I mentioned, you know, thereis this learning.
This is super powerful.
New, tool.
We have seen, you know, we're seeing
big growth and adoption by advisors.
We did this cert.
We did our first survey back inMay and June of last year.
We had like 12% of advisorswho were using AI on a daily basis.
(40:25):
Now our our current surveythat's in the field,
we're looking at, in the 24 to 26% range.
So basically a doubling in six months.
These are advisorswho are recognizing, wow,
this thing is really powerfuland can be very, very helpful with me.
It can help me save time.
It can, produce high quality work.
(40:47):
And it can also, serve to,
inform me, and brainstorm with me
across all sorts of different fronts.
But again, you know, 8 to 10 hoursis what you really need, but for what?
To start to answer your question, Devin.
Yeah, taketake one of your problems to it.
Invoke some expertise.
And, remember, you're not
(41:09):
this is not the chat box looks.
The chatbox looks too much like a search box.
We know, we see search.
We see advisorssearching on, horse's mouth all the time.
And we know people search on 1 or 2 wordsand basically.
Right.You can't do that with the AI models.
You have to have a robust conversationand back and forth
and you'll be amazed at what you learn.
So anyway, Devin, I'll I'll leave itat that if you just want to,
(41:33):
you know, pointpeople to where they can go to be, or you.
Yeah.
If they still want to sign up
for our program, which starts Thursdayand runs for 30 days in a row,
why don't you tell thema little bit about that?
And so that thank you, everyonefor being here with us today.
And thanks for joining me for this chat.
As Sean mentioned,
our AI powered financial advisor program,which is kind of our intro to AI.
(41:53):
It's a four weekprogram begins on Thursday, March 6th.
You can learn moreby going to your horse's mouth.com/ai.
We've had over
100 advisors go through this program nowat this point, and three different groups,
with great success.
So we're really looking forwardto kicking off another group,
this Thursday, March 6thand then starting in April, on
April 3rd, we are doing ourAI marketing for advisors workshop.
(42:16):
This is a five week programthat really gets into the nitty gritty
of how to use AI in your marketing.
We teach you how to kind of come awaywith a full marketing plan.
You know, that you can then implementthroughout the year, and we talk about
how you can use AI in all different partsof that plan to actually,
make your marketing come to life.
So if you want to learn more about that,you can go to horse's mouth.com/i
(42:38):
marketing. It's on the screen right nowfor you there.
We hope to see you at one of our events.
And thanks for being, part of thislive stream today.
Sean, thanks for joining me,and we hope to see you again soon.
Hey, Kenya.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.