Latest / The Efficient Advisor: Tactical Business Advice for Financial Planners / 199: AI Overhaul---Creative Ways to Use AI in Your Advisory Business w/ Guest Ben Olsen

Transcript

199: AI Overhaul---Creative Ways to Use AI in Your Advisory Business w/ Guest Ben Olsen · The Efficient Advisor: Tactical Business Advice for Financial Planners

• 0:00
  Speaker 1: Okay, so we've talked about AI on the show before, but this is next level stuff.
• 0:07
  Speaker 1: I'm talking about way more than just using chat GPT here, although that's an amazingly cool tool and I absolutely love using it.
• 0:15
  Speaker 1: But today we are getting into efficiency to the max.
• 0:19
  Speaker 1: We're going to nerd out with AI automation expert Ben Olson.
• 0:23
  Speaker 1: Ben is building AI for advisors.
• 0:26
  Speaker 1: That's literally replacing jobs, and I don't mean that in a bad way.
• 0:30
  Speaker 1: I mean replacing tasks so that you can redeploy your team members' efforts into way more effective areas like leveling up your client experience.
• 0:41
  Speaker 1: Just imagine having and training an AI assistant to write emails, do meeting summaries, organize your client service model Like literally.
• 0:51
  Speaker 1: It's crazy what's possible.
• 0:53
  Speaker 1: And we're going to talk about all of that and how it actually works.
• 0:57
  Speaker 1: My mind was blown.
• 1:00
  Speaker 1: All right, let's go.
• 1:13
  Speaker 1: Okay, but right before we dive into this conversation with Ben, I want to tease an upcoming episode that I'm super excited about and I want to share something.
• 1:22
  Speaker 1: So first, in the spirit of talking about AI, one of the biggest things stressing advisors out is security of it.
• 1:30
  Speaker 1: Right, how do we protect our clients' data when using stuff that's open source AI?
• 1:35
  Speaker 1: This has been a big hangup, especially for meeting notes, and I've got a guest coming soon that is going to answer this and so much more.
• 1:41
  Speaker 1: There is finally an incredible AI note-taking tool that is safe.
• 1:43
  Speaker 1: I cannot wait to answer this and so much more.
• 1:44
  Speaker 1: There is finally an incredible AI note-taking tool that is safe.
• 1:47
  Speaker 1: I cannot wait to share this with you and, for those of you that are like, that sounds amazing, I can't wait for this conversation.
• 1:53
  Speaker 1: I have got a link for you to check it out, and, since I'm going to be helping them by giving feedback on part of the tool, I've got a 10% discount for my listeners.
• 2:04
  Speaker 1: So, yes, I was like, yeah, sure I can do that, and you're going to have to give me a little hookup for the listeners, and they were like let's consider that done.
• 2:11
  Speaker 1: So I will have a link for that in the show notes if you can't help yourself.
• 2:14
  Speaker 1: And they've got, um, all kinds of cool illustrations out there or you know videos you can watch and how it works.
• 2:21
  Speaker 1: So and I will be sharing more, there was going to be a whole episode coming up about the security and how they back it, and it's awesome.
• 2:27
  Speaker 1: I'm so excited.
• 2:28
  Speaker 1: The other thing I wanted to share quick is we are going to be opening up registration for the group coaching program that starts this August.
• 2:34
  Speaker 1: It's going to open soon, so I've got a special discount and deal for people on the wait list.
• 2:39
  Speaker 1: So if you want to join us for our 10 month coaching program, get on the wait list and save some cash.
• 2:46
  Speaker 1: If you're not sure what the heck I'm even talking about, you can check out the coaching page of the website for more details, and I'll link both the wait list and the details page in the show notes below.
• 2:57
  Speaker 1: Ooh, and one more thing worth noting.
• 2:59
  Speaker 1: So in the new group coaching program, we are inviting all kinds of experts into the program.
• 3:08
  Speaker 1: So you've obviously got me and then you've got all of the people around you, but we are bringing in guests like Matt Jarvis, ben Olson, deirdre Van Nest, robin Crane.
• 3:18
  Speaker 1: We've got Kate Gulen.
• 3:20
  Speaker 1: There's a whole bunch of people going to be speaking into this program and creating content and giving resources for us, and I wanted to share that.
• 3:27
  Speaker 1: Ben is one of them.
• 3:28
  Speaker 1: So Ben did, and those of you who are in the group program it's coming.
• 3:32
  Speaker 1: I'm so excited for you to watch it.
• 3:33
  Speaker 1: Ben did a whole series for us on how to use chat GPT in your business.
• 3:39
  Speaker 1: So all of these really creative and unique ways that he can see advisors integrating the free or even the little premium priced chat GPT inside of your program.
• 3:50
  Speaker 1: So keep your eyes peeled for that.
• 3:52
  Speaker 1: But, ben, you're going to love him.
• 3:54
  Speaker 1: He is super interesting, knows his stuff, and so I just wanted to share that.
• 3:59
  Speaker 1: He is part of the Systems of Scale group coaching program now, all right, let's dive in.
• 4:04
  Speaker 1: Of the Systems of Scale group coaching program.
• 4:08
  Speaker 1: Now, all right, let's dive in.
• 4:10
  Speaker 1: Okay, so this interview has been a long time in the making and I feel like every time I interview someone I say I'm really excited for the interview, but this time I genuinely, genuinely mean it.
• 4:20
  Speaker 1: I have been polling advisors in the different small group coaching programs and online all around this topic.
• 4:27
  Speaker 1: So I became aware of you, ben.
• 4:30
  Speaker 1: So my bestie Michelle, who has been on the episode or has been on the podcast I think she has two episodes a couple of times was sharing with me all of this really cool kind of cutting edge AI stuff that she was doing in her business and my mind was absolutely blown and she was like you have got to have Ben on the podcast.
• 4:49
  Speaker 1: So I reached out to Ben, ben, and I have been talking I feel like we've been chatting for a really, really long time, and even from the time that we started talking to today, recording AI has changed and progressed so so much.
• 5:03
  Speaker 1: So I am extra excited, and I know the advisors in the communities that I've been talking to in preparation for this episode are really, really fired up for this conversation.
• 5:13
  Speaker 1: So, ben, thank you so so much for being here.
• 5:16
  Speaker 2: Yeah, absolutely.
• 5:17
  Speaker 2: Thank you for having me Super excited to be here.
• 5:19
  Speaker 2: I've been looking forward to it as well.
• 5:21
  Speaker 2: And yeah, you know, ai is changing exponentially.
• 5:24
  Speaker 2: So lots of, lots of covers since the last time we spoke, just six months ago or so.
• 5:29
  Speaker 1: Yeah, so I, I, I I'm going to preface this with use small words that I can understand when it comes to technology, and I was, I was sharing with you before we started recording that I was.
• 5:40
  Speaker 1: I've been joking, but my son, my 15 year old, and I started watching.
• 5:43
  Speaker 1: I started watching for the first time ever.
• 5:45
  Speaker 1: I am remiss to admit that I have never seen the Terminator movies until now, so I have been watching them diligently, taking notes in preparation for Judgment Day, because now I think I fully understand what AI is capable of.
• 5:57
  Speaker 1: But before we get really into it, why don't you share what your background is and sort of how you fell into the AI space for, specifically, financial advisors?
• 6:08
  Speaker 2: Yeah, absolutely.
• 6:09
  Speaker 2: So.
• 6:09
  Speaker 2: I went to Indiana University so shout out to any fellow Hoosiers listening I started my career at an independent RIA here in Chicago, a smaller firm of about 50 individuals and they specialize in retirement planning, so it's kind of a classic entry level job.
• 6:28
  Speaker 2: I was doing a lot of cold calling, helping advisors build their books of business, was also doing some financial planning as well.
• 6:34
  Speaker 2: After doing that for a little bit I moved over to Merrill Lynch where I was working on building my own book of business.
• 6:41
  Speaker 2: And I think one of the things that really pulled me into the world of being a financial advisor is the entrepreneurial nature of it and kind of, you know, the idea that there's really is no ceiling and you can build your business as you see fit and as large as you want it, and that's something I always valued.
• 7:00
  Speaker 2: But I found myself becoming increasingly interested in kind of coupling that with technology.
• 7:05
  Speaker 2: So I started doing a lot exploring technology in my free time, started teaching myself how to code and eventually I kind of realized I was kind of just better suited to work in technology.
• 7:17
  Speaker 2: So I left Merrill Lynch and started working at a financial technology startup.
• 7:23
  Speaker 2: Started, you know, sharpening my skills, did a master's degree in data science while I was working there as well, and fast forward to today, you know I kind of had some a few years of experience at various startups doing data science and machine learning, and now what I do is I help companies develop.
• 7:45
  Speaker 2: AI software and I've been doing it for about a year and when I first got started I actually reached out to some of my former connections that were financial advisors to see if they wanted to explore some ideas and do some business with me, and I found that they had a great need for AI and that the solutions I was building for them were awesome and they love them.
• 8:08
  Speaker 2: So I decided to kind of pivot into solely focusing on the independent advisor and RIA market and helping them use.
• 8:18
  Speaker 2: AI and workflow automation to operate more efficiently, and that is what I'm doing today.
• 8:24
  Speaker 1: Okay.
• 8:24
  Speaker 1: So I love that.
• 8:25
  Speaker 1: And I remember the first time we talked and you were kind of telling me about all of the things that AI can do, and I remember like probably playing it off really cool, like yeah, yeah, yeah, cool, cool, cool.
• 8:34
  Speaker 1: I knew that, I knew that, or maybe I didn't, maybe I was just like what that's bananas?
• 8:39
  Speaker 2: I think you did say exactly that.
• 8:41
  Speaker 2: I'm just kidding.
• 8:43
  Speaker 1: I'm not really cool at pulling off anything and playing cool.
• 8:45
  Speaker 1: I have zero.
• 8:46
  Speaker 1: I have zero chill, so I tend to just get really excited.
• 8:49
  Speaker 1: But you were sharing all of these things and my mind was kind of expanding and stretching while we were talking because up to that point, like I have you know, I use chat, gpt and I use a few other AI tools inside of my business and at the point we were talking, it was really mostly through like, and I think where most advisors are at now is how can I use AI for marketing?
• 9:14
  Speaker 1: right, like that seems to be the lens that most people like how can I have it write a blog post or social media or create different collateral inside of my business?
• 9:24
  Speaker 1: And I've seen a little bit more progression of people saying, oh my gosh, I'm using it to write investment policy statements and things that are a little bit, you know, kind of tiptoeing into the world of AI in a way to, you know, really support the business.
• 9:38
  Speaker 1: But when you and I were talking, it was like mind boggling because it was like all right, I get hyped up about operations and what you were sharing was like teaching this tool to like go into all of your different systems and read stuff and really truly I hate to even say it this way but replace the need for a lot of people on your team, like, and it was just like, wow, this is really life altering stuff for advisors.
• 10:03
  Speaker 1: So I was just totally blown away by all of it.
• 10:07
  Speaker 1: So I'm kind of curious about where AI is at, like, what is kind of the current state, what are these recent advancements and what is all of the stuff that it's kind of capable of.
• 10:19
  Speaker 1: We'll kind of start high level and then we'll start kind of chunking this down if possible of start high level and then we'll start kind of chunking this down if possible.
• 10:27
  Speaker 2: Yeah, absolutely so.
• 10:31
  Speaker 2: All the stuff you're hearing about now and the technology that's been causing all of this attention to be focused on AI in the last few years is generative AI.
• 10:35
  Speaker 2: And to kind of back up a second and talk about why that's different than what we traditionally consider AI, historically, ai is like a technology that was very good at like interpreting information, finding patterns and like classifying things.
• 10:52
  Speaker 2: So, to give like a more concrete example, think about if you were talking to like a customer support chatbot on like a website that you shop on or something, and you were like hey, I placed a refund request last week.
• 11:05
  Speaker 2: When can I expect to get my money back?
• 11:07
  Speaker 2: And then the chatbot would say something like in order to place a refund request, follow these steps right.
• 11:13
  Speaker 2: Like we've all had that experience with a chatbot where you're like oh, like you understand directionally what I'm saying, you don't actually get it.
• 11:20
  Speaker 2: The reason that was happened that happens is because the model behind that is just looking at what you said and it's looking at its kind of predefined canned answers and it's saying which ones are most similar and whichever one ranks the highest, it's going to respond with that.
• 11:36
  Speaker 2: So that's how AI traditionally works and that's like an example of how it was good at, you know, finding patterns and things like that.
• 11:45
  Speaker 2: And what we've seen recently with the advancement of some crazy technologies like deep learning and transformers which I won't get into, but if people want to look those up, please do is that AI can actually now generate information.
• 12:00
  Speaker 2: So it's not just interpreting it, but it's actually creating it.
• 12:04
  Speaker 2: So it's not just interpreting it, but it's actually creating it.
• 12:07
  Speaker 2: And that's why you probably have had maybe some experiences today where, if you talk to a chatbot and you're like hey, my name is Ben, what's the status of my refund request?
• 12:13
  Speaker 2: It's going to reply and say hi, ben, you can find more information about that here.
• 12:17
  Speaker 2: Right, it wasn't explicitly told to say your name, but it can write things on the fly.
• 12:23
  Speaker 2: So that's kind of the big shift that we've seen is that AI can now generate and this is spreading across what they call like modals.
• 12:32
  Speaker 2: So not only can it generate text, but it can also generate images, it can generate videos and things of that nature.
• 12:40
  Speaker 2: So that, in of itself, the ability for AI to generate and create information is what everyone's so excited about and what kind of creates these massive opportunities to kind of augment how we work and automate various parts of our work in our lives?
• 12:59
  Speaker 1: of crazy.
• 13:00
  Speaker 1: Okay, I love it.
• 13:01
  Speaker 1: So, so all that cool stuff you just said how does an advisor like how does it really fit into an advisor business, like on a day-to-day basis, like how, how do you see, or how are you helping?
• 13:17
  Speaker 1: Advisors currently really integrating AI in a way that makes just operating so much more efficient.
• 13:22
  Speaker 2: Yeah, so what we do is we help advisors integrate AI in kind of like a custom case-by-case basis.
• 13:30
  Speaker 2: So we'll look at the advisor's tech stack, what their pain points are and build custom solutions.
• 13:35
  Speaker 2: But even before that, I think there are a lot of things that advisors can start doing themselves and tools that they can be using themselves that are already out there on the market.
• 13:45
  Speaker 2: Obviously, like chat, gpt is a really big one.
• 13:49
  Speaker 2: I would encourage everyone to use it.
• 13:51
  Speaker 2: If they're not already, try to use it at least once or twice every day.
• 13:55
  Speaker 2: I have it pulled up on my computer all day long and I use it.
• 13:59
  Speaker 2: It's kind of like my first place to go instead of like Googling.
• 14:03
  Speaker 2: Now, also, there are some great like kind of going off that idea of Googling.
• 14:11
  Speaker 2: There's another technology called perplexity AI.
• 14:13
  Speaker 2: I use this one every day as well.
• 14:15
  Speaker 2: Essentially, it's like if Google and a chatbot combined.
• 14:19
  Speaker 2: So if you're ever you know, I remember when I was working as an advisor, I would start my morning by just like reading a bunch of like Wall Street Journal and CNBC articles to try to figure out, like what's happening in the markets today, what are the top stories, whatever, and perplexity is basically something that you can search the web with and it will read.
• 14:40
  Speaker 2: You ask a question.
• 14:41
  Speaker 2: It will read all of the search results and then summarize them to contextualize your question.
• 14:46
  Speaker 2: So that's another one.
• 14:47
  Speaker 2: And then, yeah, there's just so many more.
• 14:50
  Speaker 2: One of my favorites is like Fireflies AI.
• 14:53
  Speaker 2: It's a meeting summarizer.
• 14:54
  Speaker 2: You know, that's one that will kind of just hang out with you on your calls, record everything, give you summaries and stuff like that.
• 15:00
  Speaker 2: Zapier is another one I love.
• 15:04
  Speaker 2: That's a little bit less of a pure AI product, more of kind of a workflow automation tool that allows you to strap together different tools you use and get them talking to each other.
• 15:13
  Speaker 2: So I know that was like a long list, but these are all tools that you don't need to hire someone to build or integrate for you.
• 15:20
  Speaker 2: I would encourage advisors to go start playing around with them.
• 15:23
  Speaker 2: Try to use them a little bit every day and get comfortable with them.
• 15:27
  Speaker 1: So where?
• 15:28
  Speaker 1: So for someone who's wanting to like get good at chat, gpt or perplexity yeah, using it a couple of times a day for sure, but where else can they go to get better at prompts or asking?
• 15:41
  Speaker 1: Really, cause right, the better you become at asking for the right thing, just like anything, right?
• 15:46
  Speaker 1: Garbage in garbage out.
• 15:47
  Speaker 1: The better you get at asking for stuff, the better information or you know content that you're getting back.
• 15:53
  Speaker 1: That's right.
• 15:54
  Speaker 2: So, yeah, I think what you're describing is like what people are calling prompt engineering, um, which I think is a little bit of an overkill to like make it that big of a, because at the end of the day, it's just kind of communication skills.
• 16:06
  Speaker 2: But there is, you know the thing.
• 16:10
  Speaker 2: Like you said, garbage in, garbage out and high quality inputs generate high quality outputs.
• 16:14
  Speaker 2: So there are a lot of things you can learn on YouTube.
• 16:17
  Speaker 2: I absolutely love YouTube.
• 16:19
  Speaker 2: You could just go on YouTube and type in prompt engineering.
• 16:23
  Speaker 2: Look at a popular video that was recently uploaded and you'll learn a ton.
• 16:27
  Speaker 2: I wouldn't stress that like too much, like I wouldn't worry about becoming the best prompt engineer.
• 16:34
  Speaker 2: It's kind of one of those things where, like, the more you use it, the more comfortable you'll get, the more you'll understand how to communicate with it.
• 16:40
  Speaker 2: It's really kind of like working with a colleague, right.
• 16:42
  Speaker 2: When you first start working with somebody, you don't know how to most effectively communicate with them, so maybe at first you kind of over-communicate and then you learn how they think and then you start communicating more efficiently.
• 16:53
  Speaker 2: That's kind of been like my experience with using ChatGPT as well.
• 16:57
  Speaker 2: So I would say, go on YouTube, read about prompt engineering, but don't feel like you need to put tons of time and effort into mastering that.
• 17:09
  Speaker 1: Okay, and one more kind of basic question about chat, gpt, because I actually was just talking to an advisor and I asked him this question.
• 17:12
  Speaker 1: I was like, ooh, I'll ask Ben.
• 17:13
  Speaker 1: So you know, you always hear like it learns what you like.
• 17:16
  Speaker 1: So how do you tell it like, hey, like so for me.
• 17:20
  Speaker 1: I am a podcast host and a speaker.
• 17:23
  Speaker 1: I work with financial advisors.
• 17:24
  Speaker 1: This is the voice that I like to talk in.
• 17:32
  Speaker 2: Like how do I teach it those things that it remembers it time and time again?
• 17:38
  Speaker 2: Yeah, so there's kind of like three different ways to do that currently with ChatGPT.
• 17:40
  Speaker 2: First and foremost, they recently introduced long-term memory, so the model itself, while you're talking to it now, is actually going to pick up on things that it thinks it should remember, like it'll be.
• 17:51
  Speaker 2: Oh, that's an important thing about Libby that she just mentioned.
• 17:54
  Speaker 2: I'm going to remember that and it'll actually show you on the screen.
• 17:57
  Speaker 2: It'll say memory updated when you type in certain things.
• 18:01
  Speaker 2: So it's kind of doing itself, doing that itself in the background.
• 18:05
  Speaker 2: Already.
• 18:06
  Speaker 2: Um, with the most recent model Um, and that's a feature that, like, you can turn off if you don't want it to do that.
• 18:11
  Speaker 2: Um.
• 18:12
  Speaker 2: The other option is to create what are called um GPTs.
• 18:17
  Speaker 2: Um, I think these are part of, like the, the premium subscription Um.
• 18:21
  Speaker 2: It's like 20 bucks a month or something like that.
• 18:23
  Speaker 2: Um, and essentially that's like your own version of chat GPT that you get to kind of customize a little bit and you're able to upload documents and files that you know have information that's personalized to you.
• 18:36
  Speaker 2: So in you know, in that case you could just draft up a document explaining who you are, what your goals are, what you want to get out of chat GPT, how you want it to help you and you can upload it to the GPT and it's going to reference that every time it replies to you.
• 18:51
  Speaker 2: The last option is to just literally drop a file into the chat, or drop like text into the chat and just say like this is information about me.
• 19:00
  Speaker 2: Use this when you're replying.
• 19:03
  Speaker 1: So interesting.
• 19:03
  Speaker 1: Okay, so thinking about it, writing stuff for you, creating content for you when we were talking like however long ago that was you were sharing a lot of the cool stuff that you're creating for advisors that helps with the operational side.
• 19:20
  Speaker 1: Tell us a little bit more about how you see all of the advisors that you're working with.
• 19:25
  Speaker 1: What are some of the most common things that you're building AI to do for them?
• 19:30
  Speaker 1: So we're all prepared Everybody.
• 19:32
  Speaker 1: If you're multitasking, come back to us and prepare to have your minds blown.
• 19:35
  Speaker 1: Okay and go.
• 19:38
  Speaker 2: Yeah.
• 19:38
  Speaker 2: So this kind of gets into my favorite thing in this space right now, which is AI agents, and these are think of them as kind of chatbots that can execute tasks for you and they can go into your software and do things for you.
• 19:53
  Speaker 2: So, in terms of like how this works, it starts by obtaining access to your technology.
• 20:02
  Speaker 2: So like, if you look at the typical advisor tech stack, things like Outlook, redtail, wealthbox, emoney all these technologies that have what's called an API are able to be integrated with, and essentially what that means is that software is allowed to access the technology, kind of behind the scenes, right, like it's not going to like log in and click to the website, but it's going to kind of take a backdoor into the technology and be able to access it.
• 20:31
  Speaker 2: And what happens when you combine that with chatbots is that they can actually reason about doing things inside of these technologies.
• 20:40
  Speaker 2: So an example would be if you set up an agent that has access to your Redtail CRM, it knows how to look through all of your data on Redtail.
• 20:52
  Speaker 2: So you could say when was the last time I had a meeting with Ben Olson and it will run a search that looks for my profile and look for that data point of when the last meeting was, and it might say you know you met with Ben 12 months ago and you could say, okay, I need to get him back on the calendar, can you please schedule something with him?
• 21:12
  Speaker 2: And he'll say, okay, do you want me to draft an email?
• 21:13
  Speaker 2: Right?
• 21:13
  Speaker 2: And you can say yes, and it will pull in information about me from Redtail, draft up that email and then, if it's also synced up with Outlook, it can then put that like in your drafts folder or your outbox.
• 21:24
  Speaker 2: So yeah, that's kind of like one use case.
• 21:27
  Speaker 2: Most of the use cases that I see advisors using right now is primarily centered around like client communications, so meeting follow-ups, you know regular email updates, newsletters, things like that.
• 21:44
  Speaker 1: Okay, so, so you mentioned before, like you know, that your job is to come in and solve some of these pain points.
• 21:50
  Speaker 1: So what do you gosh?
• 21:53
  Speaker 1: I'm like okay, my like my brain is trying to keep up with all it.
• 21:56
  Speaker 1: Like I, I am operating on a very old operating system, um being that I'm in my forties now, so I'm trying to think, like how I describe okay, so, like it can go in and it can look at Redtail, it can look at eMoney, it can draft emails for you.
• 22:11
  Speaker 1: How do you like?
• 22:14
  Speaker 1: What you were telling me before, if I remember correctly, was like that you can continue to train it and like refine it and teach it to like read meeting notes.
• 22:23
  Speaker 1: And I mean, am I, am I understanding that Right?
• 22:25
  Speaker 1: Like it can go and like read things and say, like here's what you did last time.
• 22:28
  Speaker 1: Like so you can ask it right.
• 22:30
  Speaker 1: Like hey, I met with Ben, what did we talk about?
• 22:33
  Speaker 1: Like summarize what we talked about and what we said we were going to need to do in the future.
• 22:37
  Speaker 1: Or like, like give us some more really tactical stuff.
• 22:40
  Speaker 1: That like what are the pain points that advisors are expressing?
• 22:42
  Speaker 1: And then I definitely want to get into meeting notes because I feel like that is like a hands down, like one of the largest pain points for everyone everywhere.
• 22:48
  Speaker 1: But what other like super operationally stuff do you see, cause it probably feels really basic to you, but for someone like me, I'm like, what it can?
• 22:57
  Speaker 1: It can do that.
• 22:58
  Speaker 1: So what else are you?
• 22:59
  Speaker 1: What other pain points are you solving?
• 23:01
  Speaker 2: Yeah, so one of the ones that we we are doing a lot for our clients right now is creating like reports and documents that the advisors are sitting down and giving to their clients or going over in meetings.
• 23:14
  Speaker 2: So, and this kind of ties into this idea of bringing your tech stack together and using AI at the same time.
• 23:22
  Speaker 2: So I know, you know there's lots of advisors out there that when they create a report or a document to go over with a client, they're pulling in data from the CRM, from eMoney, from whatever else, and they're kind of manually typing it into, maybe like a Google Doc or an Excel file, and then it takes them quite a lot of time to create, you know this final document that they send to the client or print out and go over with them in a meeting.
• 23:47
  Speaker 2: So one of the things that you can do with AI is get all of those technologies to talk to each other, so you're not manually going in and pulling the data out of the software and dumping it into you know Google Sheets or whatever.
• 24:00
  Speaker 2: You can have that happen automatically and you can have it.
• 24:03
  Speaker 2: You know draft commentary, like draft text, about what the data is saying, so you can have it write out little summaries and things like that.
• 24:13
  Speaker 2: So yeah, I'd say like document and report generation is another huge use case.
• 24:18
  Speaker 1: That's so interesting.
• 24:19
  Speaker 1: Okay, so it could, essentially so.
• 24:22
  Speaker 1: I have a lot of advisors that are wanting to do like one page financial plans, right, and like how do I take, just instead of running hey, here's money guide pro and here's 47 pages of output to tell the story that I'm wanting to tell.
• 24:35
  Speaker 1: I really want to pull in just a couple charts and graphs from that and plop it here in the top left corner and have a little summary of what that means underneath it, and then a couple of recommendations here and maybe their rates of return from their brokerage accounts over here on the right hand side.
• 24:51
  Speaker 1: And that's basically what you're saying, that you can build it to just like do that customized to whatever the advisor is wanting to have.
• 24:59
  Speaker 2: Yeah, yeah, absolutely, and I will say, like taking a step back like this is relatively complicated.
• 25:06
  Speaker 2: To build this Like it's possible, but like this isn't something.
• 25:09
  Speaker 2: You just kind of like go log into chat GPT and you can just have it spit this out In order to like do things like this, like you do need to have like a custom built piece of software, but yeah, that's it's certainly possible and kind of the way it works is you basically give it the AI access to all of those different data points in those different softwares and you explain to it what they are and what they mean.
• 25:34
  Speaker 2: And then you also give it the ability to dump that data into a document and you explain to it how to do that.
• 25:41
  Speaker 2: And then you're able to kind of talk to it and say that like I want a report that has ABC followed by DEF, and it will suck all that information in from wherever it needs to be and format it.
• 25:55
  Speaker 2: It might not be perfectly formatted.
• 25:57
  Speaker 2: That's like a little bit of an issue that we run into sometimes.
• 26:01
  Speaker 2: The formatting can be like a little bit sloppy, but it's going to get you like 75 plus percent of the way there.
• 26:07
  Speaker 1: So yeah, that's amazing.
• 26:08
  Speaker 1: Okay, so what other pain points do your clients typically come to you with?
• 26:12
  Speaker 2: Yeah.
• 26:13
  Speaker 2: So I would say another one which we kind of talked about was was market is marketing.
• 26:17
  Speaker 2: Every advisor like wants to grow, but a lot of times they might not be doing much, especially online.
• 26:25
  Speaker 2: So AI can be really powerful for online marketing, especially like if you're on LinkedIn, in terms of generating content, sending messages to people, following up on messages and things like that.
• 26:39
  Speaker 2: So, yeah, I'd be happy to dive deeper into that if you want, but I think marketing is another big use case as well.
• 26:46
  Speaker 1: Okay, that's so interesting.
• 26:47
  Speaker 1: Okay, so some of the other stuff that I remember Michelle kind of sharing with me and I'm curious, like she was talking a lot about.
• 26:55
  Speaker 1: Yeah, like having almost like an assistant would kind of going into your client service model, right, and being able to say, hey, these are the clients that you typically meet with in July, like, can I prepare a list for you of everybody who you need to see, and I can go ahead and draft the emails to try to schedule them and like some of that kind of stuff.
• 27:16
  Speaker 1: So do you do a lot or do you see a lot around, kind of the organization of like client service model.
• 27:24
  Speaker 1: Like, hey, we said we were going to do this for this client on this day.
• 27:27
  Speaker 1: Like, can it remind you of stuff?
• 27:29
  Speaker 1: Yeah, okay.
• 27:31
  Speaker 2: Yeah, absolutely so.
• 27:32
  Speaker 2: As long as you have that information inside of one of these technologies or softwares, whether it be your CRM or wherever it, that's totally possible.
• 27:42
  Speaker 2: Um, and, yeah, that's, that's something that we we worked on with Michelle, where you know essentially we she about meetings and have follow-up points and things that she wanted to have her team do prior to the next meeting, and the system would you know, look at her calendar and say, okay, that meeting's coming up.
• 28:01
  Speaker 2: Here are the five things in her notes that need to get done, need to get done, and it would be able to notify her team members and add like to-dos for them to complete those tasks, like the week prior or whatever their desire Okay.
• 28:15
  Speaker 1: So when you say add to-dos, like can it create?
• 28:18
  Speaker 1: Okay.
• 28:18
  Speaker 1: So this is a question that I have Can it affect like your CRM, like can it go in and do like modify?
• 28:25
  Speaker 2: your.
• 28:25
  Speaker 1: CRM.
• 28:26
  Speaker 1: Yeah, that's a great point.
• 28:27
  Speaker 2: So a lot of what I've been explaining up into this point is about it reading data from your CRM.
• 28:33
  Speaker 2: But yeah, you can absolutely also have it write data or create data back into the CRM or whatever other technology, as long as those technologies support that through their API.
• 28:47
  Speaker 2: So the technology has to support the ability for the chatbot to kind of come in the back door and add information to it.
• 28:54
  Speaker 2: Most of them do not all of them do, but most of them do.
• 28:58
  Speaker 1: So, like the big ones, wealthbox Redtail those typically-.
• 29:02
  Speaker 2: E -Money Salesforce.
• 29:03
  Speaker 2: Yeah, yeah.
• 29:05
  Speaker 1: So okay, oh my gosh.
• 29:06
  Speaker 1: So you said E-money.
• 29:07
  Speaker 1: Okay, so can I tell my chat bot to like make updates to someone's e-money profile?
• 29:15
  Speaker 2: Yeah, yeah, that is supported through their API to be able to update the account.
• 29:21
  Speaker 2: So, yeah, essentially what that would look like is, without trying to get too technical here, you would basically write code that makes the updates to the account and then you would equip the chatbot with that code and it would use its ability to understand language to know when it should actually run that code.
• 29:44
  Speaker 2: So it's kind of like, yeah, you'd write the code and then the chatbot just reasons about when to actually run it and use it, if that makes sense.
• 29:55
  Speaker 1: It does.
• 29:55
  Speaker 1: I mean, my like tiny little brain is like wait what?
• 29:58
  Speaker 1: So?
• 30:00
  Speaker 1: Basically, like I would be able to go into it and be like hey, go into Jim and Jan Smith's emoney and update Jim's 401k to this dollar amount and it was like yeah, cool, Let me do that for you real quick.
• 30:14
  Speaker 2: Yeah, because it is possible to write code that does exactly that and then.
• 30:20
  Speaker 2: So you would essentially need to have a developer create that code and then equip the chatbot with it.
• 30:26
  Speaker 2: And yeah, that's totally possible.
• 30:28
  Speaker 1: Yeah, the whole time you're saying you would write the code.
• 30:30
  Speaker 1: I'm like no, no, no, you would write the code.
• 30:35
  Speaker 2: But yeah, and you do have to be careful with that, like, whenever we build systems where you can actually update information inside of the software, we always have the chatbot confirm with the person like I'm about to do this.
• 30:49
  Speaker 2: Can you please confirm, because you don't want it to just like, go wild and start updating values that are wrong or something.
• 30:55
  Speaker 1: So Right, right, because that's when it takes over and it's like you're not paying me enough, I'm going to go in and ruin all of your data, yeah.
• 31:02
  Speaker 2: I've never.
• 31:02
  Speaker 2: I've never seen it go rogue.
• 31:03
  Speaker 2: But I guess never say never.
• 31:06
  Speaker 1: Well, I'm four terminators in now, so I'm very warped on what.
• 31:11
  Speaker 1: I'm what I think this stuff's capable of.
• 31:12
  Speaker 1: Okay, so that's really amazing.
• 31:14
  Speaker 1: Like I keep thinking from like an operational perspective, like how AI can step in and just take so many of those little things that we can create workflows for.
• 31:25
  Speaker 1: But I really feel mostly like they're still done at a somewhat manual level.
• 31:31
  Speaker 1: So, like thinking through, like some of the stuff we're doing, like in our group coaching program that we're wrapping up right now, and I'm thinking like, wow, so this could really go in and do like analysis on our clients, so if I'm saying like, hey, go into Redtail and label all clients that look like this as a client, right, like anyone who has X number of dollars in AUM and or this and or this, like go in and categorize them this way for me and then set up a.
• 31:58
  Speaker 1: I want to see them every six months and then it'll remind me like hey, you have so-and-so, that's an, a client.
• 32:03
  Speaker 1: You're supposed to meet them in June and December.
• 32:06
  Speaker 1: Would you like me to draft the email for?
• 32:08
  Speaker 1: I mean, that kind of stuff to me is like unbelievably life-changing for advisors.
• 32:13
  Speaker 2: Yeah, absolutely, and everything you just described is possible.
• 32:16
  Speaker 2: It's pretty complex and it's not easy to build, but it's possible.
• 32:20
  Speaker 2: And that's where we are right now and I think there's going to be a lot of like software that comes out and starts making these things like a lot easier in the future.
• 32:30
  Speaker 2: When that will be, I'm not totally sure, but what you just described is like absolutely, in my opinion, the direction we're heading and these kind of things maybe in a few years will be kind of standard practice in my opinion.
• 32:44
  Speaker 1: Wow, are you seeing a lot of?
• 32:46
  Speaker 1: So you mentioned kind of earlier like chatbot or like I can get on Amazon and be like, hey, what's up, Where's my package and it can you know, go and look for that data and respond.
• 32:54
  Speaker 1: Are you seeing advisors using client facing chatbots in any way, shape or form?
• 33:01
  Speaker 2: So kind of and what I mean by that is something that I've built for a few advisors is like an email replying assistant.
• 33:09
  Speaker 2: So it's not actually a chat bot that the client is like physically looking at and seeing, but it's an AI model that will monitor their inbox and when a certain email comes in that meets certain criteria, it will actually draft a reply and put it like in the drafts folder or the outbox folder for the advisor so that they don't have to write it or they can at least start with the draft.
• 33:34
  Speaker 2: This is where we get kind of into the compliance conversation, like you don't want a chatbot or an email agent directly replying to a client without you seeing it.
• 33:46
  Speaker 2: You can't do that.
• 33:46
  Speaker 2: So that's why in a lot of cases, right, like we're putting these in like the drafts folder.
• 33:51
  Speaker 2: But yeah, so I haven't really seen many advisors that are having like a literal chatbot talking directly to their clients, but they do have them that are aiding them in replying themselves.
• 34:03
  Speaker 1: Okay, super, super interesting, so let's dive into that.
• 34:06
  Speaker 1: Yeah, let's talk.
• 34:07
  Speaker 1: That was the number one question that came up when I was asking advisors like hey, what do you want to know?
• 34:12
  Speaker 1: As I get ready for this, this interview, and everyone was like security, compliance, like even stuff like Finmate, ai, things that can come in and take meeting notes we're seeing like Otter being added to you know, zoom calls and taking it.
• 34:27
  Speaker 1: What do advisors need to be thoughtful about or thinking about when it comes to security and compliance and legal?
• 34:35
  Speaker 2: Yeah, so I think first and foremost, like what I just mentioned, is, you never want to be having AI directly communicate with your clients.
• 34:44
  Speaker 2: You can use it to aid in your own communication with your clients, but you always it to aid in your own communication with your clients, but you always need to be like that intermediary and I think that's like pretty self-explanatory and I think like a lot of people kind of recognize that.
• 34:57
  Speaker 2: Another thing to think about is like PII and like personal information, right?
• 35:05
  Speaker 2: So a lot of you know these softwares have information about clients, that's, you know, personal and sensitive information that you might not want to be feeding to an AI model.
• 35:16
  Speaker 2: So always be careful of that, especially like if you're using ChatGPT and maybe uploading documents to it.
• 35:22
  Speaker 2: You don't want to upload like a client's financial statement or some document that they provide you into chat GPT if it has, you know, like their social security number and things like that.
• 35:34
  Speaker 2: So definitely want to be like really careful and mindful when it comes to that.
• 35:39
  Speaker 1: With the custom stuff that you're doing is because I know you're using like open a well maybe I don't know that actually, but I like to think that I know what you're doing and I have.
• 35:51
  Speaker 1: I have no idea, but like so the custom stuff that you're doing is compliance.
• 35:55
  Speaker 1: That is that kind of a thing still a risk or a concern.
• 35:58
  Speaker 2: For sure, um, and that's something that we like that's actually like a big reason why you might want to use the custom stuff is because you can work within the boundaries of like those considerations.
• 36:09
  Speaker 2: So a lot of times, like when we're doing stuff, we just won't give it access to certain data points, so it's like literally impossible for it to access certain pieces of information.
• 36:18
  Speaker 2: And then there's some other fancy things that you can do if you really need to, where you kind of like strip certain information out and like kind of anonymize it and then send it to the model and then receive it back and then like map their real information back to it.
• 36:33
  Speaker 2: There's like some interesting research in that field right now, like how do you get around?
• 36:38
  Speaker 2: And you kind of like trick it, like you give it like fake names or whatever, if that makes sense.
• 36:42
  Speaker 2: But yeah, most of the time we're just building solutions that just flat out don't ever touch certain sensitive pieces of information.
• 36:51
  Speaker 2: And then there also is this idea of and this might get like a little too complex or technical, so cut me off from going down a rabbit hole- here.
• 37:02
  Speaker 1: Oh no, I am here for it, I'm going to take notes and Google it later.
• 37:06
  Speaker 2: So what I'm explaining right now is using like OpenAI as models, which is like the same model as ChatGPT, and when you do that, openai has the company itself, has the ability to see, and they can save the information that you're sending and receiving from it.
• 37:23
  Speaker 2: You also can host with like or sorry, it's called hosting your own model, which is where, essentially, you have control over it and no company can access it or see what you're doing with it.
• 37:36
  Speaker 2: That's much more complex and you can't use OpenAI's models to do that, but Facebook has developed some models, as an example, that are very similar to ChatGPT and you can actually host them yourselves, meaning they completely live within your technology and Facebook will never see what is information is being passed in or out of them.
• 37:58
  Speaker 2: That's a pretty complex endeavor.
• 38:01
  Speaker 2: It's a very expensive endeavor, but it is another option and I think we'll see the cost of that probably lower over time and that might become more, more normal.
• 38:10
  Speaker 1: Yeah, Okay, so fascinating.
• 38:12
  Speaker 1: So we kind of touched on it before and I want to come back to that this idea of one of the biggest pain points for advisors being meeting notes.
• 38:20
  Speaker 1: So let's just kind of take out compliance or security and everyone will have to, you know, check and think about, you know deal with that on their own.
• 38:27
  Speaker 1: What are you seeing as some of the best resources, tools or what are you building for advisors that help them not just like capture a transcript, let's say, for a 60 minute meeting, but to actually like synthesize that data down into follow-up points, tasks, Like so you're talking about what you can build, can go and then take those and actually turn them into a task inside of the CRM for people, right, yeah?
• 38:52
  Speaker 2: yeah.
• 38:53
  Speaker 1: So what do you see?
• 38:54
  Speaker 1: Like, talk us through, like, what are you guys seeing?
• 38:55
  Speaker 2: as for best practices from a meeting notes perspective, yeah, so meeting notes is a huge one, and typically it kind of depends on, like, the context of the meeting, but usually what we'll do is we'll kind of bucket meetings into various categories and then have the system do certain things with the meeting notes based on, like, what type of meeting it was, and a lot of those are exactly what you mentioned, like generating follow-up tasks and adding them to the CRM, drafting reply emails to the client and things of that nature.
• 39:27
  Speaker 2: There are some cool technologies that are focused on this exact use case and I'm sure maybe you or some of your listeners have heard of them.
• 39:37
  Speaker 2: One of them is called, like Zox, and that's exactly what they do, and another one is called Jump, like Jump app, and they are meeting note software specifically for financial advisors, and they do exactly that.
• 39:56
  Speaker 1: Okay, cool, and so again.
• 39:58
  Speaker 1: So, like my brain like looks at, like what's the highest, like okay, it can read through it, it can synthesize stuff, it can draft a follow-up email, it can create tasks, it can establish its own reminders inside of the CRM.
• 40:11
  Speaker 1: Hey, okay, this is what I want to go in a follow-up email, this is what I want to be created a task.
• 40:25
  Speaker 1: This is a reminder that I want for you, as my you know lovely assistant, to tell me in three months or six months, like how do we actually do that?
• 40:36
  Speaker 1: Yeah, so um he's laughing at me.
• 40:39
  Speaker 1: You guys can't see him.
• 40:39
  Speaker 1: He's laughing at me.
• 40:41
  Speaker 1: That's a dumb question.
• 40:42
  Speaker 2: I'm smiling because I love this stuff, I love talking about it.
• 40:45
  Speaker 2: Um, so it's basically like it's really important to have like a human in the loop and, like, going back to what I said, you always want to have like the, the assistant, like asking you, hey, here's the tasks I'm about to create in Redtail, like does this look good?
• 41:00
  Speaker 2: Should I actually go and do this?
• 41:01
  Speaker 2: But I guess, to answer your question kind of like under the hood, something that's really popular right now is this concept of like a multi-agent architecture and this might get like a little technical again, but I'll try to keep it simple where what people are doing and like what research is pointing towards seeing like the best results in, is building little agents that do specific little tasks and then teaching them how to work with each other.
• 41:28
  Speaker 2: So this is kind of crazy and really interesting to me.
• 41:31
  Speaker 2: But essentially what you do is you would take like an AI agent and say, like you are like the red tail expert, you teach you how to do everything in red tail update information, add new tasks, read information, et cetera and then you would have another one that you basically train to be like an email responder and it's like your job is to draft emails.
• 41:52
  Speaker 2: You're super good at that.
• 41:53
  Speaker 2: Here's all the information you need and then you build a piece of software that actually kicks off a workflow where those two kind of work with each other and they go back and forth and hand off tasks between each other.
• 42:04
  Speaker 2: So, like in this complex use case where you like want to get the transcript of a meeting, write out the follow-up tasks, then add them to the CRM, then write up the follow-up email, then add the email to the drafts folder.
• 42:21
  Speaker 2: It's probably going to be a lot of little agents that do these specialized things, kind of collaborating with each other.
• 42:28
  Speaker 1: Okay, that actually makes way more sense to my brain.
• 42:30
  Speaker 1: Yeah, it's not just yeah.
• 42:33
  Speaker 2: And it's, I think, over time, like this is the direction we're going in, where, like these, little agents will be spun up dynamically.
• 42:43
  Speaker 2: So when you say you want something to be done, one agent will just create all the other agents needed, but for right now you still, from a software standpoint, have to create them yourselves and explicitly show them how to work with each other.
• 42:57
  Speaker 2: But that's definitely, in my opinion, the direction we're heading in.
• 43:02
  Speaker 1: Okay, amazing.
• 43:03
  Speaker 1: So, all right, how okay?
• 43:06
  Speaker 1: So I'm going to share a quick story.
• 43:07
  Speaker 1: So my whole family is in television.
• 43:10
  Speaker 1: I actually you might even be able to see behind me there's some Emmys sitting on my shelf from family members not my Emmys as much as I think this is an award winning podcast, right, we don't have any Emmys, but um, so my whole family was in TV and you know I can remember my dad back in.
• 43:28
  Speaker 1: Like you know, the movie Anchorman is like exactly what my dad experienced back in the seventies in the TV world and I can remember as technology came about and then my brother became a director as well, and I can remember my dad telling my brother, ben, when new technologies came out.
• 43:42
  Speaker 1: He was like don't get worried about it, don't resist it, be the best at using it and then your job will always be safe.
• 43:49
  Speaker 1: So if you can be on the forefront of those technologies, if you can embrace them and get really, really good at them and be the best at using them, so I mean I can remember when TV cameras right For directing the news, like when they became automated, and then when editing software became more automated.
• 44:06
  Speaker 1: And now you know my brother does the job and it only takes one man versus a team of 12, because he was always being really intentional about staying on the forefront.
• 44:15
  Speaker 1: So what can advisors do to better equip themselves, because this stuff is coming, and it's coming in a way that will make our lives easier, even though it feels like daunting or like to those of us that aren't you like?
• 44:30
  Speaker 1: This is like how do I teach it?
• 44:31
  Speaker 1: What do you mean I teach it to do that?
• 44:33
  Speaker 1: What do you mean?
• 44:33
  Speaker 1: I have an agent creates other agents.
• 44:34
  Speaker 1: How do we get well equipped, or how do we at least become as equipped as possible, to start understanding this and then intentionally integrating it into our businesses?
• 44:47
  Speaker 2: Yeah, absolutely.
• 44:48
  Speaker 2: I'm happy you brought this up because I definitely have some opinions on that too.
• 44:52
  Speaker 2: And before I kind of get into like the tactical stuff you kind of just reminded me of like one of my favorite stories of thinking about this like technological advancements and things like that is the story of the ATM.
• 45:03
  Speaker 2: So, like when the ATM was first invented, everyone thought it was going to eliminate bank tellers because it like quite literally, is a machine that's a bank teller.
• 45:13
  Speaker 2: And what ended up happening was it actually significantly lowered the cost of operating a bank and people still had that desire to work with a human in some circumstances.
• 45:25
  Speaker 2: So a lot more banks opened and it ended up creating like a net demand for bank tellers.
• 45:30
  Speaker 2: So there was actually more bank teller jobs after the ATM was invented than before.
• 45:35
  Speaker 2: So for anyone who's like worried or concerned, I would encourage them to kind of like think in that regard, and I think it's especially true in this industry as well.
• 45:43
  Speaker 2: Right, like we all know, this is a relationship based business.
• 45:46
  Speaker 2: Clients hire advisors because of who they are, because they, they know them, they like them and they trust them.
• 45:52
  Speaker 2: So I think advisors, more so than a lot of other professions out there are really kind of protected from this technology, like coming for their jobs, I guess, if you will.
• 46:02
  Speaker 2: With that being said, though, I think your dad is spot on.
• 46:07
  Speaker 2: That's what a lot of people are saying about AI right now.
• 46:09
  Speaker 2: They're like it's not going to replace you, but someone who uses it to be twice as efficient as you will replace you.
• 46:15
  Speaker 2: So I would encourage anyone who's thinking about it might be scared of it to not be to fully embrace it and start seeing how it can make your life as an advisor easier.
• 46:27
  Speaker 2: It can automate some of those boring things that you probably don't like sitting down and doing in front of your computer, so that you can actually spend more face to face time with your clients.
• 46:37
  Speaker 2: With your clients and in terms of, like what to actually do, I think it just comes back to staying informed, following the latest trends.
• 46:45
  Speaker 2: You know reading articles, use, chat, tpt every day, thinking to yourself you know how can I use this on a daily basis, learning more about it and, if you want, you know, finding perhaps a strategic partner or a consultant.
• 46:59
  Speaker 1: You know, like us to kind of guide you along, and yeah, oh yeah, nobody should attempt to do what you're doing on their own.
• 47:08
  Speaker 1: That's one of those like absolutely not hire, absolutely not hire people.
• 47:11
  Speaker 1: You know, and I want to piggyback on what you said too, because I do hear that from advisors like, oh well, what are we going to do when, like you know, joe Schmo can get on chat GPT and it can write an entire financial plan?
• 47:22
  Speaker 1: I think it's important and tell me if I'm wrong, but it's taking all of this data and while it's, you know, doing like that generative AI, it can't create original thought, correct, or can it?
• 47:37
  Speaker 2: So that's kind of a tough question to answer.
• 47:40
  Speaker 2: So I guess we're going to get philosophical right now.
• 47:44
  Speaker 2: It kind of depends, like how you define original, so generative AI.
• 47:49
  Speaker 2: It technically is creating its own thought or its own response when you're talking to it, but really all that is is like a sentence, for example, that was formed based on everything it's trained on and what it's trained on.
• 48:05
  Speaker 2: A lot of these, like chat gbt, for example, is trained on essentially almost all the information on the internet, so it's kind of like a, a replication of everything that exists on the internet.
• 48:16
  Speaker 2: So it's like is that an original thought?
• 48:18
  Speaker 2: Maybe, I don't know.
• 48:20
  Speaker 2: It is original in the sense that the engineers who created ChatGPT didn't say answer X when you see question Y, but the way it's answering your questions is based on data and information that was created by humans, like essentially the entire internet.
• 48:47
  Speaker 1: Yeah, so it can't create like a new concept or theory or something kind of all on its own if it doesn't already exist out there in the internet.
• 48:52
  Speaker 2: No, it can.
• 48:52
  Speaker 2: And this is actually something that's called like hallucinating hallucinations, where it actually the what chatbots and chatGPT wants to do is it wants to make you really happy, like it wants to give you the best possible answer, and sometimes it'll say things that might not necessarily be true and it'll kind of make them sound true because it wants to be right and it wants to give you the right answer.
• 49:17
  Speaker 2: This is called like hallucinations is.
• 49:19
  Speaker 2: There's a whole other topic.
• 49:20
  Speaker 2: There's ways to really like mitigate that.
• 49:22
  Speaker 2: Obviously, when you put something into production that you're using to help you, you want to mitigate how often that happens.
• 49:30
  Speaker 2: But there's been some really funny examples of this recently.
• 49:33
  Speaker 2: Actually, if you look up like Google, they recently tried to integrate a chatbot into Google search and there's like I think they like shipped it too early.
• 49:44
  Speaker 2: They didn't have enough people like review it.
• 49:46
  Speaker 2: So there's been like all these funny things on the Internet of like things that it's been saying that are wrong.
• 49:51
  Speaker 2: Like.
• 49:52
  Speaker 2: An example is like someone Googled I can't get the cheese to stick to my pizza, what should I do?
• 49:58
  Speaker 2: And the google search literally said you should glue it.
• 50:01
  Speaker 2: Uh, because someone wrote that on like a reddit forum as a joke and it became a really popular reddit post.
• 50:08
  Speaker 2: So their model I'm totally expecting that their model must have some sort of weighting of like oh, if it's a popular reddit post, show it to the person who's searching.
• 50:17
  Speaker 2: Um, so if anyone wants a good laugh, just look up what the Google search has been doing recently.
• 50:23
  Speaker 1: Okay, that's hysterical, I love it.
• 50:25
  Speaker 1: Okay, we're getting into hallucinations, which feels very Skynet, very futuristic to me.
• 50:30
  Speaker 1: So with that we'll wrap up.
• 50:32
  Speaker 1: Is there anything else that you think it's important for advisors to know or to be considering or, I guess, maybe a better question, before they show up to a you right?
• 50:43
  Speaker 1: What for you in order for you to do your best work, to build something custom for advisors?
• 50:49
  Speaker 1: What is it important that they know about themselves, their practice, their pain points?
• 50:53
  Speaker 1: What can they come to you with so that it's the absolute best input to get the best output?
• 50:59
  Speaker 2: Yeah, so the best input to get the best output.
• 51:00
  Speaker 2: Yeah, so the best input to get the best output.
• 51:02
  Speaker 2: I feel like I'm a chat bot.
• 51:03
  Speaker 2: I'm just kidding.
• 51:05
  Speaker 2: So, it really like.
• 51:08
  Speaker 2: This sounds like really simple, but it really comes down to just like having a clear set of goals and like where your pain points and bottlenecks currently like.
• 51:15
  Speaker 2: A lot of times, advisors will come to us and be like OK, give me the, the playbook for AI, just tell me everything that you need to build, and I'm like okay, well, what are your goals?
• 51:26
  Speaker 2: Who's on your team?
• 51:26
  Speaker 2: What are your pain points, where are your bottlenecks?
• 51:29
  Speaker 2: Because you can really do almost anything with AI right now and it's very exploratory in nature, so there is no perfect solution or set of solutions for someone like us to build.
• 51:41
  Speaker 2: So, yeah, I would say you know getting being real about like what are your goals, what are your pain points?
• 51:47
  Speaker 2: And then also just like a list of all the different technologies you use.
• 51:52
  Speaker 2: That's pretty straightforward, but that is sometimes a little bit difficult to get out of people that is sometimes a little bit difficult to get out of people Awesome, okay.
• 52:08
  Speaker 1: So at the end of every episode, I like to ask our guests that are business owners themselves, that are entrepreneurs what are two practices and I'm going to laugh if they're AI related, they don't have to be but what are two practices or things or tools or mindset shifts, hacks, whatever you want to call them that you personally use, that you've adopted, to help you run your business more efficiently, whether that's from a software standpoint, or like a productivity or, you know, just like a personal time management?
• 52:33
  Speaker 1: What do you?
• 52:34
  Speaker 2: use that kind of has helped you, I'll give you two One is technology and one is completely not technology.
• 52:41
  Speaker 2: So, I obviously.
• 52:42
  Speaker 2: I use AI a lot, so I use the, the chat GPT mobile app and it has a voice feature where you can talk to it.
• 52:50
  Speaker 2: I use that all day long.
• 52:52
  Speaker 2: So I give a question pops in my head.
• 52:54
  Speaker 2: I just pull up the app and I speak into the app and I ask it that question.
• 52:58
  Speaker 2: I haven't Googled anything in a while because I just do that.
• 53:02
  Speaker 2: So that's like a little hack.
• 53:04
  Speaker 2: I'm doing that like probably I don't know 10 to 20 times a day and then, completely not tech related, this is super random, but I recently got into cold plunging and I absolutely love it.
• 53:16
  Speaker 2: I do it every morning.
• 53:17
  Speaker 2: There's so many good benefits.
• 53:21
  Speaker 2: It's allowed me to kind of wean off caffeine and coffee, it wakes me up in the morning, it helps me become focused and I absolutely love it.
• 53:30
  Speaker 2: So if anyone is thinking about doing that, I would encourage you to do it.
• 53:34
  Speaker 2: I'm a total nerd about the science behind it.
• 53:36
  Speaker 2: Okay, I love it.
• 53:42
  Speaker 2: So do you have a cold plunge in your apartment?
• 53:43
  Speaker 2: I have a cold plunge in my apartment, on my balcony, and I'm on the 35th floor of a high rise in downtown Chicago.
• 53:48
  Speaker 2: So believe me when I say it's not easy to keep that water cold.
• 53:51
  Speaker 2: I'm lugging ice and buckets around my apartment all the time.
• 53:56
  Speaker 2: My girlfriend thinks I'm insane, but I do what I got to do to get my my daily cold plunging.
• 54:02
  Speaker 1: Worth it Okay.
• 54:02
  Speaker 1: So how cold and for how long?
• 54:05
  Speaker 2: So the like the research says like the optimal amount of time is between three and five minutes Um.
• 54:12
  Speaker 2: I like to try to like challenge myself to stay in as long as I can, which is probably not good.
• 54:16
  Speaker 2: So usually anywhere between like three and five minutes um.
• 54:19
  Speaker 2: Temperature I don't really know what the temperature is, I'm not that fancy, I don't have a thermometer out there, but it's ice and water, so I would guess it's probably between like 35 to like 45 degrees maybe.
• 54:31
  Speaker 1: Okay, fabulous.
• 54:31
  Speaker 1: I used to do cryotherapy and I loved it yeah it's the same thing, from what I understand.
• 54:38
  Speaker 2: I've never done cryo, but from what I understand it's the same thing people do for the same reason.
• 54:43
  Speaker 1: Yeah, Just if you ever do it close your eyes so that your eyeballs don't freeze.
• 54:46
  Speaker 2: Noted.
• 54:46
  Speaker 1: Awesome.
• 54:47
  Speaker 1: Well, thank you so so much for your time.
• 54:51
  Speaker 1: If anyone listening which everyone will be like, yes, I need these things.
• 54:56
  Speaker 1: How do they find you?
• 54:57
  Speaker 1: What's the best way to find what you're doing and reach out?
• 55:01
  Speaker 2: Yeah, so you could check out our website.
• 55:03
  Speaker 2: It's beemo B-E-E-M-O automationcom.
• 55:08
  Speaker 2: My initials are BMO, by the way, so it's like a little nickname for me, bmo.
• 55:12
  Speaker 2: You can also email me, ben at beemoautomationcom, or connect with me on LinkedIn.
• 55:18
  Speaker 2: I love LinkedIn.
• 55:19
  Speaker 2: I'm on there all the time Just Ben Olson.
• 55:22
  Speaker 1: So okay, perfect, I will link all of those in the show notes.
• 55:24
  Speaker 1: Thank you so much again for your time.
• 55:26
  Speaker 1: Appreciate what you're doing.
• 55:27
  Speaker 1: We're so glad there are people like you out there that know these things to to help us.
• 55:32
  Speaker 1: Most advisors are a hot mess behind the scenes, so if anyone listening is like, yeah, that's me, you're not alone, we all are, and these little things that you're doing will help significantly.
• 55:43
  Speaker 1: You know, sometimes it seems like a small task, being automated is not doesn't have a huge impact, but when you can do that time and time again, it does ultimately end up really helping you to operate more efficiently and reduce that stress and overwhelm.
• 55:58
  Speaker 1: So thank you, ben, appreciate your time.
• 56:00
  Speaker 2: Yeah, absolutely, thanks for having me.
• 56:02
  Speaker 2: We'll have to do it again sometime maybe.
• 56:05
  Speaker 1: Oh yeah, because it's all going to change in a year, so we'll be back.
• 56:07
  Speaker 2: Yeah, that's right.
• 56:10
  Speaker 1: Okay, I hope you loved this interview as much as I did.
• 56:14
  Speaker 1: I just think this AI stuff is so interesting and I loved just some of the language that he that he gave us that we can totally utilize, like things like, um, these AI agents that go into your software and do things for you, like input stuff into your CRM.
• 56:31
  Speaker 1: Like, my mind is blown at stuff like that.
• 56:34
  Speaker 1: Um, the way you can use chat GPT to create customized documents and you can upload stuff and train it and teach it Super cool.
• 56:42
  Speaker 1: I love a lot of the tools that he mentioned.
• 56:44
  Speaker 1: I'm going to drop all of those in the show notes below and really kind of I was appreciative of him helping us understand what generative, generative AI is and how it can actually um, like what.
• 56:56
  Speaker 1: Like what's the difference between between all the different types of AI and really what can you have it start doing and what's already available to us on the market?
• 57:04
  Speaker 1: And then, how is he creating all of this cool stuff?
• 57:07
  Speaker 1: I mean, I think I said it in the episode but like I didn't even know what it was capable of, I thought it was chat GPT for marketing right or bust, and I've been slowly learning and utilizing it in my own business, but watching advisors using it to create investment policy statements, to write emails, to build out templates, I mean there's so much cool stuff it can do.
• 57:28
  Speaker 1: So I'm excited that Ben is part of the group coaching program now and that he's sharing with us all these different ways that we like dum-dums not that you're a dum-dum, I'm pointing at myself as I'm recording this but for people who are like, yeah, I'm not that super tech savvy, how can I use these AI tools, how can you take away some of the intimidation factor and how can we make it really accessible for me in my business?
• 57:55
  Speaker 1: I thought it was really interesting that he's like I don't even use Google anymore, I use ChatGPT.
• 57:59
  Speaker 1: That is just fascinating to me.
• 58:02
  Speaker 1: So, anyway, hope you found this episode helpful.
• 58:04
  Speaker 1: Per the usual, I would love for you to leave a rating for this podcast.
• 58:10
  Speaker 1: That would mean super duper much to me, and I hang out on LinkedIn and over the Efficient Advisor community on Facebook, as usual, guys.
• 58:19
  Speaker 1: So thank you so much Again.
• 58:20
  Speaker 1: Don't forget if you are looking to join the Systems to Scale group coaching program, the registration for that will start in July.
• 58:27
  Speaker 1: Unless you're well later July, unless you're on the wait list.
• 58:30
  Speaker 1: Then you get in early and you get a discount.
• 58:32
  Speaker 1: So if you want to be part of that wait list, don't forget to click the link in the show notes or head out to the efficientadvisorcom and you'll see a banner up at the top taking you to the wait list.
• 58:42
  Speaker 1: All right, thanks so much.
• 58:44
  Speaker 1: Talk soon.
• 58:45
  Speaker 1: Have a great week.