AI Guides with John Stroud | E109

welcome to the financial planning for Canadian business owners podcast you will hear about industry insights with award-winning financial planner and entrepreneur Jason Pereira through the interviews with different experts with their stories and advice you will learn how you can navigate the challenges of being an entrepreneur plan for success and make the most of your business and life and now your host Jason Pereira hello and welcome today I'm John Stroud CEO at AI guides AI guides is an AI strategy and consulting company and I brought John on the show specifically to

talk about what everybody else is talking about which is artificial intelligence specifically around generative Ai and how artificial intelligence can be harnessed to basically make your business better and no we're not talking about having to work at a large-scale Enterprise in order to make that happen we're talking about even single person operations there is a buffet of options these days and with that here's my interview with John John thanks for your time today oh thanks for having me show John Stroud of AI guides tell us about AI guides sure so we are AI strategy consultants and we got started a couple of years ago because my partner Janan and I could see

that for all the potential of AI many organizations were struggling with it and the failure rate could be quite high they were failing because they were getting mad touched to the wrong solution and so what we wanted to do was get in early in their planning process to give them independent advice on what was the best fit for their particular organization and just get them on the road map to a successful implementation excellent so quite the Hot Topic right now I mean I I look at the entire open you know the chat EPT Revolution quarter Revolution that when that emerged on the scene it was one of those seminal moments in technological history yeah let's go back a bit so I look back at when chat GPT came on the scene and caught the attention of everyone very quickly as one of those seminal moments in the forward progress of humanity quite honestly it was it was this thing that basically as far as as technology is can be intimidating in our official intimidating the simple nature of the interface and the ability to do from the silliest to the most incredible things with it very quickly captured everyone's imaginations in incredibly fast so I can't remember the last time we had a technology that did that maybe the iPhone maybe not yeah it's a big one it's a big seminal moment in history at least the way I look at it so so that happened and that of course brought AI back to the Hot Topic on technology for everybody and let's talk about first of all what we're talking about when we're talking about artificial intelligence because that is an Ever moving goal post quite honestly so when we're talking about AI what does that mean to you I think of AI in non-technical terms it's a prediction machine so you give it data and it ingests it but puts it through an algorithm and it makes a prediction about the world and in order to get that prediction machine you can think of it as as having like three main parts it's kind of like an analogy for a car you had your data which is like the oil but then you need a pipeline that refines it so that's cleaning up your data and then you need an algorithm which is just the engine of your car so you take your oil you refine it and then you put it in your engine and your car can go so if you've got data that gets cleaned and you put it in your algorithm then you can get it to make predictions about just about anything yeah it's funny because I was you know I've had guests on my other podcasts about data being the new talking about data being the new oil and yes artificial intelligence is the engine that goes over top of that and I guess that represented for most people the biggest obstacle to implementation of artificial intelligence stores out there I mean don't get me wrong it was technological limitations for the layperson to even being understand but data stewardship up until the last like five to ten years has been relatively poor in most organizations quite honestly like people would just the amount of human error in terms of input of this stuff and there was no actual like policing of that properly and even old Legacy systems just suddenly deciding that a certain column was or certain field is going to be used for something new after a certain point like just it was garbage absolute garbage so the ability to leverage that a large organizations was limited by their just terrible stewardship for decades on end that's absolutely right and I talked about you know the high failure rate of projects a big reason for that was around their data stewardship is it's a if you don't take the time to get the data right then you have a garbage in garbage out problem and so many organizations might have been better off not even starting an AI project when you think of the opportunity cost of going in the wrong direction and then get something at the end that doesn't work not only did you spend all that money but you wasted all your time and you dispirited the team in the process right so it it's hard to do well when it works it's great but it does require effort well it's it's amazing because I mean mean typically these people get showroomed something that does work and it does work if all of the infrastructure beneath it is set up for the tower to not fall over the problem is they sign up for it maybe overestimate their ability to get that infrastructure in place or to solve their infrastructure problem and then they put the tower up and it falls down right like Bingo and another example that we use is they get sold a light bulb and they say wouldn't it be great if you had a light bulb instead of candle and the light bulb really works but you can buy it but if you don't have an outlet to plug it into and electricity going to it the problem's not the light bulb the problem is everything on the back end exactly as for the reticence I mean there's been more than one massive blow up of Executives getting fired over money getting blown on something that couldn't be implemented so I mean as usual basically Executives don't want to stick their neck out for something that could basically blow open their face right so that was the big obstacle limitation to large scale prizes and and small scale companies look they're not going to sit back and think about their data you know data stewardship they're going to hire they're going to buy off the shelf software they're going to implement that and there's not many who have the wherewithal to basically look at layering other systems over top of those things and integrating and whatnot so that's that's always been a difficulty but with the current world of AI how has that changed there's a big change with generative Ai and the threshold for anyone to get involved with it is just so much lower and there's a couple of reasons for them one you talked about the excitement that people had of using it for the first time and it it wasn't so much the underlying technology that had changed all that much it was just the interface it was just a chat interface so it was completely intuitive to people so it was easy to access and then the second thing is is once you actually got in there there were these pre-trained models they done all of that hard work that we were talking about getting the data getting it organized getting it to you so you had this little Christmas present for you of something nicely prepared and so the big companies like open AI have presented it to you it's almost free practice and ready to go right it's pretty exactly yeah so that opens up any number of doors for implementation so what have you seen as kind of like the simple base level uses of artificial intelligence let's also take a step back and set set expectations here right when people think of artificial intelligence I'm sure the first thing that comes to mind is The Terminator or some other character or previous generation would have been Hal you know whatever whatever it is right and the reality is we're talking about theirs we're talking about general purpose AI like that's the white whale and the the the the long-term goal of the industry is to basically get to a point where you can just have a natural conversation with an AI and that AI understands you fully like a human being would and be able to undertake any number of things without you having to think about it basically what Syrian and Alexa hope to be but are very very limited in their capacities so that's general purpose AI we're not talking about that with anything we're talking about today we're talking about very narrow use cases for AI so very specific hyper usage it's a lot easier to nail one thing really well than it is to try to beat all things to all people so basically that's the setting the stage so let's talk about the easy simple implementations that you've seen that are powerful that basically in let's let's face this face it no one thing is going to run your business for you right like no one thing's gonna do that but combining a bunch of stuff will have the impact of creating efficiencies that will add up to significant amounts of time as you implement more and more so using that kind of framing give me some examples of some simple implementations that can be time saving productivity improving and basically that can compound over time for small businesses sure so the easiest way to get in is just simply tapping into open Ai and the chat gbt so you can buy you can get it for free or you can buy a subscription that gets you to the head of the line and you can do a number of creative and productivity tasks you can make it your co-author or your co-pilot on writing emails or doing marketing and so forth and the models are not just open AI but there are going to be new ones that are emerging for specific purposes for example Bloomberg has come up with its own gbt now that's focused on financial services and that's going to be at a much higher price point but you can see that other General models are going to evolve that are going to be similar to what uh Bloomberg has so for your domain so that's doing no work then the next step over is there's companies that have essentially built a niche offering mostly on top of the open AI so if you're a financial planner there are Niche applications on how you can use Excel better or there are Niche applications on how you can provide better customer support so it's got all of the strength of the open AI model but with a little tweaking and that's still very easy to do it's a small subscription fee and then if you want to go further it's actually taking your own data and adding it to the open AI model and fine-tuning it so that it's giving answers that are specific to your context the uh fair enough and that's that's you kind of went top down to some degree so and then back out but a couple things to touch upon there all right let's talk about some of the simple general ones right the one I like to use as an example because you can you can literally get access to chat gbt so the powered stuff in Google Docs in Microsoft Office now right like it can literally start writing your blog posts your letters whatever else it is give you ideas so it's a matter of learning how to use the proper prompts which we'll come back to later in order to do that but something even more simple the simplest example I use is grammarly go which I've been using grammarly for years that basically cleans up my my spelling and my my typos which I'm not notorious for typos but now grammarly go well actually I can tell it what kind of email to write right so some of my email responses if if I don't want to spend a lot of time drafting something overly wordy that simply you know rejects a prospect or to politely invite someone to something I can simply type in invite this person do blah blah blah and it comes up with a fully composed email right and that is thoughtful and well done and whatever tone I want it to be when you think about the so let's just think about how much time you spend an email right like depending on your job if I can shave one to two minutes off a link to email every time add that up over the course of a year that's a lot of days saved quite honestly so simple implementation right and then when it's not intimidating to anyone yeah absolutely and that's just one example and they're now websites so one of them is called there's an AI for that and they're up over 5 000 different applications that are just like that grammarly is one of the big ones uh Jasper's another one and so there's some some big players and then tons and tons of little niche things so almost like any part of your job you can get better and faster at it so there's been a switch from just being more creative to actually being more productive with these tools totally and let's go up to the we talked about financial you know Services specific so for those of you who are in the financial planning industry or financial business you've already seen a bunch of stuff like this now not generative AI more algorithm based but Conquest planning which basically will actually tell you what strategies to basically use and also potentially even just run a plan for you to a company I started working with recently called CFA to the U.S sorry out of the UK that is actually its own chat large language model protocol that basically is looking to automate back in middle office information the Gathering and and workflows so being able to throw in questionnaires and whatnot and be able to and linking into crms and being able to use it as an interface is to say okay I mean with a client what's their outstanding balance on their mortgage boom pulls up that information what is this boom what is that so being able to draw in all that information to one unit 5u while also simultaneously being able to pull extract data when necessary so yeah I think that we're going to see it's interesting because what typically happens with technology is the broad stuff is handled for the General market and then people slowly get into like building Niche Services I feel like this is all happening at the same time quite honestly like the broad Market solution came out and now suddenly everybody just immediately basically doing every level of depth and service that there is yeah I think that's totally right and for every domain more and more of these tools are emerging and so you uh you got to check it out to make sure that if it's a good fit for you and some are more reliable than the others but the problem is not going to be that these aren't these tools aren't there it's going to be how do I pick of the dozens or hundreds of tools that could work what's the best fit for me well I also think that just part of this that has to yeah so let's talk about it doesn't actually deal with a couple things let's talk about the journeys so like someone wants to get started potentially looking to harness artificial intelligence maybe they're detective file whatever else is how do they get started down this journey of figuring out what's out there how can they use it what areas of their business it could help in like what does it look like sure I think a really easy way to do this I mean you start with some literacy about just like what AI is how could it impact my business and if you want to have a conversation about that chatbt is actually not a bad place to start because it has that chat format and it can point you to resources but once you've done a little bit of work you can experiment some of these things are free a lot of them only cost a little bit a month to try them out if you're gonna get going on it what I would try to do is think about something that's internal facing as opposed to external facing and something that has a human in the loop as opposed to something that the AI is going to do on its own think of it like a science experiment where are the opportunities for me and I want to manage the downside so that if something goes wrong and I get a suggested email back to a customer that I don't like I get a chance to review it before it goes out so I think that's good and then if you really start to want to expand then I mean you can start looking at bringing in outside help to to give you some advice on that so a good place to start and I would also say that frankly like the reality is is that they're probably having it thrown at them by their current vendors right like a lot of artificial almost every major platform I see is either basically already incorporated or working to incorporate it shortly so I mean everything from Salesforce who's already got to be a test going on on their large language model and one of my favorites notion who was very very picked out of date on this where they can basically there's a lot of things they can do with notion AI it just keeps on expanding but the but I think the key to understand is like you said is the awareness and this can be even like the simplest of businesses can find ways to harness that and one of the examples I use for this is you know I found a I think was called PDF or chat PDF or something like that and it's a system that you just upload any pef into it and start asking questions right so you can upload a contract and ask like what it means for you in certain contexts and actually get information that's relatively true so I think that is that is the key so one of the things I want to talk about though is and this comes up from time to times concerns about privacy because basically open AI is effectively a open protocol right so technically it is a open source project so this has come up more than once it's like hey wait a sec am I pushing confidential information to a open source project so what is the answer to that and let's talk about the implications yeah don't share personal information or sensitive information with open AI because the controls that you would want in place are not there and so if you share that information you don't know where it might Bubble Up Against again and so you want to be very cautious about the the information you share until you can get to the point where openai is going to allow for more private containers where you make sure that your information will be protected sort of like people used to be nervous about putting information in the cloud and they now feel comfortable you'll want to wait until they get to those level of protections before you you start sharing your information yeah so I think I think you're absolutely right there but I also think the other piece of this too to keep in mind is that that's the open AI protocol now there's also large language models that are private within companies and different vendors right so one of the big concerns I have with the company I was advising was okay hold on you're actually using the open protocols and pushing all the supply information to them and the answer was no we are sidling that we've got our own large language model so I think it's important to understand that there is a there's a difference that there's not everybody's not everybody's just using open AI they I think it's a very important question that when you're doing due diligence they understand what the Privacy boundaries are of this thing yeah exactly you're right and so different players have different rules and if it's free then you should really be worried about it but if you're setting up specific agreement with the company then you'll want to check that in advance right so that's the Privacy aspect we discussed by finding it let's talk about basically like your company and your role like you talked about the Consulting like give me kind of the start to finish experience of what it is to deal with the consultant around AI like how does that look what kind of companies do you typically deal with you know take me through that entire process sure so we typically deal with small and medium-sized organizations the leaders of those tend not to be tech savvy and instead they're more optimistic about the potential for AI to be transformative their business and they're also a little bit worried about what all of this change means and so they have a sense of where they'd like to go with their business but they know they know they don't know enough about Ai and so what we do is we sit down with them and take them through three steps so one is their Vision it's what is the goal that they're trying to accomplish and then we frame that in an in AI terms for them and then secondly we look at their particular text Act and the skills of the the employees that they have so the more sophisticated they are the more options that are available to them so the third step is we say well if that's the vision of where you want to go and these are your I.T constraints points then here's a road map on how you can accomplish that goal and then we hand them off to the team that can do the implementation and what's particularly nice if you if you do the work now is that there's a government of Canada grant for creating a digital strategy so for companies that have 500 000 in Revenue they probably qualify for this Grant and that gives them fifteen thousand dollars towards the preparation of the strategy so if you got a hunch about how you might be able to use AI then the vast majority of the cost is covered through a grant and so you get that independent assessment of what you could do and then if you want to implement that then there's a further up to a hundred thousand dollar interest-free loan to implement your plan excellent well and it's a smart program honestly I get people's feet wagged in the grant let them get the assessment done for free and then give them the ability to basically finance that you know smart move very smart so okay so now we covered that let's go back to kind of like the foundational steps we talked about the the basically how to get started the what are the things that people need to be aware of that maybe they're not thinking about or we haven't talked about it just yet because we talked about like where to find this stuff how to get implemented how to get started how to think about a little bit and and what privacy concerns you have what else should they be aware of before they go down this journey one thing to keep in mind is that sometimes people think of AI as being just like an I.T project and we like to say AI is not it AI is more like a science experiment think of it as you know you're teaching the machine it's like a child that needs to learn and you ask it some questions and initially get some wrong answers and with training you can make it better so you have to have that mindset that there's incredible potential here but it's not necessarily going to work right out of the box for some of the simpler Niche applications for sure it's more like software but the the more you get involved with AI the more to have this experimental mindset and part of the the ROI from a project is not just what happens on this particular project that you're working on but it's the learning that you get it's to thinking about projects differently so if you can do a small project with AI then you can think oh now I get how this is different from it and then applying those lessons in something more important more valuable to the organization yep agreed and let's talk about that training because I mean here's the thing people have this this every every now there's a funny story about this first off like yes anyone who's played around with it long enough knows that it will give you wrong answers all right and it's partially the software it's what partially the AI is partially how we use it right so the first challenge becomes what we're asking before right telling it not to make stuff up is actually an important thing it was a lawyer who got in trouble for citing a case that did not actually exist so understanding the limitations on how to communicate with it are hugely important the most one of the more popular list of jobs right now is prompt Engineers so let's talk about what a prompt is in regards to a generative Ai and why it's important sure so I'll prompt is just the instructions that you give to the chatbot this is what you're asking and the more precise you can be the more parameters you can put around bands the higher the quality and response you're going to get and now there's this community of people that are sharing tips on how to make the best prompts and they're not always intuitive and so there's these libraries that are emerging of best practices if you wanted prom for uh customer sales then then use these magic words so that's evolving pretty quickly but it's being as directive as possible gets you better response excellent absolutely and just again it's so it's funny it's its own skill set I mean just like a lot of ways just like training a human being his own skill set right we all think we communicate very effectively but if you think of your community very effectively you know just verbally tell someone how to do a job and then see how that comes back right it's just it doesn't come back clean every time so I want to also deal with the Elephant Room on artificial intelligence and that's that's less a concern for owners more so than this for employees is is this replacing people right like what does that look like and how true is that fear yeah that's a very legitimate fear and I think that certainly within a number of technologies that say around robotic process automation where companies have been very keen on this because they take their routine tasks and they turn around and lay people off so that's an understandable fear and people are reluctant to say well why would I teach this bot simply to replace me but I think when you look at organizations the greater threat comes from working at a company that's not introducing technology because if your company isn't introducing Ai and your competitors are then the threat is going out of business because you're not able to meet the demands of of your customers so I think it's companies that invest in technology and take the approach that we're going to free people up to do other roles that they're better suited to are the ones that are most likely to succeed so it's it's not as a negligible risk people should be aware of this but I think the greater threat is that if you're at a company that isn't investing in technology true and I also think at the end of the day you know it's not the highest best use of people's time to do a lot of the stuff that this stuff can do right and you don't want to do either right that they don't want to do right so look I get it if you're someone who just likes pushing paper and not like a term square peg square into a square hole like you know the reality is is that that's a low-hanging fruit piece of infra of work that basically artificial intelligence and Robotics are going to take over right so the reality is is you're right like a business's job is to stay alive first and foremost right and you know holding yourself back from developing the business and creating greater efficiencies is just going to long-term come to your margins and jeopardize the entire future of that business all together and then guess what all those jobs are done so it's a it's a transformational thing now I also say that for most people any form of artificial intelligence and the productivity gains from from it are not something that's like light switch doesn't like turn on overnight and suddenly like oh half my staff is not necessary it doesn't happen it's it's a little bit more of a slow boiling frog in a pot of water and natural attrition through basically people leaving or through basically retirement often solves the problem of well I don't have to lay anybody off right it's just it depends on how you look at things yeah I agree and the technology also creates new tasks new problems that require new skills so we talked about prompt engineer well that was a job that didn't exist two years ago but now it does and you can think of like what are the tasks that you're doing in your company well solving it with AI is going to create some new set of responsibilities and new tasks that people are going to be better suited to so if you're worried about your job I think you want to be someone who's adaptable and open-minded to it but there's going to be so many opportunities to contribute yep absolutely so before we wrap up let's go over some kind of last minute kind of tips and tricks or best practices so basically what is kind of your all-encompassing some memory of advice for people looking to look at this face if you feel like things are moving quickly now but just watch uh we're on an exponential curve and the rate of change is only going to increase and so our advice is to get in with something pick a way that you can improve a workflow some pain that's really bothering in your organization and see is there a way that AI would be able to do this cheaper faster better for us and start learning because once you start to see the opportunities here it'll snowball for you and you can make this switch from looking at a problem in terms of just efficiency how can we do our current work better faster cheaper over time you can start to think about well how can this change my business model if I'm not doing this anymore what's a different way that I could serve my customers and that can lead to Innovation opportunities so get in start learning get more efficient and then get more Innovative absolutely and I would also say that it's kind of fun it's always perspective everybody you know so much of your concept of the speed of the departure depends on your starting point right and I was watching a show the other day that was based in the 1960s and the character was going on about how the rate of change is like unprecedented compared to previous generations and whatever which was true at the time because we're always sitting on a bit of an exponential at least in the last I would say post-industrial Revolution era for sure we're always sitting on an exponential curve but when you're sitting on that the hill initially does not look that big but if you don't start progressing up that hill then basically and you're left behind it just starts to move to the Horizon and eventually beyond that right so I would always say General best practices always be a continuous learner because this stuff will never leave you behind but this is why kids always find it to be less intimidating because it's their starting point right a starting point is something way back in the past yeah if you're worried about adapting to AI I think I would say this is real and this is going to have widespread implications in all parts of society it's the most important in Innovation since fire so it's going to continue and it doesn't get easier a year from now it gets harder this is what you know we all think about like oh I'm going to the bank I'm using the bank machine here's all these old people or people who basically can't don't even know it like just refuse to use the baby because they're intimidated by it right like they didn't start off as like luddites they started off understanding the tools of the era they were born in and they were that they were used to and then basically stop learning all right like that's the reality of it and it's as I tell anyone who thinks that they can skirt like oh I don't know how to use this I'm just going to ignore it it's like you think this is gonna get easier for you imagine how much harder this is going to be in 30 years right I would rather not be someone scared of every tool in the universe and the bank teller one is a good example of automation right when that was introduced the idea was well this is going to eliminate all of the jobs for tellers but reducing the cost to providing the services allowed Banks to increase the number of branches that they opened and so more people were served and the number of people who were tellers didn't actually go down and this is the funny thing is that like it's the economic concept of consuming the marginal Improvement right so let's imagine you can wave magic wand and everyone in an industry could increase their margins by by 100 percent tomorrow like they could double their profitability tomorrow what do you think is going to happen do you think that they're just going to pocket that or do you think that they're going to use that to invest to try to gain more market share or expand their services right like it is the natural in actually inbred into all businesses to grow right so they're going to invest that in growth which basically means expansion of services compression of margin is probably back to where they were before because people were used to it so it's acceptable and basically and but it'll be smaller margin on larger pies right so This Is Not Unusual like because if we didn't follow that pattern then the reality is most businesses would just still have the footprint they did 30 40 years ago and would be less profitable in general but nope everybody wants to grow the pie and take a slice yeah exactly excellent Sean thank you so much for taking the time we really appreciate it we meandered quite around this a bit but this is not a linear constant to understand right you have to understand a bunch of different facets and dimensions and Concepts around it right indeed and these were really great questions I really enjoyed this excellent and where can people find you the easiest way to find this is on our website so it's aiguides.co and you can also connect with me on LinkedIn so John strap be happy to chat with anyone excellent thank you very much thank thank you so that was my interview with John Stroud at AI guys hope you enjoyed that hope you got some sense that you know what you there are easy ways to get started here and that awareness and education here is vital to preserving a business at any stage at any stage of your career quite honestly so basically by all means take a look at AI guys what they're putting out and just try to keep abreast of some of this stuff as always if you enjoyed this podcast please leave a review on Apple podcast SoundCloud Stitch or Spotify or whatever is that your podcast until next time take care this podcast was brought to you by Woodgate Financial an award-winning financial planning firm catering to high net worth individuals business owners and their families to learn more go to woodgate.com you can subscribe to this podcast on Apple podcast Stitcher Google Play and Spotify or find more episodes at jasonparera.ca you can even ask