In this episode of Digital Marketing Masters, host Matt Rouse is joined by Kate vanderVoort, founder and CEO of Social Mediology and the AI Success Lab, to discuss the integration of AI into business practices. Kate shares her insights on how companies can transition from having no AI involvement to effectively incorporating AI into their operations.
She emphasizes the importance of understanding AI's implications and the need for businesses to develop a human-centered, responsible approach to AI adoption. The conversation highlights the challenges companies face in navigating the rapid advancements in AI technology and the necessity of creating a supportive culture for AI integration.
Kate and Matt delve into practical strategies for businesses to start their AI journey, including conducting AI business audits and fostering AI champions within organizations. They discuss the significance of leadership involvement in AI training and policy development, and how small to medium-sized businesses are often more agile in adopting AI solutions. The episode also touches on the importance of addressing employee concerns about AI, such as job security, and the potential for AI to streamline operations and save time. Listeners are encouraged to consider the strategic implementation of AI to enhance productivity and innovation within their companies.
Sponsored by SMB Autopilot, your local marketing on autopilot.
https://aisuccesslab.com/strategy
Kate in Linkedin: https://www.linkedin.com/in/katevandervoort-social-mediology/
Matt on LinkedIn: https://www.linkedin.com/in/mattmrouse/
https://matthewrouse.com
Hook Digital Marketing:
Canada
USA
[00:00:04]
AI Narrator #1:
Today on the digital marketing masters podcast. All the way from Australia, it's Kate vanderVoort. And we're talking about getting your business started with
[00:00:32] Matt Rouse:
AI. Hey, everybody, and welcome back to Digital Marketing Masters. I'm your host, Matt Rouse. Today, my guest is Kate vanderVoort. How are you doing, Kate? I'm great. Thanks, Matt. Thanks for having me. So, Kate, why don't you give us a quick, you know, like, one or two breadth of, who you are and what you do?
[00:00:50] Kate vanerVoort:
Sure. So I'm founder and CEO of Social Mediology and the AI Success Lab, and we work with companies and businesses and individuals sometimes on how to use AI and integrate AI into all aspects of their business in really human centered, responsible, and ethical ways.
[00:01:11] Matt Rouse:
And you're
[00:01:12] Kate vanerVoort:
in Australia. Right? I am.
[00:01:14] Matt Rouse:
Nice. And I'm in Canada, and my business partner who's listening in the background right now is in Costa Rica. We're just all over. We got, like, half the Earth covered here, I think. Kate, I wanted to have you on the show specifically because this has come up for me, and when you and I were talking, this has come up for you a lot in your own work, is that companies basically are not not are they just not prepared for AI. They have no idea what they're doing when it comes to AI. And so I think what we wanna try and capture here today is how does a company go from zero, like, no AI use whatsoever, to getting started or at least on the right path, you know, to getting AI to work for them?
[00:02:02] Kate vanerVoort:
Yeah. So I think it's worth stepping back just a little bit to, you know, many of us who are early adopters of technology and have moved into working with AI had our roots and our start in sort of digital marketing and often social media or community building or online marketing in some way. And in so many ways, this reminds me of when I started my company fifteen years ago when everybody was grappling with social media. And it was this big black hole of how does this change how we communicate with our customers, and it's taken a fair chunk of time for you know, there were the early adopter companies, but it took quite a few years for a lot of businesses and companies to get their head around what this two way communication suddenly meant with their customers.
And so in some ways, AI, the chat g p t moment, the last couple of years that's been really focused on what does AI mean now for us personally, professionally, for humanity, companies and everyone really are really grappling with what those you you know, what this technology actually means for us. And unlike social media and digital marketing, which was a fairly big transition, this has far greater reaching implications. It's happening at a pace that, you know, hold on to your hat. I can't keep up, and I'm, you know, I'm sure you feel the same, Matt. We're in this all day every day, and keeping up is more than a full time job for a team, let alone individuals.
And so companies at the moment are really struggling with this this raging river or a fire hose of new apps and tools and, you know, LLMs and rag and do we build our own? Do we use the consumer stuff? Do we use Copilot? Do we use Gemini? Really trying to figure out where to even dip their toe in or take that first sip. Right. And so that's where we kinda are at the moment is I'm not sure how it is in Canada, but here in Australia, a lot of companies are sort of frozen not knowing where to start, not knowing how to even, you know, dip their toe in the water. And we talked about when you were on my podcast, you know, the people who are sneaking home with the USB sticks using AI at home because their companies have not yet put anything in place that allows them to do that within the workplace.
[00:04:39] Matt Rouse:
Well, and not just companies, but organizations in general. Right? Your educational institutions and government institutions and nonprofits and
[00:04:49] Narrator AI:
Your host, Matt Rouse, wrote a science fiction short story that is included in the upcoming anthology, Tales of Galactic Pest Control, an Amazing Stories anthology edited by David Gerald and Tom Easton. Amazing Stories has a long tradition of publishing thought provoking and entertaining speculative fiction. With Tales of Galactic Pest Control, we're working with talented writers like Matt and more than 30 other authors to bring you fresh, fun, and unexpected stories about pest control in all its strange and fascinating forms. Get your digital or printed copy by backing this book on Kickstarter for less than it will cost in the bookstore.
[00:05:29] Matt Rouse:
If you're not a for profit company, one of the best places you can start, and we'll talk about for profit companies in a second, but one of the best places you can start is just getting a policy down. Right? Having something that says, these are the apps that our our employees are allowed to use, and this is the process they have to follow when they use them. And, I mean, there's lots of there's lots of ways you can kind of get some drafts of those policies. But I think that, you know, just to to have something down on paper kind of thing for your employees is a really big step in the right direction.
[00:06:10] Kate vanerVoort:
Yeah. And it's interesting because and I'll I'll comment on that in just a moment, but the ones that are the early adopters that are running with this and seeing massive benefit are the small businesses. The the small to medium sized businesses. They're more agile. They've got less, you know, legacy systems potentially that they've gotta get on board. And I resisted for a long time setting up a membership and online courses for that community, but that's what they want. They're they're ready, you know, go. And they're the ones that have really embraced it. And from that perspective, every company or business and you're right. I was using the word company, but it's every I do a lot of work in the not for profit sector and in that more small to medium sized sector.
But every organization, business, company needs to assume that a portion of their workforce are already using AI. Right. And rather than you know, the approach I take is is actually not to start with a policy because I've found that a lot of teams, a lot of people, employees within a business are quite fearful of what AI means. And it's everything from, will AI take my job? You've written a book on that. It's fair. If I use AI, am I just showing my boss that I'm replaceable? Or if I use AI, will I seem to be lazy? You know, there's a lot of misconceptions about AI. And until you address those, if you actually come in and put a policy in over the top of people,
[00:07:40] Matt Rouse:
it can be met with quite a bit of resistance if people have not yet grappled with some of those issues. So a lot of I think there's definitely a training thing that has to happen there as well. But I I mean, when I talk about putting a policy in place is to to get started.
[00:07:55] Kate vanerVoort:
Right? Yeah. Absolutely. But what I suggest with that when we first start having conversations is be really open with the whole organization and say, okay. We're dipping our toe into AI. And what we wanna know, what we wanna understand is who are our AI champions already within the business? Who's already started dabbling with AI?
[00:08:16] Matt Rouse:
Just find the lazy ones. No. I'm just saying Yeah. Well, you're not
[00:08:20] Kate vanerVoort:
no. Well, I yeah. Again, I think, culturally, you wanna make sure that that's not how people feel. But if they can come to the fore and then you work with those people who actually have an champions to then get other staff and other, you know, other teams on board within a a business. And so, absolutely, I think there needs to be some high level, you know, we're only gonna use Copilot or we're only go we're not gonna put any of our data in as we're getting started. We're only going to use them to give us frameworks and strategies that we can then tailor to ourselves. So whatever that approach is, some of those high level guardrails. But I really like to encourage organizations to work with their staff on developing those policies because it doesn't take long to hit that resistance. And once you hit that resistance, there's a lot of work to do to get past that often.
[00:09:25] Matt Rouse:
It's true. I mean, I've been in, you know, back in in my IT contracting days, I've been hired to put a software system in place that changes how people do their jobs, and then they just throw up the walls and everything. And I'm like, I wasn't paid to, like,
[00:09:43] Kate vanerVoort:
break down everyone's resistance. Right? I I was, like, paid to, you know, make Microsoft project work, you know, that kind of thing. You see, I get paid to break down resistance. That's right. The whole change piece is a big part of the work that we're doing at the moment.
[00:09:56] Matt Rouse:
I I you know, I think that one thing that I find with companies kind of in the small to medium business size range, especially on the maybe on the larger side, you know, your hundred employee business kind of thing. The, like, president, owner, CEO, C suite, whoever it is who's who's kind of at the top of the pyramid in that, you know, in that org chart is they already have a more than full time job, and they rarely have had a chance to even touch an AI system. Like, they haven't opened chat GPT. They've just heard some stuff. They've seen some videos, you know, whatever, but they haven't really used it.
And I feel like when I talk to businesses where there's been no AI use whatsoever by the kind of people in charge, that it's gonna be extremely difficult for them to implement anything because they don't even understand what it is. Right? And there's definitely, like, a training gap there. And it's hard to say, you know, okay, missus CEO of, you know, x y z corp. We should get an AI policy in place, and we should have, you know, training for employees. And and we need to get legal involved and IT involved, and we need to kind of get this whole system up and running. Maybe we're gonna have an AI counsel, and they're like, I don't even know what it does. Right?
So do you find that the approach is kind of best to kind of train the trainer kind of thing. Right? Like, train the leaders, demo stuff for them first, or is it more of a from the inside up in the organization where you find those AI champions and you say, show us how you're using it so that we can show the boss how it's being used?
[00:11:47] Kate vanerVoort:
I think it's different for every organization. I have got quite a large, more than one handful of companies or businesses where I've done the initial piece of work, and now they're frozen because leadership doesn't understand. Right. Because what they did was just say, oh, train my staff or do some work with my staff, but because they were not involved in that process or didn't engage in the way that maybe could have been really beneficial, they're then not in a position to approve and help with the rollout beyond that. So a lot of the work I've done has gotten has stopped at the moment, and I think the reason is, you know, they hear uncle Bob at the barbecue on the weekend talk about how AI is breaching copyright or the geopolitical issues or all these issues that get are getting talked about outside of the workplace. And sometimes that seeps into the workplace. But you're absolutely right. Unless that leadership long term, unless that leadership is on board, Unless that leadership long term, unless that leadership is on board, they're never gonna approve what needs to be approved to have the technology integrated. So I think this year is a really funny year, and we're we're going to see those that just dive in. Although, you know, one of the other things that I'm seeing a lot of is we've had the companies or the businesses that wanna build their own models. They wanna build their own internal solutions.
And I actually think that those that wait a little bit longer, some of that stuff are gonna be really big investments that were not needed because of the tech coming down the line. And so in many places, I'm saying just sit tight for a bit and let's see where autonomous agents get to because maybe, you know, educate your staff now, have them understand what LLMs are, how to prompt, how to work with AI, how to decide what aspects of their role, you know, can be done by AI or assisted with or collaborating in collaboration with AI. So do that culture change in education piece, but maybe sit tight just for a little bit on the big investments in the tech. And it's the same with some companies only wanna use Copilot because they're in the Microsoft environment.
Right. And that's a place to start, but it's not a cheap place to start. You know, if you've got thousands of staff, Copilot licenses, that adds up. Now you're blowing $20 a month on licenses and Yeah. And maybe Copilot is not the best solution. Maybe there are some other solutions that are either available already or about to be. And so it's tricky for for senior management to be making those decisions at the moment. And as much as that's the work that I do, companies need to be pulling in the expertise to help them navigate that. And I think that that will and I think you said on on my podcast that that they've already got AI expertise, and absolutely, most companies do, but they don't necessarily have the strategic expertise on integrating AI across every task, every role, every department, and across the, you know, the whole of an organization. People have often got a real handle on what it means for their role or their team or their part of the organization, but maybe not necessarily a whole of organization approach.
[00:15:11] AI Narrator #1:
Alfonso here, and I want you to get more customers for your local business on autopilot with SMB Autopilot. Boost your business's visibility with our AI driven local marketing, and it's done for you so that you can focus on what you do best and leave the local marketing to us. Visit smbautopilot.ai to learn more.
[00:15:31] Matt Rouse:
Yeah. I think, kind of along those lines, one thing that I've seen that seems to be working fairly well in some organizations is kind of these, like, the equivalent of, like, a hackathon. You know, we're gonna get all the marketing staff together or the marketing and sales. Depends on the size of your company, obviously. You know? Maybe if you have three salespeople, one marketing person, and an admin, you get those five people together. And the idea is we're gonna basically list all of our all of our tasks that are kind of, like, almost like what's our job description. Right? Like, what are all the things that we have to do on a day to day basis that are not having conversations with people, building relationships, you know, stuff that your AI is not capable of?
You list down those tasks, and then you start going,
[00:16:23] Kate vanerVoort:
which of these things do I think an AI could possibly do, and how much time is it gonna save me? Right? Yeah. And that's actually the first step we do when consulting with companies is that AI business audit. So we have a process whereby every person in the organization lists or goes through, so tracks for a week what are the tasks that they're doing on a repeatable basis. And then we have a process whereby we can analyze all of that information across the organization. It's a great amount of data that we get when we do that process. And we can actually go back and say, well, look. And in fact, we did some work recently with a childcare franchise where we were able to save them forty hours a week across a team of 15 people by doing a business audit and showing them how they could do that. And so in one month, they were able to save that amount of time. And so data doesn't lie, and dollars are a big motivator often in terms of getting senior leadership on board. But
[00:17:25] Matt Rouse:
really, I'm just great is you collect all that data, and you can use the AI to sort it and analyze it for you. Absolutely.
[00:17:31] Kate vanerVoort:
Absolutely. It does tend to be a little we've refined our process because it does tend to be a little overzealous in what it thinks it can do.
[00:17:40] Matt Rouse:
Chatty chatty speak talks It's chatty. Yeah. You gotta you gotta tell it to calm down a little bit. You don't wanna use that, what's it called, high output SONNET 3.7 that just, like, gives you, like, pages and pages of of tokens. But Yeah. That's aside the point. So I think something that's really interesting there is that if you're a solo business owner, like or maybe you just have, like, a couple people in your business, the same process still holds true. Right? You can list all the things that you have to do repeatedly, that you have to do on a regular basis. Right? Like, okay. I gotta collect data from these five systems and put them all in a report and email it to, you know, Janet. So then you could look at those things. You can say, how much time am I spending on these things?
And you can say, okay. Well, you know, maybe I can use a combination of, you know, an automation system like Zapier or Pabbly Connect or Zoho Flow or whatever it is. And I can use it to pull the data from each of those systems that I'm collecting manually, and then I can have the AI collect all of that stuff and write the report for me and email it to Janet. So and then I can schedule it. And now I've just saved myself. You know, if that takes three hours every time you do it and you gotta do it every week, you know, you just save yourself twelve hours a month. And now you got twelve hours a month you can go out and find more business. Right? Or, you know, do whatever the more important task is. Yes. Sorry.
[00:19:12] Kate vanerVoort:
That process that we do with companies, we've actually turned into a custom GPT for our coaching program. So they do exactly the same process. They fill out an Excel spreadsheet of what are the tasks that they're doing on a repeatable basis. They give it to the custom GPT. The custom GPT makes a list of recommendations of where they should be looking at to save time and then gives them a road map of how to roll that out across their business. And in the four weeks of our coaching program, by the end of it, they've built a team of custom GPTs that address that initial business order that helps them to do that, kinda closes that that loop.
[00:19:51] Matt Rouse:
You know, this is kind of an aside, but I have not sold on Jeep custom GPTs. And Uh-huh. No. Tell you why. Let me tell you why. I use another program called Sym Theory, right, which is made in Australia, actually. Yep. And I can essentially do the same thing as a custom GBT, but I can put a whole bunch more kind of data and input into it. Mhmm. But the only downside is I can't give access to that to someone else. Right? Whereas chat like, custom GBTs, you could. And I'll tell you why I think I feel a little like I got burned on custom GBTs is that they were like, you're gonna make custom GBTs, and you we're gonna have a marketplace, and you could sell them. And I was like, I'm gonna sell the shit out of some custom GPTs. Yeah. And I built a whole bunch, and I figured out how they work. Then when they didn't marketplace, never appeared, and I was like, oh,
[00:20:42] Kate vanerVoort:
come on. You're not alone. A lot of people did exactly the same. But I'm still very sold on custom GPTs for two main reasons. One, it's in a workflow that most small business, small businesses feel comfortable using. So when you mentioned that platform, I went and had a look at it. I think the costs are a bit prohibitive for small businesses. And there's a a tech
[00:21:08] Matt Rouse:
learning tool that's right. It is a little bit of techy tool, but
[00:21:12] Kate vanerVoort:
the And for a lot of people, they wanna use things in the platforms that they're already using. And so most of the token cost to it. Right? That's where it it eats them. Yeah. But as soon as you start do that. Yeah. As soon as you start talking about APIs and it that then goes into a level that, to be really honest, the majority of small business owners are there. Nobody's gonna do that. And they're not able to do that. And so the custom GPTs, you know, we use them internally. We use them as lead magnets. Every workshop, every course, everything that we run externally in terms of having people in our membership or our community, we create a custom GPT that supports them with that process.
Right. So I have a free video on how to train your Flux LoRa, so how to create AI photo shoots. It's a pretty complicated process. Right? But I created a custom GPT that steps people through that process. So it's like they've got a little assistant, they can watch the video, but they've also got a little assistant where they're actually getting coached and supported on the side. And the other way I love using custom GPTs is, as you say, giving access to other people. Right. And VAs or contractors, I create custom GPTs for every task I wanna do, but then they can use those on my behalf and sound more like me and produce content that is, you know, in line with the brand voice, which is largely me.
[00:22:45] Matt Rouse:
Right. You know, I think if they connect up, like, the scheduling and, like, operator and stuff with custom GBTs, that that'll be a selling feature for me. I'll probably head back to it again.
[00:22:58] Kate vanerVoort:
Done a lot with operator. And when Manus came out, operator has been shelved, unfortunately, for $200 US a month. Yeah. I'm gonna be hard pressed to to,
[00:23:10] Matt Rouse:
yeah, renew. Manus seems pretty good. I mean, when it comes to that kind of stuff because we I mean, we're essentially a tech technical marketing company. We have developers on staff and stuff like that. Right? So, you know, if we're gonna build an automation, we're you know, even if it's something like and this you know, if you are somewhat development inclined, this is an easy way you can do some a lot of complex tasks is you can just boot up WordPress. Right? Or if you already have a WordPress website, use WS form, which has access to call APIs, and it also has webhooks.
And then you can connect that into Zapier, whatever your automation system is, do any of the processing you want, and send it back to WordPress. So you could actually make entire applications with that system if you wanted to without writing any code. Right?
[00:24:05] Kate vanerVoort:
So See, we're an education stuff there. And consulting business. We're not a tech tech implementation business. We partner with people like you when we when it comes to building the solutions. And and most of the people that we're working with, whether it's in organizations or companies or small businesses, not at the tech. So I kinda do that first piece before handing over to the tech implementation, which is that what does this mean for our business? How how are we going to create a culture here that allows for a transition into a new way of doing business? So we're really right at ground zero back at that step one before handing off for the tech implementation side of things.
[00:24:48] Matt Rouse:
I think something that's interesting is if you do have companies that are large enough that they have either IT staff or they have, like, a managed services company or something that they work with, there's definitely an opportunity there to get them to help you build solutions as well. Right? If you need something more complex, then, you know, maybe you could build yourself. That's it's not simple if you're not a developer. A lot of this stuff can be handled by your IT management people without going to an outside developer, which can save you a ton of money. Right? Like, building a piece of software now, even though it's a lot cheaper than it was before, I mean, it's still gonna be $6.06 6 to $10,000 for something reasonably simple, right, that just does, you know, data retrieval and analysis and stuff like that. Your IT people might be able to throw something together with what you have and some AI and and get that done too. So that's kind of a way for your company to sneak sneaks in my eye in the back door.
[00:25:46] Kate vanerVoort:
Sure. But we're still in that sticky taping things together phase. Right? And and we did that with social media in the early days, but this is moving a lot faster. So I remember when you couldn't, you couldn't schedule to like half the platforms.
[00:26:02] Matt Rouse:
You were like, oh, I can schedule this, but I can't schedule that. And I got a natively post on Instagram, but I can't do it on Twitter. And you know, like.
[00:26:10] Kate vanerVoort:
I remember when there were no images.
[00:26:12] Matt Rouse:
Oh, yeah. I know it's
[00:26:13] Kate vanerVoort:
So that makes me feel really old. Yeah. Me too. But I think for most companies, they'll have, like, those burning challenges, those really big issues, and you can just address that with a simple automation, and you're right. Your internal teams can often manage that. But I think we're undergoing a huge transition, and the companies that become AI first companies where it's really embedded into the culture of the organization, I think it's too early to be investing in big solutions around that. So Yeah. No. I think it's definitely in in building the the
[00:26:52] Matt Rouse:
the culture of using it rather than building your own systems.
[00:26:57] Kate vanerVoort:
But if you say to every person in the organization, imagine if you had a Kate two point o or a Matt two point o. You had a little assistant that was sitting next to you. What are the things that you would hand over to that that AI? And that's really the best place to start within the guardrails or parameters that the business sets, but that's when people can start to see the tangible benefit of that. And Right. Then if you are really and but doing that after doing an audit so that you're actually getting the economy of scale, that if everybody email is something that everybody does, is there some layer that we can put in where AI supports that across the organization, not just at the individual level? And I think that's why a lot of companies or businesses are frozen at the moment. There's so many ways that you can come at the AI conversation, And it's just about finding what works naturally for each business. I wish I could bottle it. I wish I had a system that works for every business, but it's different for every business. And tech comfort with technology, because it's often been quite a siloed part of a business, that can be a really big barrier. If you've got people who are really resistant to AI, there's a there's some work to be done before they'll even consider using AI. So
[00:28:16] Matt Rouse:
Well, there's another consideration that you know, I was just talking with somebody last week about, which which is kind of the idea of this siloed AI technology. Like, let's say my company is is a % Google Workspace shop. Right? Everything we do is on Google Workspace. The only AI anybody uses in our company is Gemini. And then Gemini updates their system, and now it doesn't do the thing that we used to use it for. Right? Well, now you're stuck. Right? But if you do have a culture in your company of testing and and messing around with some AI systems to kind of kind of get used to how they work and not just how one specific program works like Copilot or something, right, then you've got the ability to pivot. Right? If you if the AI system changes, something better comes out, whatever it is, you know, then you can go back with your teams and say, you know, oh, let's research this. Let's figure out, you know, how to use this other better system.
And, you know, one other shout out I wanted to have, you know, Kate, for using, you know, what is essentially consulting firm, right, like yourself, is that audit piece is super important if you have regulation in your industry.
[00:29:32] Kate vanerVoort:
Absolutely.
[00:29:33] Matt Rouse:
Like in Canada, if you are a medical related company, you have to host all your patient data in Canada, which means you can't send it to OpenAI with a question about a patient or something, right, Unless you anonymize it first. And so these are things that a consultant is gonna know that your business may not have considered.
[00:29:54] Kate vanerVoort:
And Australia and Canada are quite similar in terms of how legislated we are. And I was actually gonna say that when we were speaking earlier around. And just to your point there actually around most companies wanna use AI within the ecosystem that they're already using. So if that is Microsoft or if that is the Google Workspace, they wanna use Gemini or Copilot. The problem with that or sorry. The the benefit of that is that it can unlock their data in what is a secure environment that they're already familiar with. They've already got their files in the cloud. They already have addressed their data security issues. But both of those models are fairly limited in what you can actually create using AI in those models.
And so that's where you look at, okay, what other tools might be available, but then you've got all of the issues around data privacy, where's the data stored. A lot of companies do not wanna use platforms out of China even though they're probably using platforms out of China and maybe just don't realize it. Right. And so that's where it because most of the demonstrations that I use when I'm in a boardroom are not within Copilot or within Workspace because within Google because they're not the wow. Oh my goodness. Imagine if we could do that. Imagine if I could build that 20 slide slide deck in three minutes instead of three days.
And so, yeah, we're we're just still in this funny place where there's a ton of different ways that you can do AI, but most wanna start in that closed environment. But they're gonna be pretty limited in what they're able to do within that environment to start with. Yeah. And I I think,
[00:31:36] Matt Rouse:
you know, the the AI scaling law right now is averaging seven months for the doubling of of AI power. If you compare that with, like, Moore's law, which is eighteen months for the, you know, doubling of of the processing power of microprocessors, so it's essentially twice as fast of, you know, the the doubling principle. And, like, I'm been here three months, just in the last couple of weeks. Well, it's it's averaged out to seven. Right? But sometimes it's three, sometimes it's 10. Right? You know, like But averaged over what? Average. Well, it's been averaged. They've kind of tracked back since 2023. But, also, they can track back before that to when the transformer was invented, right, which I believe was back in 02/2012.
So that was a DeepMind invention originally. So Yeah. But, like, what I was gonna say was, like, I was doing my introduction to AI presentation. I finished it at 04:00 the night before the presentation. And at 5PM, OpenAI announced native images in their system where you could do image editing and all this stuff. And I had a slide where I was like, don't use OpenAI for images because Dolly three sucks. Right? And so I gotta go back and change, you know, two or three slides of my presentation to talk about this new thing, and then we gotta make a demo for it, and we gotta set this all up. And that's how quick it moves. You know? I I barely get a chance to do a presentation that I don't have to redo the same day of the time I'm presenting it because it moves so quickly.
Happens to the whole time. That's right. But the other thing is that I mean, there's it's a two two sided coin. Right? One side is it's getting better so quick that something you can't really do now or it doesn't work great might work great six months from now or three months from now or let next week maybe. It's nice. Yep. And the other side of that is it's hard to keep up. So I think it definitely helps having somebody on your team, you know, somebody who can help you, whether that's vendor, consultant, whatever. You know? Like, Kate's perfect example. Kate, if somebody wants to reach out to you to get more information or an audit or something, how would they do that? Yeah. So probably the best place to connect with me is on LinkedIn under Kate Vandevoort.
[00:33:54] Kate vanerVoort:
And we also have the AI Success Lab Facebook group where we share a lot of the breaking news, but real strategies on how you can use AI for business today that moves the dial in your business. So they're probably the best two places to connect. AI. Perfect. You can also connect with me on LinkedIn if you wanna hear about chickens because that's mostly what I write about. Chickens and AI. Great to see everyone again out, and, Kate,
[00:34:18] Matt Rouse:
so excited to have you on my show. It was a great great talk on the AI Grapple podcast as well, which everybody should go listen to, not just because I was on it. There's other great guests as well. But, Kate, thanks so much for being on the show today.
[00:34:31] Kate vanerVoort:
Thanks, Matt.
[00:34:34] Matt Rouse:
It's been fun.
Today on the digital marketing masters podcast. All the way from Australia, it's Kate vanderVoort. And we're talking about getting your business started with
[00:00:32] Matt Rouse:
AI. Hey, everybody, and welcome back to Digital Marketing Masters. I'm your host, Matt Rouse. Today, my guest is Kate vanderVoort. How are you doing, Kate? I'm great. Thanks, Matt. Thanks for having me. So, Kate, why don't you give us a quick, you know, like, one or two breadth of, who you are and what you do?
[00:00:50] Kate vanerVoort:
Sure. So I'm founder and CEO of Social Mediology and the AI Success Lab, and we work with companies and businesses and individuals sometimes on how to use AI and integrate AI into all aspects of their business in really human centered, responsible, and ethical ways.
[00:01:11] Matt Rouse:
And you're
[00:01:12] Kate vanerVoort:
in Australia. Right? I am.
[00:01:14] Matt Rouse:
Nice. And I'm in Canada, and my business partner who's listening in the background right now is in Costa Rica. We're just all over. We got, like, half the Earth covered here, I think. Kate, I wanted to have you on the show specifically because this has come up for me, and when you and I were talking, this has come up for you a lot in your own work, is that companies basically are not not are they just not prepared for AI. They have no idea what they're doing when it comes to AI. And so I think what we wanna try and capture here today is how does a company go from zero, like, no AI use whatsoever, to getting started or at least on the right path, you know, to getting AI to work for them?
[00:02:02] Kate vanerVoort:
Yeah. So I think it's worth stepping back just a little bit to, you know, many of us who are early adopters of technology and have moved into working with AI had our roots and our start in sort of digital marketing and often social media or community building or online marketing in some way. And in so many ways, this reminds me of when I started my company fifteen years ago when everybody was grappling with social media. And it was this big black hole of how does this change how we communicate with our customers, and it's taken a fair chunk of time for you know, there were the early adopter companies, but it took quite a few years for a lot of businesses and companies to get their head around what this two way communication suddenly meant with their customers.
And so in some ways, AI, the chat g p t moment, the last couple of years that's been really focused on what does AI mean now for us personally, professionally, for humanity, companies and everyone really are really grappling with what those you you know, what this technology actually means for us. And unlike social media and digital marketing, which was a fairly big transition, this has far greater reaching implications. It's happening at a pace that, you know, hold on to your hat. I can't keep up, and I'm, you know, I'm sure you feel the same, Matt. We're in this all day every day, and keeping up is more than a full time job for a team, let alone individuals.
And so companies at the moment are really struggling with this this raging river or a fire hose of new apps and tools and, you know, LLMs and rag and do we build our own? Do we use the consumer stuff? Do we use Copilot? Do we use Gemini? Really trying to figure out where to even dip their toe in or take that first sip. Right. And so that's where we kinda are at the moment is I'm not sure how it is in Canada, but here in Australia, a lot of companies are sort of frozen not knowing where to start, not knowing how to even, you know, dip their toe in the water. And we talked about when you were on my podcast, you know, the people who are sneaking home with the USB sticks using AI at home because their companies have not yet put anything in place that allows them to do that within the workplace.
[00:04:39] Matt Rouse:
Well, and not just companies, but organizations in general. Right? Your educational institutions and government institutions and nonprofits and
[00:04:49] Narrator AI:
Your host, Matt Rouse, wrote a science fiction short story that is included in the upcoming anthology, Tales of Galactic Pest Control, an Amazing Stories anthology edited by David Gerald and Tom Easton. Amazing Stories has a long tradition of publishing thought provoking and entertaining speculative fiction. With Tales of Galactic Pest Control, we're working with talented writers like Matt and more than 30 other authors to bring you fresh, fun, and unexpected stories about pest control in all its strange and fascinating forms. Get your digital or printed copy by backing this book on Kickstarter for less than it will cost in the bookstore.
[00:05:29] Matt Rouse:
If you're not a for profit company, one of the best places you can start, and we'll talk about for profit companies in a second, but one of the best places you can start is just getting a policy down. Right? Having something that says, these are the apps that our our employees are allowed to use, and this is the process they have to follow when they use them. And, I mean, there's lots of there's lots of ways you can kind of get some drafts of those policies. But I think that, you know, just to to have something down on paper kind of thing for your employees is a really big step in the right direction.
[00:06:10] Kate vanerVoort:
Yeah. And it's interesting because and I'll I'll comment on that in just a moment, but the ones that are the early adopters that are running with this and seeing massive benefit are the small businesses. The the small to medium sized businesses. They're more agile. They've got less, you know, legacy systems potentially that they've gotta get on board. And I resisted for a long time setting up a membership and online courses for that community, but that's what they want. They're they're ready, you know, go. And they're the ones that have really embraced it. And from that perspective, every company or business and you're right. I was using the word company, but it's every I do a lot of work in the not for profit sector and in that more small to medium sized sector.
But every organization, business, company needs to assume that a portion of their workforce are already using AI. Right. And rather than you know, the approach I take is is actually not to start with a policy because I've found that a lot of teams, a lot of people, employees within a business are quite fearful of what AI means. And it's everything from, will AI take my job? You've written a book on that. It's fair. If I use AI, am I just showing my boss that I'm replaceable? Or if I use AI, will I seem to be lazy? You know, there's a lot of misconceptions about AI. And until you address those, if you actually come in and put a policy in over the top of people,
[00:07:40] Matt Rouse:
it can be met with quite a bit of resistance if people have not yet grappled with some of those issues. So a lot of I think there's definitely a training thing that has to happen there as well. But I I mean, when I talk about putting a policy in place is to to get started.
[00:07:55] Kate vanerVoort:
Right? Yeah. Absolutely. But what I suggest with that when we first start having conversations is be really open with the whole organization and say, okay. We're dipping our toe into AI. And what we wanna know, what we wanna understand is who are our AI champions already within the business? Who's already started dabbling with AI?
[00:08:16] Matt Rouse:
Just find the lazy ones. No. I'm just saying Yeah. Well, you're not
[00:08:20] Kate vanerVoort:
no. Well, I yeah. Again, I think, culturally, you wanna make sure that that's not how people feel. But if they can come to the fore and then you work with those people who actually have an champions to then get other staff and other, you know, other teams on board within a a business. And so, absolutely, I think there needs to be some high level, you know, we're only gonna use Copilot or we're only go we're not gonna put any of our data in as we're getting started. We're only going to use them to give us frameworks and strategies that we can then tailor to ourselves. So whatever that approach is, some of those high level guardrails. But I really like to encourage organizations to work with their staff on developing those policies because it doesn't take long to hit that resistance. And once you hit that resistance, there's a lot of work to do to get past that often.
[00:09:25] Matt Rouse:
It's true. I mean, I've been in, you know, back in in my IT contracting days, I've been hired to put a software system in place that changes how people do their jobs, and then they just throw up the walls and everything. And I'm like, I wasn't paid to, like,
[00:09:43] Kate vanerVoort:
break down everyone's resistance. Right? I I was, like, paid to, you know, make Microsoft project work, you know, that kind of thing. You see, I get paid to break down resistance. That's right. The whole change piece is a big part of the work that we're doing at the moment.
[00:09:56] Matt Rouse:
I I you know, I think that one thing that I find with companies kind of in the small to medium business size range, especially on the maybe on the larger side, you know, your hundred employee business kind of thing. The, like, president, owner, CEO, C suite, whoever it is who's who's kind of at the top of the pyramid in that, you know, in that org chart is they already have a more than full time job, and they rarely have had a chance to even touch an AI system. Like, they haven't opened chat GPT. They've just heard some stuff. They've seen some videos, you know, whatever, but they haven't really used it.
And I feel like when I talk to businesses where there's been no AI use whatsoever by the kind of people in charge, that it's gonna be extremely difficult for them to implement anything because they don't even understand what it is. Right? And there's definitely, like, a training gap there. And it's hard to say, you know, okay, missus CEO of, you know, x y z corp. We should get an AI policy in place, and we should have, you know, training for employees. And and we need to get legal involved and IT involved, and we need to kind of get this whole system up and running. Maybe we're gonna have an AI counsel, and they're like, I don't even know what it does. Right?
So do you find that the approach is kind of best to kind of train the trainer kind of thing. Right? Like, train the leaders, demo stuff for them first, or is it more of a from the inside up in the organization where you find those AI champions and you say, show us how you're using it so that we can show the boss how it's being used?
[00:11:47] Kate vanerVoort:
I think it's different for every organization. I have got quite a large, more than one handful of companies or businesses where I've done the initial piece of work, and now they're frozen because leadership doesn't understand. Right. Because what they did was just say, oh, train my staff or do some work with my staff, but because they were not involved in that process or didn't engage in the way that maybe could have been really beneficial, they're then not in a position to approve and help with the rollout beyond that. So a lot of the work I've done has gotten has stopped at the moment, and I think the reason is, you know, they hear uncle Bob at the barbecue on the weekend talk about how AI is breaching copyright or the geopolitical issues or all these issues that get are getting talked about outside of the workplace. And sometimes that seeps into the workplace. But you're absolutely right. Unless that leadership long term, unless that leadership is on board, Unless that leadership long term, unless that leadership is on board, they're never gonna approve what needs to be approved to have the technology integrated. So I think this year is a really funny year, and we're we're going to see those that just dive in. Although, you know, one of the other things that I'm seeing a lot of is we've had the companies or the businesses that wanna build their own models. They wanna build their own internal solutions.
And I actually think that those that wait a little bit longer, some of that stuff are gonna be really big investments that were not needed because of the tech coming down the line. And so in many places, I'm saying just sit tight for a bit and let's see where autonomous agents get to because maybe, you know, educate your staff now, have them understand what LLMs are, how to prompt, how to work with AI, how to decide what aspects of their role, you know, can be done by AI or assisted with or collaborating in collaboration with AI. So do that culture change in education piece, but maybe sit tight just for a little bit on the big investments in the tech. And it's the same with some companies only wanna use Copilot because they're in the Microsoft environment.
Right. And that's a place to start, but it's not a cheap place to start. You know, if you've got thousands of staff, Copilot licenses, that adds up. Now you're blowing $20 a month on licenses and Yeah. And maybe Copilot is not the best solution. Maybe there are some other solutions that are either available already or about to be. And so it's tricky for for senior management to be making those decisions at the moment. And as much as that's the work that I do, companies need to be pulling in the expertise to help them navigate that. And I think that that will and I think you said on on my podcast that that they've already got AI expertise, and absolutely, most companies do, but they don't necessarily have the strategic expertise on integrating AI across every task, every role, every department, and across the, you know, the whole of an organization. People have often got a real handle on what it means for their role or their team or their part of the organization, but maybe not necessarily a whole of organization approach.
[00:15:11] AI Narrator #1:
Alfonso here, and I want you to get more customers for your local business on autopilot with SMB Autopilot. Boost your business's visibility with our AI driven local marketing, and it's done for you so that you can focus on what you do best and leave the local marketing to us. Visit smbautopilot.ai to learn more.
[00:15:31] Matt Rouse:
Yeah. I think, kind of along those lines, one thing that I've seen that seems to be working fairly well in some organizations is kind of these, like, the equivalent of, like, a hackathon. You know, we're gonna get all the marketing staff together or the marketing and sales. Depends on the size of your company, obviously. You know? Maybe if you have three salespeople, one marketing person, and an admin, you get those five people together. And the idea is we're gonna basically list all of our all of our tasks that are kind of, like, almost like what's our job description. Right? Like, what are all the things that we have to do on a day to day basis that are not having conversations with people, building relationships, you know, stuff that your AI is not capable of?
You list down those tasks, and then you start going,
[00:16:23] Kate vanerVoort:
which of these things do I think an AI could possibly do, and how much time is it gonna save me? Right? Yeah. And that's actually the first step we do when consulting with companies is that AI business audit. So we have a process whereby every person in the organization lists or goes through, so tracks for a week what are the tasks that they're doing on a repeatable basis. And then we have a process whereby we can analyze all of that information across the organization. It's a great amount of data that we get when we do that process. And we can actually go back and say, well, look. And in fact, we did some work recently with a childcare franchise where we were able to save them forty hours a week across a team of 15 people by doing a business audit and showing them how they could do that. And so in one month, they were able to save that amount of time. And so data doesn't lie, and dollars are a big motivator often in terms of getting senior leadership on board. But
[00:17:25] Matt Rouse:
really, I'm just great is you collect all that data, and you can use the AI to sort it and analyze it for you. Absolutely.
[00:17:31] Kate vanerVoort:
Absolutely. It does tend to be a little we've refined our process because it does tend to be a little overzealous in what it thinks it can do.
[00:17:40] Matt Rouse:
Chatty chatty speak talks It's chatty. Yeah. You gotta you gotta tell it to calm down a little bit. You don't wanna use that, what's it called, high output SONNET 3.7 that just, like, gives you, like, pages and pages of of tokens. But Yeah. That's aside the point. So I think something that's really interesting there is that if you're a solo business owner, like or maybe you just have, like, a couple people in your business, the same process still holds true. Right? You can list all the things that you have to do repeatedly, that you have to do on a regular basis. Right? Like, okay. I gotta collect data from these five systems and put them all in a report and email it to, you know, Janet. So then you could look at those things. You can say, how much time am I spending on these things?
And you can say, okay. Well, you know, maybe I can use a combination of, you know, an automation system like Zapier or Pabbly Connect or Zoho Flow or whatever it is. And I can use it to pull the data from each of those systems that I'm collecting manually, and then I can have the AI collect all of that stuff and write the report for me and email it to Janet. So and then I can schedule it. And now I've just saved myself. You know, if that takes three hours every time you do it and you gotta do it every week, you know, you just save yourself twelve hours a month. And now you got twelve hours a month you can go out and find more business. Right? Or, you know, do whatever the more important task is. Yes. Sorry.
[00:19:12] Kate vanerVoort:
That process that we do with companies, we've actually turned into a custom GPT for our coaching program. So they do exactly the same process. They fill out an Excel spreadsheet of what are the tasks that they're doing on a repeatable basis. They give it to the custom GPT. The custom GPT makes a list of recommendations of where they should be looking at to save time and then gives them a road map of how to roll that out across their business. And in the four weeks of our coaching program, by the end of it, they've built a team of custom GPTs that address that initial business order that helps them to do that, kinda closes that that loop.
[00:19:51] Matt Rouse:
You know, this is kind of an aside, but I have not sold on Jeep custom GPTs. And Uh-huh. No. Tell you why. Let me tell you why. I use another program called Sym Theory, right, which is made in Australia, actually. Yep. And I can essentially do the same thing as a custom GBT, but I can put a whole bunch more kind of data and input into it. Mhmm. But the only downside is I can't give access to that to someone else. Right? Whereas chat like, custom GBTs, you could. And I'll tell you why I think I feel a little like I got burned on custom GBTs is that they were like, you're gonna make custom GBTs, and you we're gonna have a marketplace, and you could sell them. And I was like, I'm gonna sell the shit out of some custom GPTs. Yeah. And I built a whole bunch, and I figured out how they work. Then when they didn't marketplace, never appeared, and I was like, oh,
[00:20:42] Kate vanerVoort:
come on. You're not alone. A lot of people did exactly the same. But I'm still very sold on custom GPTs for two main reasons. One, it's in a workflow that most small business, small businesses feel comfortable using. So when you mentioned that platform, I went and had a look at it. I think the costs are a bit prohibitive for small businesses. And there's a a tech
[00:21:08] Matt Rouse:
learning tool that's right. It is a little bit of techy tool, but
[00:21:12] Kate vanerVoort:
the And for a lot of people, they wanna use things in the platforms that they're already using. And so most of the token cost to it. Right? That's where it it eats them. Yeah. But as soon as you start do that. Yeah. As soon as you start talking about APIs and it that then goes into a level that, to be really honest, the majority of small business owners are there. Nobody's gonna do that. And they're not able to do that. And so the custom GPTs, you know, we use them internally. We use them as lead magnets. Every workshop, every course, everything that we run externally in terms of having people in our membership or our community, we create a custom GPT that supports them with that process.
Right. So I have a free video on how to train your Flux LoRa, so how to create AI photo shoots. It's a pretty complicated process. Right? But I created a custom GPT that steps people through that process. So it's like they've got a little assistant, they can watch the video, but they've also got a little assistant where they're actually getting coached and supported on the side. And the other way I love using custom GPTs is, as you say, giving access to other people. Right. And VAs or contractors, I create custom GPTs for every task I wanna do, but then they can use those on my behalf and sound more like me and produce content that is, you know, in line with the brand voice, which is largely me.
[00:22:45] Matt Rouse:
Right. You know, I think if they connect up, like, the scheduling and, like, operator and stuff with custom GBTs, that that'll be a selling feature for me. I'll probably head back to it again.
[00:22:58] Kate vanerVoort:
Done a lot with operator. And when Manus came out, operator has been shelved, unfortunately, for $200 US a month. Yeah. I'm gonna be hard pressed to to,
[00:23:10] Matt Rouse:
yeah, renew. Manus seems pretty good. I mean, when it comes to that kind of stuff because we I mean, we're essentially a tech technical marketing company. We have developers on staff and stuff like that. Right? So, you know, if we're gonna build an automation, we're you know, even if it's something like and this you know, if you are somewhat development inclined, this is an easy way you can do some a lot of complex tasks is you can just boot up WordPress. Right? Or if you already have a WordPress website, use WS form, which has access to call APIs, and it also has webhooks.
And then you can connect that into Zapier, whatever your automation system is, do any of the processing you want, and send it back to WordPress. So you could actually make entire applications with that system if you wanted to without writing any code. Right?
[00:24:05] Kate vanerVoort:
So See, we're an education stuff there. And consulting business. We're not a tech tech implementation business. We partner with people like you when we when it comes to building the solutions. And and most of the people that we're working with, whether it's in organizations or companies or small businesses, not at the tech. So I kinda do that first piece before handing over to the tech implementation, which is that what does this mean for our business? How how are we going to create a culture here that allows for a transition into a new way of doing business? So we're really right at ground zero back at that step one before handing off for the tech implementation side of things.
[00:24:48] Matt Rouse:
I think something that's interesting is if you do have companies that are large enough that they have either IT staff or they have, like, a managed services company or something that they work with, there's definitely an opportunity there to get them to help you build solutions as well. Right? If you need something more complex, then, you know, maybe you could build yourself. That's it's not simple if you're not a developer. A lot of this stuff can be handled by your IT management people without going to an outside developer, which can save you a ton of money. Right? Like, building a piece of software now, even though it's a lot cheaper than it was before, I mean, it's still gonna be $6.06 6 to $10,000 for something reasonably simple, right, that just does, you know, data retrieval and analysis and stuff like that. Your IT people might be able to throw something together with what you have and some AI and and get that done too. So that's kind of a way for your company to sneak sneaks in my eye in the back door.
[00:25:46] Kate vanerVoort:
Sure. But we're still in that sticky taping things together phase. Right? And and we did that with social media in the early days, but this is moving a lot faster. So I remember when you couldn't, you couldn't schedule to like half the platforms.
[00:26:02] Matt Rouse:
You were like, oh, I can schedule this, but I can't schedule that. And I got a natively post on Instagram, but I can't do it on Twitter. And you know, like.
[00:26:10] Kate vanerVoort:
I remember when there were no images.
[00:26:12] Matt Rouse:
Oh, yeah. I know it's
[00:26:13] Kate vanerVoort:
So that makes me feel really old. Yeah. Me too. But I think for most companies, they'll have, like, those burning challenges, those really big issues, and you can just address that with a simple automation, and you're right. Your internal teams can often manage that. But I think we're undergoing a huge transition, and the companies that become AI first companies where it's really embedded into the culture of the organization, I think it's too early to be investing in big solutions around that. So Yeah. No. I think it's definitely in in building the the
[00:26:52] Matt Rouse:
the culture of using it rather than building your own systems.
[00:26:57] Kate vanerVoort:
But if you say to every person in the organization, imagine if you had a Kate two point o or a Matt two point o. You had a little assistant that was sitting next to you. What are the things that you would hand over to that that AI? And that's really the best place to start within the guardrails or parameters that the business sets, but that's when people can start to see the tangible benefit of that. And Right. Then if you are really and but doing that after doing an audit so that you're actually getting the economy of scale, that if everybody email is something that everybody does, is there some layer that we can put in where AI supports that across the organization, not just at the individual level? And I think that's why a lot of companies or businesses are frozen at the moment. There's so many ways that you can come at the AI conversation, And it's just about finding what works naturally for each business. I wish I could bottle it. I wish I had a system that works for every business, but it's different for every business. And tech comfort with technology, because it's often been quite a siloed part of a business, that can be a really big barrier. If you've got people who are really resistant to AI, there's a there's some work to be done before they'll even consider using AI. So
[00:28:16] Matt Rouse:
Well, there's another consideration that you know, I was just talking with somebody last week about, which which is kind of the idea of this siloed AI technology. Like, let's say my company is is a % Google Workspace shop. Right? Everything we do is on Google Workspace. The only AI anybody uses in our company is Gemini. And then Gemini updates their system, and now it doesn't do the thing that we used to use it for. Right? Well, now you're stuck. Right? But if you do have a culture in your company of testing and and messing around with some AI systems to kind of kind of get used to how they work and not just how one specific program works like Copilot or something, right, then you've got the ability to pivot. Right? If you if the AI system changes, something better comes out, whatever it is, you know, then you can go back with your teams and say, you know, oh, let's research this. Let's figure out, you know, how to use this other better system.
And, you know, one other shout out I wanted to have, you know, Kate, for using, you know, what is essentially consulting firm, right, like yourself, is that audit piece is super important if you have regulation in your industry.
[00:29:32] Kate vanerVoort:
Absolutely.
[00:29:33] Matt Rouse:
Like in Canada, if you are a medical related company, you have to host all your patient data in Canada, which means you can't send it to OpenAI with a question about a patient or something, right, Unless you anonymize it first. And so these are things that a consultant is gonna know that your business may not have considered.
[00:29:54] Kate vanerVoort:
And Australia and Canada are quite similar in terms of how legislated we are. And I was actually gonna say that when we were speaking earlier around. And just to your point there actually around most companies wanna use AI within the ecosystem that they're already using. So if that is Microsoft or if that is the Google Workspace, they wanna use Gemini or Copilot. The problem with that or sorry. The the benefit of that is that it can unlock their data in what is a secure environment that they're already familiar with. They've already got their files in the cloud. They already have addressed their data security issues. But both of those models are fairly limited in what you can actually create using AI in those models.
And so that's where you look at, okay, what other tools might be available, but then you've got all of the issues around data privacy, where's the data stored. A lot of companies do not wanna use platforms out of China even though they're probably using platforms out of China and maybe just don't realize it. Right. And so that's where it because most of the demonstrations that I use when I'm in a boardroom are not within Copilot or within Workspace because within Google because they're not the wow. Oh my goodness. Imagine if we could do that. Imagine if I could build that 20 slide slide deck in three minutes instead of three days.
And so, yeah, we're we're just still in this funny place where there's a ton of different ways that you can do AI, but most wanna start in that closed environment. But they're gonna be pretty limited in what they're able to do within that environment to start with. Yeah. And I I think,
[00:31:36] Matt Rouse:
you know, the the AI scaling law right now is averaging seven months for the doubling of of AI power. If you compare that with, like, Moore's law, which is eighteen months for the, you know, doubling of of the processing power of microprocessors, so it's essentially twice as fast of, you know, the the doubling principle. And, like, I'm been here three months, just in the last couple of weeks. Well, it's it's averaged out to seven. Right? But sometimes it's three, sometimes it's 10. Right? You know, like But averaged over what? Average. Well, it's been averaged. They've kind of tracked back since 2023. But, also, they can track back before that to when the transformer was invented, right, which I believe was back in 02/2012.
So that was a DeepMind invention originally. So Yeah. But, like, what I was gonna say was, like, I was doing my introduction to AI presentation. I finished it at 04:00 the night before the presentation. And at 5PM, OpenAI announced native images in their system where you could do image editing and all this stuff. And I had a slide where I was like, don't use OpenAI for images because Dolly three sucks. Right? And so I gotta go back and change, you know, two or three slides of my presentation to talk about this new thing, and then we gotta make a demo for it, and we gotta set this all up. And that's how quick it moves. You know? I I barely get a chance to do a presentation that I don't have to redo the same day of the time I'm presenting it because it moves so quickly.
Happens to the whole time. That's right. But the other thing is that I mean, there's it's a two two sided coin. Right? One side is it's getting better so quick that something you can't really do now or it doesn't work great might work great six months from now or three months from now or let next week maybe. It's nice. Yep. And the other side of that is it's hard to keep up. So I think it definitely helps having somebody on your team, you know, somebody who can help you, whether that's vendor, consultant, whatever. You know? Like, Kate's perfect example. Kate, if somebody wants to reach out to you to get more information or an audit or something, how would they do that? Yeah. So probably the best place to connect with me is on LinkedIn under Kate Vandevoort.
[00:33:54] Kate vanerVoort:
And we also have the AI Success Lab Facebook group where we share a lot of the breaking news, but real strategies on how you can use AI for business today that moves the dial in your business. So they're probably the best two places to connect. AI. Perfect. You can also connect with me on LinkedIn if you wanna hear about chickens because that's mostly what I write about. Chickens and AI. Great to see everyone again out, and, Kate,
[00:34:18] Matt Rouse:
so excited to have you on my show. It was a great great talk on the AI Grapple podcast as well, which everybody should go listen to, not just because I was on it. There's other great guests as well. But, Kate, thanks so much for being on the show today.
[00:34:31] Kate vanerVoort:
Thanks, Matt.
[00:34:34] Matt Rouse:
It's been fun.
