WEBVTT
00:00:01.919 --> 00:00:05.280
Companies will spend millions of dollars on AI.
00:00:05.440 --> 00:00:13.199
I mean, they'll buy thousands of enterprise seats for Microsoft 365 Copilot or ChatGPT Enterprise or whatever it is.
00:00:13.519 --> 00:00:22.000
CEOs will stuff as many AI-related buzzwords as they can at their weekly all-hands meeting to make it seem like they're on the cutting edge.
00:00:22.399 --> 00:00:24.239
But they're skipping something.
00:00:24.399 --> 00:00:32.640
The most basic thing: teaching their employees how to actually use whatever large language model they're providing.
00:00:33.200 --> 00:00:40.560
I've seen a sizable gap in 2026 starting to form in corporate America between the have and the have nots.
00:00:40.960 --> 00:00:45.439
I'm talking about the have trained their employees and the have not trained their employees.
00:00:45.520 --> 00:00:52.000
And if I'm being honest, I would say like 90% or more of companies fall in the latter group.
00:00:52.240 --> 00:00:55.439
Your company would probably fall there too.
00:00:57.439 --> 00:00:58.399
But why?
00:00:59.119 --> 00:01:01.200
Well, don't worry.
00:01:01.520 --> 00:01:10.719
Even if you are in that ladder group of, yeah, my company hasn't trained uh us, or if you're the CEO, we haven't trained our people on AI.
00:01:10.799 --> 00:01:13.680
I'm gonna lay that out for you on today's show.
00:01:13.840 --> 00:01:25.359
As part of our Start Here series, volume six, we're going over how to train your team on AI, the seven steps to educate your organization on using large language models.
00:01:25.519 --> 00:01:25.840
All right.
00:01:25.920 --> 00:01:27.439
I hope you're excited for this one.
00:01:27.599 --> 00:01:28.319
I am too.
00:01:28.480 --> 00:01:30.000
If you're new here, great.
00:01:30.159 --> 00:01:32.000
This is a great place to start.
00:01:32.239 --> 00:01:41.599
After doing this everyday AI thing for uh three years and more than 700 episodes, one of the most common questions is where do I start?
00:01:41.760 --> 00:01:42.879
You have so many podcasts.
00:01:42.959 --> 00:01:45.359
That's why we started the Start Here series.
00:01:45.439 --> 00:01:49.519
Uh so if you haven't already, please go to starthirseries.com.
00:01:49.680 --> 00:01:53.840
That'll give you free access uh to our inner circle community.
00:01:53.920 --> 00:01:58.239
And then in that start here series space, you can go and listen to every single podcast.
00:01:58.319 --> 00:01:59.359
You don't got to look around.
00:01:59.439 --> 00:02:02.159
Uh, you know, if you want to watch the video version, it's all there.
00:02:02.400 --> 00:02:11.520
But this is the essential podcast series to both learn the AI basics or to double down on your AI knowledge.
00:02:11.840 --> 00:02:17.520
So uh last week in volume five, so you can go listen to that episode 703.
00:02:17.599 --> 00:02:28.560
We talked about AI hallucinations, what they are, why they happen, and the right way to reduce the risk, which is a great transition into actually training your team on AI.
00:02:28.719 --> 00:02:29.199
Uh right.
00:02:29.360 --> 00:02:31.360
Notice I didn't start with that on volume one, right?
00:02:31.439 --> 00:02:32.479
We had to go over the basics.
00:02:32.639 --> 00:02:35.039
We had to, you know, understand all the jargon.
00:02:35.120 --> 00:02:41.120
We had to really tap into the AI mindset and, you know, talk about hallucinations and some of these other things.
00:02:41.280 --> 00:02:43.599
But uh let's get straight into it.
00:02:43.840 --> 00:02:52.719
And this is gonna be a quicker episode because I don't want it to get drowned out in the normal, you know, sometimes fluff that I might bring.
00:02:52.879 --> 00:02:55.520
But the gap that I'm talking about is alarming.
00:02:55.680 --> 00:03:04.080
So according to a McKinsey study, 92% of companies plan higher AI investment, but only 1% call their deployments mature.
00:03:05.199 --> 00:03:09.520
Yeah, I wasn't making it up when I said the gap is sizable.
00:03:09.599 --> 00:03:15.680
So that's like every single executive saying, yeah, AI is the most important thing and we're doubling down.
00:03:15.840 --> 00:03:21.120
Yet only 1% are saying, yeah, we are actually getting it right across the board.
00:03:21.360 --> 00:03:30.639
And uh another uh study from Forrester said that only 30% of large enterprises will even mandate AI training by the end of this year.
00:03:30.879 --> 00:03:34.240
This wasn't by the end of 2023 or 2024.
00:03:34.479 --> 00:03:45.280
By the end of the year 2026, only less than a third of organizations are even going to require AI training, which is absolutely bonkers, right?
00:03:45.680 --> 00:03:55.599
To think that just about every single organization in America is saying that yes, using AI and deploying AI is the most one of the most important things that we're trying to do as a company.
00:03:55.759 --> 00:04:02.719
We'll spend any money, we'll throw all the buzzwords, but we're not even gonna learn to understand it, or we're not even gonna take the time to learn to understand this thing.
00:04:02.800 --> 00:04:04.960
So here is the seven steps you need to go through.
00:04:05.120 --> 00:04:10.960
Step one, and I'm gonna break these, uh, I'm gonna break them down, but I'm gonna give them to you all here up front, not gonna make you wait to the end.
00:04:11.199 --> 00:04:13.360
Step one, leadership must go first.
00:04:13.520 --> 00:04:17.199
Step two, you've got to fix broken workflows before adding AI.
00:04:17.360 --> 00:04:20.240
Step three, you need to pick one AI platform and commit.
00:04:20.399 --> 00:04:24.480
Step four, you need to train in three layers, all right, not upskilling.
00:04:24.560 --> 00:04:26.399
We are rebuilding and unlearning.
00:04:26.560 --> 00:04:30.079
Step five, you need to document your procedures, not just your data.
00:04:30.319 --> 00:04:34.639
Step six, you need to mandate hands-on practice with real outputs.
00:04:34.800 --> 00:04:40.000
And step seven, you've got to go from operator to orchestrator, all right.
00:04:40.079 --> 00:04:44.079
Uh, and probably step zero would be take our free prime prompt polish course.
00:04:44.160 --> 00:04:44.399
All right.
00:04:44.480 --> 00:04:54.959
So if you do go to the starthearseries.com, uh, that will give you free access to our inner circle community, but you can also go take our free prime prompt polish course, uh, self-paced.
00:04:55.040 --> 00:04:57.759
It's about an hour and a half, two hours, right?
00:04:57.839 --> 00:05:00.079
More than 15,000 people have taken it.
00:05:00.160 --> 00:05:01.680
So yeah, go take that first.
00:05:01.759 --> 00:05:03.839
That's step zero, and then dive into step one.
00:05:03.920 --> 00:05:08.319
All right, so let's talk a little bit more about step one and why AI leadership must go first.
00:05:08.560 --> 00:05:12.480
And I don't want you to take this the wrong way.
00:05:12.720 --> 00:05:17.279
I'm not saying AI needs to be a top-down mandate because if it is, it will fail.
00:05:17.439 --> 00:05:22.000
When I say leadership must go first, I'm saying the CEO must use it daily.
00:05:22.079 --> 00:05:22.480
Uh, right.
00:05:22.560 --> 00:05:30.560
I had a great conversation a couple of years ago um with WWT's um CEO, Jim Kavanaugh, right?
00:05:30.800 --> 00:05:33.839
CEO of a$20 plus billion dollar revenue company.
00:05:33.920 --> 00:05:42.879
Um, and you know, we were chatting both, you know, on camera and before and after about how he's using AI every single day and has been for years, right?
00:05:43.040 --> 00:05:47.120
And even when employees come to him, you know, he's like, hey, is this AI native?
00:05:47.360 --> 00:05:47.680
Right?
00:05:47.920 --> 00:05:51.600
Like, don't don't bring anything to him first if it's not AI native.
00:05:51.759 --> 00:05:52.879
It's not AI first.
00:05:53.040 --> 00:05:59.920
And I think that's a great um leadership, leadership example of how you need to be, right?
00:06:00.000 --> 00:06:06.079
So if you are the CEO of a company or if you're speaking with the CEO of the company, your CEO can't be saying, yeah, go, go use AI.
00:06:06.160 --> 00:06:08.000
And they're still doing it the old way, right?
00:06:08.079 --> 00:06:10.000
This is change management 101.
00:06:10.079 --> 00:06:16.160
Uh, and you can't just, you know, have a couple, you know, key stakeholders, a couple AI champions, and expect this thing to work.
00:06:16.319 --> 00:06:20.560
It needs to start with leadership at the top using it.
00:06:20.639 --> 00:06:22.000
Again, not top-down mandate.
00:06:22.079 --> 00:06:23.519
You need to get everyone's, right?
00:06:23.600 --> 00:06:31.439
We'll probably do another episode in the Start Here series about, you know, proper ways to build AI uh strategies, et cetera, right?
00:06:31.600 --> 00:06:38.720
But you need to create those cultural moments, the wins of the week, the all hands demos, internal spotlights, and we're gonna get to more on that later.
00:06:38.879 --> 00:06:44.800
But if your top people are not actually using it, it's not actually going to work.
00:06:45.600 --> 00:06:50.160
Step two, fix broken workflows before adding AI, right?
00:06:50.319 --> 00:06:51.360
I called this before.
00:06:51.439 --> 00:06:55.839
You can throw uh, you know, makeup on an ugly pig, it's still an ugly pig.
00:06:56.000 --> 00:06:56.399
All right.
00:06:56.560 --> 00:07:05.279
If you think that adding AI to an antiquated, uh, you know, on tech workflow is going to help, it's probably not.
00:07:05.519 --> 00:07:09.040
In many cases, it could make things much worse, right?
00:07:09.279 --> 00:07:21.279
I can't tell you the number of use cases that I've come across personally, you know, when we consult companies, uh, you know, just stories, just chatting with people, you know, saying, like, hey, we try to implement AI in this process and it didn't work.
00:07:21.360 --> 00:07:23.360
And I'll be like, okay, tell me about this process.
00:07:23.519 --> 00:07:24.319
And they'll tell me about it.
00:07:24.399 --> 00:07:27.120
And I'm like, this process is completely broken, right?
00:07:27.439 --> 00:07:30.000
AI or not, this process isn't working, right?
00:07:30.079 --> 00:07:38.720
Because a lot of times, you you know, these processes, they're they're they're duct taped, they're uh, you know, just randomly glued together, they're MacGyver together.
00:07:38.879 --> 00:07:40.560
And it's like, okay, well, where did you get this?
00:07:41.040 --> 00:07:45.360
Okay, well, you know, this is you know, Jane in marketing, she she did this piece this way.
00:07:45.519 --> 00:07:48.720
Um, you know, when Bob and IT, I got this from him.
00:07:48.800 --> 00:07:50.240
And no, right?
00:07:50.959 --> 00:07:55.759
Your your standard workflows, if they don't make sense without AI, right?
00:07:56.079 --> 00:07:59.920
You're like, oh, this is uh uh the right way to do this this task.
00:08:00.560 --> 00:08:02.079
AI is not gonna help, right?
00:08:02.160 --> 00:08:07.199
People think AI, and this is again not to pound home on this one, right?
00:08:07.279 --> 00:08:11.040
But people think AI will fix broken processes, right?
00:08:11.120 --> 00:08:14.879
And it's not like I said, it'll make things worse.
00:08:15.120 --> 00:08:22.000
You need to redesign the entire workflow by being AI first, but you also need to be able to measure it as well.
00:08:23.439 --> 00:08:28.879
All right, all right, step three pick one AI platform and commit.
00:08:29.040 --> 00:08:34.240
All right, so we went into this one in depth on volume three of the start here series.
00:08:34.320 --> 00:08:35.919
That's episode 695.
00:08:36.399 --> 00:08:42.559
And I'm gonna spend a little bit more time on this one uh because I think it's important when I say pick one and commit, okay?
00:08:43.679 --> 00:08:52.799
Because I think a lot of organizations are Microsoft Copilot 365 organizations, and they still might need Chat GPT Enterprise, right?
00:08:52.960 --> 00:09:02.399
So I like to say, you know, Microsoft 365 Copilot, it is both the the worst and the best, right?
00:09:02.639 --> 00:09:14.559
Um, it's the worst because it can be extremely hard for organizations to understand where they use Copilot, how to enable it, permissions, right?
00:09:14.720 --> 00:09:19.440
It can it can be a nightmare if you don't have your ducks in order.
00:09:19.519 --> 00:09:22.639
It's kind of similar to your your your workflows, right?
00:09:22.799 --> 00:09:31.600
If your organization is not in order from an IT from a from a Microsoft Windows perspective, Copilot's probably not going to work very well.
00:09:31.679 --> 00:09:39.919
And a lot of those organizations realize that early on, and they're like, okay, you know, cope, it's it's almost like too much, right?
00:09:40.080 --> 00:09:42.960
In some instances, it can be too much.
00:09:43.120 --> 00:09:46.559
And in those instances, yeah, you might have to pick one of the others, right?
00:09:46.639 --> 00:09:58.399
But for the most part, I say your online operating systems, your AI operating systems, it's Anthropics Claude, it's Google Gemini, it's open AI's Chat GPT.
00:09:58.559 --> 00:10:07.360
And amongst those three, you should probably to get started with, you should be choosing one and moving most of your critical day-to-day processes in there.
00:10:07.519 --> 00:10:10.720
No matter what your kind of uh AI go-to-market is, right?
00:10:10.799 --> 00:10:16.080
Whether you're starting with the champion team, whether you're starting with certain departments, whether you're doing the entire company, right?
00:10:16.159 --> 00:10:17.360
I can't say what's the best.
00:10:17.440 --> 00:10:23.200
It just depends on the size of your company, how much training you have, what your processes look like, et cetera, right?
00:10:23.279 --> 00:10:23.519
Are you?
00:10:23.679 --> 00:10:26.720
Do you have 10,000 employees or do you have 50?
00:10:26.799 --> 00:10:27.440
Um, right.
00:10:27.600 --> 00:10:34.080
But for the most part, you should all be learning and sandboxing in the same AI operating system, one of those three, right?
00:10:34.159 --> 00:10:42.399
And then eventually, yeah, maybe you're uh, you know, if you have a software engineering department, you know, maybe they might end up using Claude or something like that.
00:10:42.559 --> 00:10:53.120
Maybe your creative team might end up using Gemini, but you should all be learning and bringing the processes in and measuring ROI and finding your use cases and scaling in a single operating system.
00:10:53.279 --> 00:11:00.639
Because what happens when you start using, oh, we're gonna use ChatGPT for this, use Gemini for this, use Claude for this, use Copilot for this.
00:11:00.799 --> 00:11:07.840
All of a sudden, then you're gonna find it very easy to say, okay, well, then we're gonna use these 10 other tools as well.
00:11:08.080 --> 00:11:11.679
And that 10 other tools becomes 30, becomes 50, right?
00:11:11.840 --> 00:11:26.720
A lot of uh did a lot of consulting calls in 2025, and so many of them, so many of them, I was blown away, right, by the number of AI tools, especially medium-sized businesses were using, right?
00:11:26.799 --> 00:11:40.399
It's it's it's like some companies had, you know, 10 different AI tools that they were using just for writing different kinds of copy, literally 10 different tools, not 10 different GPTs, 10 different projects, 10 different gems within one system.
00:11:40.559 --> 00:11:43.759
10 different tools that they were using for writing, right?
00:11:44.159 --> 00:11:49.840
I think that not focusing on a single AI operating system leads to that shiny AI object syndrome.
00:11:49.919 --> 00:11:50.720
So pick one.
00:11:50.879 --> 00:11:54.159
And then also, you gotta pick one.
00:11:54.399 --> 00:11:55.039
All right.
00:11:55.600 --> 00:12:00.639
Studies have shown, right, whether you want to call it shadow IT, AI sprawl.
00:12:00.799 --> 00:12:04.240
I've been calling it second computer AI since 2023.
00:12:04.320 --> 00:12:07.519
Uh, you know, I called it happening back then, and it is prominent.
00:12:07.759 --> 00:12:10.559
Even if you think that your people are not using, right?
00:12:10.720 --> 00:12:17.200
Your team, if you're a department head, if you're the CEO, whatever, if you're an HR, everyone's using AI.
00:12:18.399 --> 00:12:19.519
Okay, they are.
00:12:19.679 --> 00:12:27.519
Whether they're using it once a week or for every single project, whether you've greenlit it or not, people are using AI.
00:12:27.600 --> 00:12:33.120
A lot of times on a second computer on a different device, uh, right, incognito window, whatever it may be.
00:12:33.519 --> 00:12:42.960
So if you have some hesitation about bringing your data, bringing your processes into a large language model, they're already there, right?
00:12:43.120 --> 00:12:44.399
A lot of people don't know.
00:12:44.639 --> 00:12:47.600
Everything on your website, it's already in large language models.
00:12:47.679 --> 00:12:48.320
People don't know, right?
00:12:48.399 --> 00:12:51.360
They they upload PDFs or these documents from five years ago.
00:12:51.519 --> 00:12:53.840
And it's like, yeah, that's that's in a large language model.
00:12:53.919 --> 00:13:00.080
That's I can access it by looking at a at a site map uh on your on your website.
00:13:00.240 --> 00:13:05.919
People don't understand the amount of data that is already available about your company on social media, etc.
00:13:06.320 --> 00:13:11.200
Everything, almost everything from a public front-facing perspective is already in large language models.
00:13:11.279 --> 00:13:14.559
And then when you think about, oh, my company's data, blah, blah, blah, it's so special.
00:13:14.639 --> 00:13:15.519
No, it's not, right?
00:13:15.600 --> 00:13:16.320
What do you use?
00:13:16.399 --> 00:13:18.159
You use a cloud provider, right?
00:13:18.320 --> 00:13:18.879
Guess what?
00:13:19.039 --> 00:13:22.879
Probably the AI operating system to start with is whatever cloud provider you use.
00:13:23.039 --> 00:13:29.679
So if you use, you know, Microsoft Azure, SharePoint, OneDrive, whatever, well, probably start with Microsoft 365 Copilot.
00:13:29.840 --> 00:13:32.480
If you're using Google, well, you should probably start with Google, you know.
00:13:32.720 --> 00:13:33.919
AWS, I don't know.
00:13:34.000 --> 00:13:36.720
You can roll the dice and and try the Nova, right?
00:13:36.799 --> 00:13:38.960
I wouldn't personally recommend it, but you can try that.
00:13:39.039 --> 00:13:46.240
Uh, but you have to stop tool sprawl early because once it sprawls, it's too much to handle.
00:13:46.559 --> 00:13:50.480
So get your AI platform and commit to it.
00:13:50.799 --> 00:13:53.759
Step four, train in three layers.
00:13:54.000 --> 00:13:54.639
All right.
00:13:55.600 --> 00:14:02.399
This is where I might rub a lot of the people the wrong way that came up with their AI slogans uh in 2024 and 2025.
00:14:02.559 --> 00:14:04.159
People at big tech companies, right?
00:14:04.240 --> 00:14:07.600
Everyone's you know, sprinkling the word upskill and reskill.
00:14:07.759 --> 00:14:09.200
That's a great way to fail.
00:14:09.440 --> 00:14:16.000
Um AI, large language models are smarter than us, right?
00:14:16.879 --> 00:14:18.960
Um, I was a Pulitzer fellow.
00:14:19.120 --> 00:14:23.519
I won all these writing awards back in the day, ACP story of the year, right?
00:14:23.600 --> 00:14:24.799
All these things.
00:14:25.440 --> 00:14:27.440
AI is way better at writing than me.
00:14:27.679 --> 00:14:28.000
Right?
00:14:28.159 --> 00:14:31.200
If you're looking at the normal work slot that's out there, absolutely not.
00:14:31.279 --> 00:14:36.159
But I know how to use AI, I can make it write better than 99.9% of writers.
00:14:36.240 --> 00:14:41.840
And I can say that because I've been getting paid to write for 25 years and I've been writing at a very high level, right?
00:14:42.000 --> 00:14:47.679
Same thing with anything financial analysis, um, marketing, comms.
00:14:50.159 --> 00:14:52.480
Like pick your department, HR.
00:14:53.519 --> 00:14:54.320
Doesn't matter.
00:14:54.879 --> 00:14:59.840
AI models, if you're using them the right way, they're they're better than us and they're smarter than us.
00:15:00.000 --> 00:15:15.200
So to think that we can just uh sprinkle in AI at the top and the side here and there with upskilling, reskilling, no, no, you have to rebuild because we should probably be um taking a complete shift in how we work, right?
00:15:15.679 --> 00:15:17.360
Complete mindset shift, right?
00:15:17.919 --> 00:15:22.639
We are orchestrating but not pushing buttons, right?
00:15:22.799 --> 00:15:28.320
Because these models, for the most part, they're getting things done 10 to 100 times faster.
00:15:28.559 --> 00:15:38.559
So kind of the the role, I think, of the average domain-specific knowledge worker over the last next five to 10 years is going to change, change drastically.
00:15:38.799 --> 00:15:42.240
How me and you are working in 10 years, right?
00:15:42.399 --> 00:15:51.759
If you're if if you work a traditional desk job and you get paid for what you know in your brain, in five to 10 years, sorry, it's not gonna be like that anymore, right?
00:15:51.919 --> 00:15:56.559
It might be like two to three years, but in five to 10 years, it's not gonna be like that.
00:15:56.639 --> 00:16:00.639
So you can't just think of upskilling and reskilling, you have to unlearn and rebuild.
00:16:00.799 --> 00:16:01.120
All right.
00:16:01.200 --> 00:16:05.279
So what that means, layer one, you have to get literacy, you have to understand.
00:16:05.440 --> 00:16:09.039
So go listen to the first couple volumes of this series and you'll be there.
00:16:09.200 --> 00:16:15.039
Uh, layer two of this uh training process, it's domain specific by role.
00:16:15.279 --> 00:16:15.600
Okay.
00:16:15.919 --> 00:16:20.240
Um, you can't just have one training for your entire organization, right?
00:16:20.480 --> 00:16:27.279
You should be breaking it down and saying, hey, here's how uh as a, we'll just use marketing, right?
00:16:27.440 --> 00:16:38.799
As a marketer, here's how we use our company's data and produce better marketing collateral with our AI operating system, right?
00:16:38.960 --> 00:16:42.000
You do need AI literacy that works organization-wide.
00:16:42.080 --> 00:16:48.320
And that's why, you know, something like what we do with the chat uh with the Prime Prompt Polish, or, you know, we do that for businesses.
00:16:48.480 --> 00:16:50.480
I think that's great for everyone across the board.
00:16:50.639 --> 00:16:55.840
But then after that, you have to get into domain-specific uh training by role or by department type.
00:16:55.919 --> 00:16:59.840
And then number three, you have to understand data and procedures.
00:17:00.080 --> 00:17:08.720
Bad data has plagued everyone for decades before AI, and it's going to plague you and your company even worse.
00:17:08.880 --> 00:17:16.960
So if you thought that somehow AI could be a band-aid or makeup on your data problem, no, it is only going to expose it and make it worse.
00:17:17.119 --> 00:17:19.920
All right, because I've been, I've been saying this for a long time.
00:17:20.079 --> 00:17:29.680
Your good habits over the last 10, 15, 20, 30 years, or your companies or your departments are not going to matter anymore when we're talking about AI native workplaces.
00:17:29.920 --> 00:17:34.720
Step five, document your procedures, not just your data.
00:17:34.880 --> 00:17:35.279
All right.
00:17:35.440 --> 00:17:36.160
That's huge.
00:17:36.319 --> 00:17:40.400
And when I say document your procedures, all right.
00:17:40.799 --> 00:17:42.720
And I'm sorry if you're a longtime listener.
00:17:42.799 --> 00:17:56.480
Uh, you know, if you've listened to all 700 and some episodes, you know, some of these things I'm gonna be repeating myself, but documenting your processes is about collecting that human intelligence, right?
00:17:56.880 --> 00:18:00.160
Think of that your coworkers that have been there for a very long time.
00:18:00.240 --> 00:18:04.240
And you're like, man, if this person ever left, we'd be screwed, right?
00:18:04.480 --> 00:18:06.880
Once Deborah's gone, we're done.
00:18:07.119 --> 00:18:08.640
We're done without Deborah, right?
00:18:08.799 --> 00:18:10.720
Okay, we'll start documenting what Deborah does.
00:18:10.880 --> 00:18:13.279
Start documenting what Deborah knows, right?
00:18:13.440 --> 00:18:23.119
Start seeing how AI and and Deborah can better coexist, uh, right, because that is, I think, the last mile problem of AI implementation.
00:18:23.279 --> 00:18:34.640
Uh, so much of the early uh, you know, connecting your data um with large language models has been focused on, you know, some version of retrieval uh retrievalog meta generation or RAG, uh, right?
00:18:34.960 --> 00:18:52.799
Um and now these AI operating systems have a version of that very simple to bring your data in one click from whatever cloud storage, but you have to start getting and collecting and curating and cleaning different types of data that we historically have never needed as companies, right?
00:18:52.960 --> 00:19:02.240
Which is how our smartest people think, how they solve problems, you know, their own internal uh decision tree that's not structured data, right?
00:19:02.400 --> 00:19:04.960
I think of like flow charts and decision trees, right?
00:19:05.039 --> 00:19:08.799
So even if you're not a big, you know, traditional machine learning, right?
00:19:09.200 --> 00:19:18.079
Traditional artificial intelligence, machine learning, you can think of one type as like a decision tree or like a if this then or if if else uh tree, right?
00:19:18.319 --> 00:19:22.079
So it's kind of like that, but with unstructured data.
00:19:22.240 --> 00:19:23.759
That's great for structured data, right?
00:19:23.839 --> 00:19:25.359
For things that you can put in spreadsheet.
00:19:25.519 --> 00:19:28.720
If a value is more than five, this happens, right?
00:19:28.880 --> 00:19:33.920
If we have more than 30 of these in stock, then this email gets sent out, right?
00:19:34.160 --> 00:19:37.599
But what about those things that aren't quantifiable in a spreadsheet?
00:19:37.920 --> 00:19:40.960
That's what you have to start documenting in step five.
00:19:41.119 --> 00:19:43.839
You have to document your procedures, not just your data.
00:19:44.000 --> 00:19:47.039
That is the last mile problem of AI implementation.
00:19:47.200 --> 00:19:54.240
Because right now, let's be honest, for the most part, if you know what you're doing, right?
00:19:54.319 --> 00:20:00.160
If you're using the right AI model and you're using the right mode and you're connecting your data, that's tablescapes now, right?
00:20:00.400 --> 00:20:01.519
Everyone's doing that.
00:20:01.680 --> 00:20:05.920
This is what's gonna, at least for 2026, maybe 2027, I think.
00:20:06.160 --> 00:20:18.880
This is what's gonna help you, your department, your company create separation if you start documenting your procedures, your company's IP, your department's special sauce, Deborah's brain.
00:20:19.039 --> 00:20:20.720
You've got to start collecting it all.
00:20:21.279 --> 00:20:26.319
Step six, mandate hands-on practice with real outputs, right?
00:20:27.200 --> 00:20:32.240
The simplest way I tell people do do Friday lunch, Friday lunch days, right?
00:20:32.480 --> 00:20:35.920
Block out 90 minutes, have a good lunch, get people together, right?
00:20:36.000 --> 00:20:37.839
Especially if you're a remote organization.
00:20:38.079 --> 00:20:40.960
This is your AI session, right?
00:20:41.599 --> 00:20:45.920
But employees need to have hands-on keyboard, right?
00:20:46.160 --> 00:20:47.200
Not just on their phone.
00:20:47.359 --> 00:20:49.039
No, hands-on keyboard.
00:20:49.279 --> 00:20:51.359
You need to be sharing, you need to be workshopping.
00:20:51.519 --> 00:20:54.319
This is like, you know, mini hackathon styles.
00:20:54.480 --> 00:20:57.759
Uh, you know, yeah, you can do a hackathon in 30 minutes, right?
00:20:57.839 --> 00:21:02.319
You go hack one of your department's biggest problems, and everyone go take their computer.
00:21:02.480 --> 00:21:04.880
You know, everyone takes 10, 15, 20 minutes.
00:21:04.960 --> 00:21:07.440
Someone gets up, they say, they say, this is the problem.
00:21:07.599 --> 00:21:09.839
Here's how I've been trying to solve this with AI.
00:21:10.000 --> 00:21:11.759
Here's what's working, here's what's not.
00:21:11.920 --> 00:21:21.200
Everyone takes 15 to 30, 15 to 30 minutes to try and go build their version, take what Bill uh in IT made, made better, try something else completely different, right?
00:21:21.359 --> 00:21:25.440
You have to be doing this continually because AI is constantly changing.
00:21:25.759 --> 00:21:36.480
Y'all, my only job, my only job is to talk to smart AI people every day, share that with you, read about read about AI, test it.
00:21:36.559 --> 00:21:38.160
That's all I do every single day.
00:21:38.880 --> 00:21:41.680
Eight to 14 hours, right?
00:21:42.160 --> 00:21:42.720
Depends.
00:21:42.960 --> 00:21:46.079
Um, but I can't keep up, right?
00:21:46.559 --> 00:21:47.519
This is my job.
00:21:47.920 --> 00:21:50.640
So you can't keep up, your department can't keep up.
00:21:50.799 --> 00:21:53.839
No one can keep up unless you are intentional about it.
00:21:54.079 --> 00:21:54.480
All right.
00:21:54.720 --> 00:21:59.119
You have to have measured adoption with outputs, not usage rates, right?
00:21:59.200 --> 00:22:05.039
People look at You know, oh, I look in ChatGPT Enterprise and you know our utilization went up from 10 to 15.
00:22:05.359 --> 00:22:06.000
So that's good.
00:22:06.240 --> 00:22:09.039
No, no, you need to look at outputs.
00:22:09.200 --> 00:22:16.799
You need to look at what people are actually producing, and you need to give the space to learn and you need to be able to measure what's good and what's not, right?
00:22:16.960 --> 00:22:21.519
And not just, okay, these 10 people, you know, everyone go show me what you created.
00:22:21.680 --> 00:22:23.759
No, now let's see why it didn't work.
00:22:23.920 --> 00:22:29.200
If it didn't work, or if it did work, let's go and let's go ahead and look at that chain of thought and see why it did work.
00:22:29.440 --> 00:22:31.599
All right, step seven, this one's a big one.
00:22:31.759 --> 00:22:33.279
Another mindset shift.
00:22:33.359 --> 00:22:40.240
So many of these things, mindset, um, process, behavioral, but shifting from operator to orchestrator.
00:22:40.640 --> 00:22:41.039
Right.
00:22:41.279 --> 00:22:51.440
So much, I think, of what successful people in AI did uh between 2022 when ChatGPT came out until 2024, 2025, right?
00:22:51.599 --> 00:22:54.000
We were the we were the glue, right?
00:22:54.319 --> 00:22:57.119
A lot of copying and pasting, chaining things together, right?
00:22:57.200 --> 00:22:59.519
N8N workflows, Zapier, right?
00:22:59.680 --> 00:23:00.799
Make.com, right?
00:23:00.960 --> 00:23:02.319
Doing all these things.
00:23:05.039 --> 00:23:08.480
Agents and agentic models are doing all those, right?
00:23:08.640 --> 00:23:09.920
We've seen this explosion.
00:23:10.000 --> 00:23:12.400
I know this is a a cheeky example, right?
00:23:12.559 --> 00:23:19.440
We've seen the explosion of the, you know, clawed bot, molt bot, uh, open claw bot, right?
00:23:19.519 --> 00:23:24.880
But um I do think that's going to lead to something bigger, where we are gonna see very soon, right?
00:23:24.960 --> 00:23:33.440
I don't know if it's gonna be 2026 or early 2027, but we're gonna see enterprise versions of AI agents that work 24-7.
00:23:33.839 --> 00:23:36.480
And all that most many of us are gonna do, right?
00:23:36.559 --> 00:23:38.240
This is what I said in five to 10 years.
00:23:38.319 --> 00:23:42.000
I think our jobs are all gonna be for the most part, many of us are gonna be orchestrating.
00:23:42.160 --> 00:23:43.119
That's all it's gonna be.
00:23:43.279 --> 00:23:44.400
In the same way now, right?
00:23:44.480 --> 00:23:52.480
If you would have told someone 30 years ago, all you're gonna do every day is something that has to do with the internet, and you're just gonna, you know, push buttons on the internet.
00:23:52.799 --> 00:24:00.559
People are like that, that federal government project, the the that thing, no, right?
00:24:01.119 --> 00:24:06.799
Good, good, good work is done with a typewriter and a phone book and walking door to door, right?
00:24:06.960 --> 00:24:09.279
We look back at that now and we're like, no, right?
00:24:09.440 --> 00:24:10.319
The same thing.
00:24:10.480 --> 00:24:15.039
Everything is gonna be orchestrating agents on the front end and on the back end.
00:24:15.119 --> 00:24:21.839
I've I've used the word, you know, tastemaker, like whatever you want to say, but you are gonna be orchestrating agents.
00:24:22.160 --> 00:24:28.559
You are gonna be giving agents the knowledge, the data, the procedural data, right?
00:24:28.640 --> 00:24:30.880
That I talked about in uh step four, step five.
00:24:30.960 --> 00:24:33.119
Uh that's what you're gonna be giving agents.
00:24:33.200 --> 00:24:36.240
And then you're gonna be making sure that they produce their work.
00:24:36.319 --> 00:24:44.559
You're gonna be looking at their work, observing, tracing, making sure they say in your company's guardrails, your department's uh set rules, and that's it, right?
00:24:44.720 --> 00:24:48.240
Don't think human in the loop, expert-driven loops.
00:24:48.319 --> 00:24:52.880
And you need to put your smartest people at those oversight points.
00:24:53.119 --> 00:24:53.680
All right.
00:24:53.839 --> 00:24:55.519
And the biggest thing here, right?
00:24:55.599 --> 00:24:59.200
And this is something to tell you the truth, this is something I'm doing right now, right?
00:24:59.279 --> 00:25:02.880
I'm I'm going through process by process everything I do from a day-to-day basis.
00:25:02.960 --> 00:25:07.200
And I'm saying, how do I take myself out from being the operator of AI, right?
00:25:07.359 --> 00:25:15.680
Oh, I go into, you know, this AI program, you know, I type this in, or I, you know, I have this GPT and, you know, I paste this transcript in, whatever it may be, right?
00:25:15.759 --> 00:25:20.640
That's still me operating, even though I'm using AI and that's an AI first workflow.
00:25:20.799 --> 00:25:22.000
I'm still operating it.
00:25:22.160 --> 00:25:23.680
How do I get myself out of that?
00:25:23.839 --> 00:25:25.519
Just orchestrating it, right?
00:25:25.599 --> 00:25:27.359
It's being done automatically for me.
00:25:27.519 --> 00:25:28.880
It's scheduled, right?
00:25:28.960 --> 00:25:32.880
It's automatically pulling the dynamic data, and I don't have to do anything with it.
00:25:33.200 --> 00:25:36.400
That's the big push, and that is step seven.
00:25:36.559 --> 00:25:37.440
Easier said than done.
00:25:37.599 --> 00:25:41.839
And like I said, y'all, these steps, like AI, they're continually moving.
00:25:41.920 --> 00:25:45.440
But I think if you go through these seven steps, right?
00:25:45.680 --> 00:25:51.839
Go listen to our first uh couple volumes of the Start Here series as well, and go take our free uh prime prompt polish course as well.
00:25:52.000 --> 00:25:59.839
But I think following these seven steps, all right, putting them into practice, easier said than done, staying up to date because everything changes.
00:26:00.000 --> 00:26:02.000
But I wanted to make it easy for you.
00:26:02.240 --> 00:26:07.920
Because if you listen to this show, I want you to be in the haves, not the have nots.
00:26:08.160 --> 00:26:08.480
Okay.
00:26:08.799 --> 00:26:10.720
So I'm making it simple for you.
00:26:10.880 --> 00:26:13.039
If you have questions, let me know.
00:26:13.279 --> 00:26:15.440
But I hope this one was helpful.
00:26:15.599 --> 00:26:25.599
As we wrap up volume six of the Start Here series, now you know how to train your team and the seven essential steps to educate your organization on using large language models.
00:26:25.759 --> 00:26:26.559
Thank you for tuning in.
00:26:26.640 --> 00:26:30.880
If this was helpful, remember please go to startheareseries.com.
00:26:31.039 --> 00:26:35.680
That is going to give you free access to our inner circle community.
00:26:35.759 --> 00:26:42.880
Um, and you can also go take our uh Prime Prompt Polish course and go listen to all the Start Here series in one place.
00:26:43.039 --> 00:26:49.519
Network with other people who are trying to do the same thing you are doing right now, which is doubling down on AI in 2026.
00:26:49.599 --> 00:26:50.319
Thank you for tuning in.
00:26:50.480 --> 00:26:53.119
Hope to see you back tomorrow in every day for more everyday AI.
00:26:53.200 --> 00:26:54.240
Thanks, y'all.
00:00:01.919 --> 00:00:05.280
Companies will spend millions of dollars on AI.
00:00:05.440 --> 00:00:13.199
I mean, they'll buy thousands of enterprise seats for Microsoft 365 Copilot or ChatGPT Enterprise or whatever it is.
00:00:13.519 --> 00:00:22.000
CEOs will stuff as many AI-related buzzwords as they can at their weekly all-hands meeting to make it seem like they're on the cutting edge.
00:00:22.399 --> 00:00:24.239
But they're skipping something.
00:00:24.399 --> 00:00:32.640
The most basic thing: teaching their employees how to actually use whatever large language model they're providing.
00:00:33.200 --> 00:00:40.560
I've seen a sizable gap in 2026 starting to form in corporate America between the have and the have nots.
00:00:40.960 --> 00:00:45.439
I'm talking about the have trained their employees and the have not trained their employees.
00:00:45.520 --> 00:00:52.000
And if I'm being honest, I would say like 90% or more of companies fall in the latter group.
00:00:52.240 --> 00:00:55.439
Your company would probably fall there too.
00:00:57.439 --> 00:00:58.399
But why?
00:00:59.119 --> 00:01:01.200
Well, don't worry.
00:01:01.520 --> 00:01:10.719
Even if you are in that ladder group of, yeah, my company hasn't trained uh us, or if you're the CEO, we haven't trained our people on AI.
00:01:10.799 --> 00:01:13.680
I'm gonna lay that out for you on today's show.
00:01:13.840 --> 00:01:25.359
As part of our Start Here series, volume six, we're going over how to train your team on AI, the seven steps to educate your organization on using large language models.
00:01:25.519 --> 00:01:25.840
All right.
00:01:25.920 --> 00:01:27.439
I hope you're excited for this one.
00:01:27.599 --> 00:01:28.319
I am too.
00:01:28.480 --> 00:01:30.000
If you're new here, great.
00:01:30.159 --> 00:01:32.000
This is a great place to start.
00:01:32.239 --> 00:01:41.599
After doing this everyday AI thing for uh three years and more than 700 episodes, one of the most common questions is where do I start?
00:01:41.760 --> 00:01:42.879
You have so many podcasts.
00:01:42.959 --> 00:01:45.359
That's why we started the Start Here series.
00:01:45.439 --> 00:01:49.519
Uh so if you haven't already, please go to starthirseries.com.
00:01:49.680 --> 00:01:53.840
That'll give you free access uh to our inner circle community.
00:01:53.920 --> 00:01:58.239
And then in that start here series space, you can go and listen to every single podcast.
00:01:58.319 --> 00:01:59.359
You don't got to look around.
00:01:59.439 --> 00:02:02.159
Uh, you know, if you want to watch the video version, it's all there.
00:02:02.400 --> 00:02:11.520
But this is the essential podcast series to both learn the AI basics or to double down on your AI knowledge.
00:02:11.840 --> 00:02:17.520
So uh last week in volume five, so you can go listen to that episode 703.
00:02:17.599 --> 00:02:28.560
We talked about AI hallucinations, what they are, why they happen, and the right way to reduce the risk, which is a great transition into actually training your team on AI.
00:02:28.719 --> 00:02:29.199
Uh right.
00:02:29.360 --> 00:02:31.360
Notice I didn't start with that on volume one, right?
00:02:31.439 --> 00:02:32.479
We had to go over the basics.
00:02:32.639 --> 00:02:35.039
We had to, you know, understand all the jargon.
00:02:35.120 --> 00:02:41.120
We had to really tap into the AI mindset and, you know, talk about hallucinations and some of these other things.
00:02:41.280 --> 00:02:43.599
But uh let's get straight into it.
00:02:43.840 --> 00:02:52.719
And this is gonna be a quicker episode because I don't want it to get drowned out in the normal, you know, sometimes fluff that I might bring.
00:02:52.879 --> 00:02:55.520
But the gap that I'm talking about is alarming.
00:02:55.680 --> 00:03:04.080
So according to a McKinsey study, 92% of companies plan higher AI investment, but only 1% call their deployments mature.
00:03:05.199 --> 00:03:09.520
Yeah, I wasn't making it up when I said the gap is sizable.
00:03:09.599 --> 00:03:15.680
So that's like every single executive saying, yeah, AI is the most important thing and we're doubling down.
00:03:15.840 --> 00:03:21.120
Yet only 1% are saying, yeah, we are actually getting it right across the board.
00:03:21.360 --> 00:03:30.639
And uh another uh study from Forrester said that only 30% of large enterprises will even mandate AI training by the end of this year.
00:03:30.879 --> 00:03:34.240
This wasn't by the end of 2023 or 2024.
00:03:34.479 --> 00:03:45.280
By the end of the year 2026, only less than a third of organizations are even going to require AI training, which is absolutely bonkers, right?
00:03:45.680 --> 00:03:55.599
To think that just about every single organization in America is saying that yes, using AI and deploying AI is the most one of the most important things that we're trying to do as a company.
00:03:55.759 --> 00:04:02.719
We'll spend any money, we'll throw all the buzzwords, but we're not even gonna learn to understand it, or we're not even gonna take the time to learn to understand this thing.
00:04:02.800 --> 00:04:04.960
So here is the seven steps you need to go through.
00:04:05.120 --> 00:04:10.960
Step one, and I'm gonna break these, uh, I'm gonna break them down, but I'm gonna give them to you all here up front, not gonna make you wait to the end.
00:04:11.199 --> 00:04:13.360
Step one, leadership must go first.
00:04:13.520 --> 00:04:17.199
Step two, you've got to fix broken workflows before adding AI.
00:04:17.360 --> 00:04:20.240
Step three, you need to pick one AI platform and commit.
00:04:20.399 --> 00:04:24.480
Step four, you need to train in three layers, all right, not upskilling.
00:04:24.560 --> 00:04:26.399
We are rebuilding and unlearning.
00:04:26.560 --> 00:04:30.079
Step five, you need to document your procedures, not just your data.
00:04:30.319 --> 00:04:34.639
Step six, you need to mandate hands-on practice with real outputs.
00:04:34.800 --> 00:04:40.000
And step seven, you've got to go from operator to orchestrator, all right.
00:04:40.079 --> 00:04:44.079
Uh, and probably step zero would be take our free prime prompt polish course.
00:04:44.160 --> 00:04:44.399
All right.
00:04:44.480 --> 00:04:54.959
So if you do go to the starthearseries.com, uh, that will give you free access to our inner circle community, but you can also go take our free prime prompt polish course, uh, self-paced.
00:04:55.040 --> 00:04:57.759
It's about an hour and a half, two hours, right?
00:04:57.839 --> 00:05:00.079
More than 15,000 people have taken it.
00:05:00.160 --> 00:05:01.680
So yeah, go take that first.
00:05:01.759 --> 00:05:03.839
That's step zero, and then dive into step one.
00:05:03.920 --> 00:05:08.319
All right, so let's talk a little bit more about step one and why AI leadership must go first.
00:05:08.560 --> 00:05:12.480
And I don't want you to take this the wrong way.
00:05:12.720 --> 00:05:17.279
I'm not saying AI needs to be a top-down mandate because if it is, it will fail.
00:05:17.439 --> 00:05:22.000
When I say leadership must go first, I'm saying the CEO must use it daily.
00:05:22.079 --> 00:05:22.480
Uh, right.
00:05:22.560 --> 00:05:30.560
I had a great conversation a couple of years ago um with WWT's um CEO, Jim Kavanaugh, right?
00:05:30.800 --> 00:05:33.839
CEO of a$20 plus billion dollar revenue company.
00:05:33.920 --> 00:05:42.879
Um, and you know, we were chatting both, you know, on camera and before and after about how he's using AI every single day and has been for years, right?
00:05:43.040 --> 00:05:47.120
And even when employees come to him, you know, he's like, hey, is this AI native?
00:05:47.360 --> 00:05:47.680
Right?
00:05:47.920 --> 00:05:51.600
Like, don't don't bring anything to him first if it's not AI native.
00:05:51.759 --> 00:05:52.879
It's not AI first.
00:05:53.040 --> 00:05:59.920
And I think that's a great um leadership, leadership example of how you need to be, right?
00:06:00.000 --> 00:06:06.079
So if you are the CEO of a company or if you're speaking with the CEO of the company, your CEO can't be saying, yeah, go, go use AI.
00:06:06.160 --> 00:06:08.000
And they're still doing it the old way, right?
00:06:08.079 --> 00:06:10.000
This is change management 101.
00:06:10.079 --> 00:06:16.160
Uh, and you can't just, you know, have a couple, you know, key stakeholders, a couple AI champions, and expect this thing to work.
00:06:16.319 --> 00:06:20.560
It needs to start with leadership at the top using it.
00:06:20.639 --> 00:06:22.000
Again, not top-down mandate.
00:06:22.079 --> 00:06:23.519
You need to get everyone's, right?
00:06:23.600 --> 00:06:31.439
We'll probably do another episode in the Start Here series about, you know, proper ways to build AI uh strategies, et cetera, right?
00:06:31.600 --> 00:06:38.720
But you need to create those cultural moments, the wins of the week, the all hands demos, internal spotlights, and we're gonna get to more on that later.
00:06:38.879 --> 00:06:44.800
But if your top people are not actually using it, it's not actually going to work.
00:06:45.600 --> 00:06:50.160
Step two, fix broken workflows before adding AI, right?
00:06:50.319 --> 00:06:51.360
I called this before.
00:06:51.439 --> 00:06:55.839
You can throw uh, you know, makeup on an ugly pig, it's still an ugly pig.
00:06:56.000 --> 00:06:56.399
All right.
00:06:56.560 --> 00:07:05.279
If you think that adding AI to an antiquated, uh, you know, on tech workflow is going to help, it's probably not.
00:07:05.519 --> 00:07:09.040
In many cases, it could make things much worse, right?
00:07:09.279 --> 00:07:21.279
I can't tell you the number of use cases that I've come across personally, you know, when we consult companies, uh, you know, just stories, just chatting with people, you know, saying, like, hey, we try to implement AI in this process and it didn't work.
00:07:21.360 --> 00:07:23.360
And I'll be like, okay, tell me about this process.
00:07:23.519 --> 00:07:24.319
And they'll tell me about it.
00:07:24.399 --> 00:07:27.120
And I'm like, this process is completely broken, right?
00:07:27.439 --> 00:07:30.000
AI or not, this process isn't working, right?
00:07:30.079 --> 00:07:38.720
Because a lot of times, you you know, these processes, they're they're they're duct taped, they're uh, you know, just randomly glued together, they're MacGyver together.
00:07:38.879 --> 00:07:40.560
And it's like, okay, well, where did you get this?
00:07:41.040 --> 00:07:45.360
Okay, well, you know, this is you know, Jane in marketing, she she did this piece this way.
00:07:45.519 --> 00:07:48.720
Um, you know, when Bob and IT, I got this from him.
00:07:48.800 --> 00:07:50.240
And no, right?
00:07:50.959 --> 00:07:55.759
Your your standard workflows, if they don't make sense without AI, right?
00:07:56.079 --> 00:07:59.920
You're like, oh, this is uh uh the right way to do this this task.
00:08:00.560 --> 00:08:02.079
AI is not gonna help, right?
00:08:02.160 --> 00:08:07.199
People think AI, and this is again not to pound home on this one, right?
00:08:07.279 --> 00:08:11.040
But people think AI will fix broken processes, right?
00:08:11.120 --> 00:08:14.879
And it's not like I said, it'll make things worse.
00:08:15.120 --> 00:08:22.000
You need to redesign the entire workflow by being AI first, but you also need to be able to measure it as well.
00:08:23.439 --> 00:08:28.879
All right, all right, step three pick one AI platform and commit.
00:08:29.040 --> 00:08:34.240
All right, so we went into this one in depth on volume three of the start here series.
00:08:34.320 --> 00:08:35.919
That's episode 695.
00:08:36.399 --> 00:08:42.559
And I'm gonna spend a little bit more time on this one uh because I think it's important when I say pick one and commit, okay?
00:08:43.679 --> 00:08:52.799
Because I think a lot of organizations are Microsoft Copilot 365 organizations, and they still might need Chat GPT Enterprise, right?
00:08:52.960 --> 00:09:02.399
So I like to say, you know, Microsoft 365 Copilot, it is both the the worst and the best, right?
00:09:02.639 --> 00:09:14.559
Um, it's the worst because it can be extremely hard for organizations to understand where they use Copilot, how to enable it, permissions, right?
00:09:14.720 --> 00:09:19.440
It can it can be a nightmare if you don't have your ducks in order.
00:09:19.519 --> 00:09:22.639
It's kind of similar to your your your workflows, right?
00:09:22.799 --> 00:09:31.600
If your organization is not in order from an IT from a from a Microsoft Windows perspective, Copilot's probably not going to work very well.
00:09:31.679 --> 00:09:39.919
And a lot of those organizations realize that early on, and they're like, okay, you know, cope, it's it's almost like too much, right?
00:09:40.080 --> 00:09:42.960
In some instances, it can be too much.
00:09:43.120 --> 00:09:46.559
And in those instances, yeah, you might have to pick one of the others, right?
00:09:46.639 --> 00:09:58.399
But for the most part, I say your online operating systems, your AI operating systems, it's Anthropics Claude, it's Google Gemini, it's open AI's Chat GPT.
00:09:58.559 --> 00:10:07.360
And amongst those three, you should probably to get started with, you should be choosing one and moving most of your critical day-to-day processes in there.
00:10:07.519 --> 00:10:10.720
No matter what your kind of uh AI go-to-market is, right?
00:10:10.799 --> 00:10:16.080
Whether you're starting with the champion team, whether you're starting with certain departments, whether you're doing the entire company, right?
00:10:16.159 --> 00:10:17.360
I can't say what's the best.
00:10:17.440 --> 00:10:23.200
It just depends on the size of your company, how much training you have, what your processes look like, et cetera, right?
00:10:23.279 --> 00:10:23.519
Are you?
00:10:23.679 --> 00:10:26.720
Do you have 10,000 employees or do you have 50?
00:10:26.799 --> 00:10:27.440
Um, right.
00:10:27.600 --> 00:10:34.080
But for the most part, you should all be learning and sandboxing in the same AI operating system, one of those three, right?
00:10:34.159 --> 00:10:42.399
And then eventually, yeah, maybe you're uh, you know, if you have a software engineering department, you know, maybe they might end up using Claude or something like that.
00:10:42.559 --> 00:10:53.120
Maybe your creative team might end up using Gemini, but you should all be learning and bringing the processes in and measuring ROI and finding your use cases and scaling in a single operating system.
00:10:53.279 --> 00:11:00.639
Because what happens when you start using, oh, we're gonna use ChatGPT for this, use Gemini for this, use Claude for this, use Copilot for this.
00:11:00.799 --> 00:11:07.840
All of a sudden, then you're gonna find it very easy to say, okay, well, then we're gonna use these 10 other tools as well.
00:11:08.080 --> 00:11:11.679
And that 10 other tools becomes 30, becomes 50, right?
00:11:11.840 --> 00:11:26.720
A lot of uh did a lot of consulting calls in 2025, and so many of them, so many of them, I was blown away, right, by the number of AI tools, especially medium-sized businesses were using, right?
00:11:26.799 --> 00:11:40.399
It's it's it's like some companies had, you know, 10 different AI tools that they were using just for writing different kinds of copy, literally 10 different tools, not 10 different GPTs, 10 different projects, 10 different gems within one system.
00:11:40.559 --> 00:11:43.759
10 different tools that they were using for writing, right?
00:11:44.159 --> 00:11:49.840
I think that not focusing on a single AI operating system leads to that shiny AI object syndrome.
00:11:49.919 --> 00:11:50.720
So pick one.
00:11:50.879 --> 00:11:54.159
And then also, you gotta pick one.
00:11:54.399 --> 00:11:55.039
All right.
00:11:55.600 --> 00:12:00.639
Studies have shown, right, whether you want to call it shadow IT, AI sprawl.
00:12:00.799 --> 00:12:04.240
I've been calling it second computer AI since 2023.
00:12:04.320 --> 00:12:07.519
Uh, you know, I called it happening back then, and it is prominent.
00:12:07.759 --> 00:12:10.559
Even if you think that your people are not using, right?
00:12:10.720 --> 00:12:17.200
Your team, if you're a department head, if you're the CEO, whatever, if you're an HR, everyone's using AI.
00:12:18.399 --> 00:12:19.519
Okay, they are.
00:12:19.679 --> 00:12:27.519
Whether they're using it once a week or for every single project, whether you've greenlit it or not, people are using AI.
00:12:27.600 --> 00:12:33.120
A lot of times on a second computer on a different device, uh, right, incognito window, whatever it may be.
00:12:33.519 --> 00:12:42.960
So if you have some hesitation about bringing your data, bringing your processes into a large language model, they're already there, right?
00:12:43.120 --> 00:12:44.399
A lot of people don't know.
00:12:44.639 --> 00:12:47.600
Everything on your website, it's already in large language models.
00:12:47.679 --> 00:12:48.320
People don't know, right?
00:12:48.399 --> 00:12:51.360
They they upload PDFs or these documents from five years ago.
00:12:51.519 --> 00:12:53.840
And it's like, yeah, that's that's in a large language model.
00:12:53.919 --> 00:13:00.080
That's I can access it by looking at a at a site map uh on your on your website.
00:13:00.240 --> 00:13:05.919
People don't understand the amount of data that is already available about your company on social media, etc.
00:13:06.320 --> 00:13:11.200
Everything, almost everything from a public front-facing perspective is already in large language models.
00:13:11.279 --> 00:13:14.559
And then when you think about, oh, my company's data, blah, blah, blah, it's so special.
00:13:14.639 --> 00:13:15.519
No, it's not, right?
00:13:15.600 --> 00:13:16.320
What do you use?
00:13:16.399 --> 00:13:18.159
You use a cloud provider, right?
00:13:18.320 --> 00:13:18.879
Guess what?
00:13:19.039 --> 00:13:22.879
Probably the AI operating system to start with is whatever cloud provider you use.
00:13:23.039 --> 00:13:29.679
So if you use, you know, Microsoft Azure, SharePoint, OneDrive, whatever, well, probably start with Microsoft 365 Copilot.
00:13:29.840 --> 00:13:32.480
If you're using Google, well, you should probably start with Google, you know.
00:13:32.720 --> 00:13:33.919
AWS, I don't know.
00:13:34.000 --> 00:13:36.720
You can roll the dice and and try the Nova, right?
00:13:36.799 --> 00:13:38.960
I wouldn't personally recommend it, but you can try that.
00:13:39.039 --> 00:13:46.240
Uh, but you have to stop tool sprawl early because once it sprawls, it's too much to handle.
00:13:46.559 --> 00:13:50.480
So get your AI platform and commit to it.
00:13:50.799 --> 00:13:53.759
Step four, train in three layers.
00:13:54.000 --> 00:13:54.639
All right.
00:13:55.600 --> 00:14:02.399
This is where I might rub a lot of the people the wrong way that came up with their AI slogans uh in 2024 and 2025.
00:14:02.559 --> 00:14:04.159
People at big tech companies, right?
00:14:04.240 --> 00:14:07.600
Everyone's you know, sprinkling the word upskill and reskill.
00:14:07.759 --> 00:14:09.200
That's a great way to fail.
00:14:09.440 --> 00:14:16.000
Um AI, large language models are smarter than us, right?
00:14:16.879 --> 00:14:18.960
Um, I was a Pulitzer fellow.
00:14:19.120 --> 00:14:23.519
I won all these writing awards back in the day, ACP story of the year, right?
00:14:23.600 --> 00:14:24.799
All these things.
00:14:25.440 --> 00:14:27.440
AI is way better at writing than me.
00:14:27.679 --> 00:14:28.000
Right?
00:14:28.159 --> 00:14:31.200
If you're looking at the normal work slot that's out there, absolutely not.
00:14:31.279 --> 00:14:36.159
But I know how to use AI, I can make it write better than 99.9% of writers.
00:14:36.240 --> 00:14:41.840
And I can say that because I've been getting paid to write for 25 years and I've been writing at a very high level, right?
00:14:42.000 --> 00:14:47.679
Same thing with anything financial analysis, um, marketing, comms.
00:14:50.159 --> 00:14:52.480
Like pick your department, HR.
00:14:53.519 --> 00:14:54.320
Doesn't matter.
00:14:54.879 --> 00:14:59.840
AI models, if you're using them the right way, they're they're better than us and they're smarter than us.
00:15:00.000 --> 00:15:15.200
So to think that we can just uh sprinkle in AI at the top and the side here and there with upskilling, reskilling, no, no, you have to rebuild because we should probably be um taking a complete shift in how we work, right?
00:15:15.679 --> 00:15:17.360
Complete mindset shift, right?
00:15:17.919 --> 00:15:22.639
We are orchestrating but not pushing buttons, right?
00:15:22.799 --> 00:15:28.320
Because these models, for the most part, they're getting things done 10 to 100 times faster.
00:15:28.559 --> 00:15:38.559
So kind of the the role, I think, of the average domain-specific knowledge worker over the last next five to 10 years is going to change, change drastically.
00:15:38.799 --> 00:15:42.240
How me and you are working in 10 years, right?
00:15:42.399 --> 00:15:51.759
If you're if if you work a traditional desk job and you get paid for what you know in your brain, in five to 10 years, sorry, it's not gonna be like that anymore, right?
00:15:51.919 --> 00:15:56.559
It might be like two to three years, but in five to 10 years, it's not gonna be like that.
00:15:56.639 --> 00:16:00.639
So you can't just think of upskilling and reskilling, you have to unlearn and rebuild.
00:16:00.799 --> 00:16:01.120
All right.
00:16:01.200 --> 00:16:05.279
So what that means, layer one, you have to get literacy, you have to understand.
00:16:05.440 --> 00:16:09.039
So go listen to the first couple volumes of this series and you'll be there.
00:16:09.200 --> 00:16:15.039
Uh, layer two of this uh training process, it's domain specific by role.
00:16:15.279 --> 00:16:15.600
Okay.
00:16:15.919 --> 00:16:20.240
Um, you can't just have one training for your entire organization, right?
00:16:20.480 --> 00:16:27.279
You should be breaking it down and saying, hey, here's how uh as a, we'll just use marketing, right?
00:16:27.440 --> 00:16:38.799
As a marketer, here's how we use our company's data and produce better marketing collateral with our AI operating system, right?
00:16:38.960 --> 00:16:42.000
You do need AI literacy that works organization-wide.
00:16:42.080 --> 00:16:48.320
And that's why, you know, something like what we do with the chat uh with the Prime Prompt Polish, or, you know, we do that for businesses.
00:16:48.480 --> 00:16:50.480
I think that's great for everyone across the board.
00:16:50.639 --> 00:16:55.840
But then after that, you have to get into domain-specific uh training by role or by department type.
00:16:55.919 --> 00:16:59.840
And then number three, you have to understand data and procedures.
00:17:00.080 --> 00:17:08.720
Bad data has plagued everyone for decades before AI, and it's going to plague you and your company even worse.
00:17:08.880 --> 00:17:16.960
So if you thought that somehow AI could be a band-aid or makeup on your data problem, no, it is only going to expose it and make it worse.
00:17:17.119 --> 00:17:19.920
All right, because I've been, I've been saying this for a long time.
00:17:20.079 --> 00:17:29.680
Your good habits over the last 10, 15, 20, 30 years, or your companies or your departments are not going to matter anymore when we're talking about AI native workplaces.
00:17:29.920 --> 00:17:34.720
Step five, document your procedures, not just your data.
00:17:34.880 --> 00:17:35.279
All right.
00:17:35.440 --> 00:17:36.160
That's huge.
00:17:36.319 --> 00:17:40.400
And when I say document your procedures, all right.
00:17:40.799 --> 00:17:42.720
And I'm sorry if you're a longtime listener.
00:17:42.799 --> 00:17:56.480
Uh, you know, if you've listened to all 700 and some episodes, you know, some of these things I'm gonna be repeating myself, but documenting your processes is about collecting that human intelligence, right?
00:17:56.880 --> 00:18:00.160
Think of that your coworkers that have been there for a very long time.
00:18:00.240 --> 00:18:04.240
And you're like, man, if this person ever left, we'd be screwed, right?
00:18:04.480 --> 00:18:06.880
Once Deborah's gone, we're done.
00:18:07.119 --> 00:18:08.640
We're done without Deborah, right?
00:18:08.799 --> 00:18:10.720
Okay, we'll start documenting what Deborah does.
00:18:10.880 --> 00:18:13.279
Start documenting what Deborah knows, right?
00:18:13.440 --> 00:18:23.119
Start seeing how AI and and Deborah can better coexist, uh, right, because that is, I think, the last mile problem of AI implementation.
00:18:23.279 --> 00:18:34.640
Uh, so much of the early uh, you know, connecting your data um with large language models has been focused on, you know, some version of retrieval uh retrievalog meta generation or RAG, uh, right?
00:18:34.960 --> 00:18:52.799
Um and now these AI operating systems have a version of that very simple to bring your data in one click from whatever cloud storage, but you have to start getting and collecting and curating and cleaning different types of data that we historically have never needed as companies, right?
00:18:52.960 --> 00:19:02.240
Which is how our smartest people think, how they solve problems, you know, their own internal uh decision tree that's not structured data, right?
00:19:02.400 --> 00:19:04.960
I think of like flow charts and decision trees, right?
00:19:05.039 --> 00:19:08.799
So even if you're not a big, you know, traditional machine learning, right?
00:19:09.200 --> 00:19:18.079
Traditional artificial intelligence, machine learning, you can think of one type as like a decision tree or like a if this then or if if else uh tree, right?
00:19:18.319 --> 00:19:22.079
So it's kind of like that, but with unstructured data.
00:19:22.240 --> 00:19:23.759
That's great for structured data, right?
00:19:23.839 --> 00:19:25.359
For things that you can put in spreadsheet.
00:19:25.519 --> 00:19:28.720
If a value is more than five, this happens, right?
00:19:28.880 --> 00:19:33.920
If we have more than 30 of these in stock, then this email gets sent out, right?
00:19:34.160 --> 00:19:37.599
But what about those things that aren't quantifiable in a spreadsheet?
00:19:37.920 --> 00:19:40.960
That's what you have to start documenting in step five.
00:19:41.119 --> 00:19:43.839
You have to document your procedures, not just your data.
00:19:44.000 --> 00:19:47.039
That is the last mile problem of AI implementation.
00:19:47.200 --> 00:19:54.240
Because right now, let's be honest, for the most part, if you know what you're doing, right?
00:19:54.319 --> 00:20:00.160
If you're using the right AI model and you're using the right mode and you're connecting your data, that's tablescapes now, right?
00:20:00.400 --> 00:20:01.519
Everyone's doing that.
00:20:01.680 --> 00:20:05.920
This is what's gonna, at least for 2026, maybe 2027, I think.
00:20:06.160 --> 00:20:18.880
This is what's gonna help you, your department, your company create separation if you start documenting your procedures, your company's IP, your department's special sauce, Deborah's brain.
00:20:19.039 --> 00:20:20.720
You've got to start collecting it all.
00:20:21.279 --> 00:20:26.319
Step six, mandate hands-on practice with real outputs, right?
00:20:27.200 --> 00:20:32.240
The simplest way I tell people do do Friday lunch, Friday lunch days, right?
00:20:32.480 --> 00:20:35.920
Block out 90 minutes, have a good lunch, get people together, right?
00:20:36.000 --> 00:20:37.839
Especially if you're a remote organization.
00:20:38.079 --> 00:20:40.960
This is your AI session, right?
00:20:41.599 --> 00:20:45.920
But employees need to have hands-on keyboard, right?
00:20:46.160 --> 00:20:47.200
Not just on their phone.
00:20:47.359 --> 00:20:49.039
No, hands-on keyboard.
00:20:49.279 --> 00:20:51.359
You need to be sharing, you need to be workshopping.
00:20:51.519 --> 00:20:54.319
This is like, you know, mini hackathon styles.
00:20:54.480 --> 00:20:57.759
Uh, you know, yeah, you can do a hackathon in 30 minutes, right?
00:20:57.839 --> 00:21:02.319
You go hack one of your department's biggest problems, and everyone go take their computer.
00:21:02.480 --> 00:21:04.880
You know, everyone takes 10, 15, 20 minutes.
00:21:04.960 --> 00:21:07.440
Someone gets up, they say, they say, this is the problem.
00:21:07.599 --> 00:21:09.839
Here's how I've been trying to solve this with AI.
00:21:10.000 --> 00:21:11.759
Here's what's working, here's what's not.
00:21:11.920 --> 00:21:21.200
Everyone takes 15 to 30, 15 to 30 minutes to try and go build their version, take what Bill uh in IT made, made better, try something else completely different, right?
00:21:21.359 --> 00:21:25.440
You have to be doing this continually because AI is constantly changing.
00:21:25.759 --> 00:21:36.480
Y'all, my only job, my only job is to talk to smart AI people every day, share that with you, read about read about AI, test it.
00:21:36.559 --> 00:21:38.160
That's all I do every single day.
00:21:38.880 --> 00:21:41.680
Eight to 14 hours, right?
00:21:42.160 --> 00:21:42.720
Depends.
00:21:42.960 --> 00:21:46.079
Um, but I can't keep up, right?
00:21:46.559 --> 00:21:47.519
This is my job.
00:21:47.920 --> 00:21:50.640
So you can't keep up, your department can't keep up.
00:21:50.799 --> 00:21:53.839
No one can keep up unless you are intentional about it.
00:21:54.079 --> 00:21:54.480
All right.
00:21:54.720 --> 00:21:59.119
You have to have measured adoption with outputs, not usage rates, right?
00:21:59.200 --> 00:22:05.039
People look at You know, oh, I look in ChatGPT Enterprise and you know our utilization went up from 10 to 15.
00:22:05.359 --> 00:22:06.000
So that's good.
00:22:06.240 --> 00:22:09.039
No, no, you need to look at outputs.
00:22:09.200 --> 00:22:16.799
You need to look at what people are actually producing, and you need to give the space to learn and you need to be able to measure what's good and what's not, right?
00:22:16.960 --> 00:22:21.519
And not just, okay, these 10 people, you know, everyone go show me what you created.
00:22:21.680 --> 00:22:23.759
No, now let's see why it didn't work.
00:22:23.920 --> 00:22:29.200
If it didn't work, or if it did work, let's go and let's go ahead and look at that chain of thought and see why it did work.
00:22:29.440 --> 00:22:31.599
All right, step seven, this one's a big one.
00:22:31.759 --> 00:22:33.279
Another mindset shift.
00:22:33.359 --> 00:22:40.240
So many of these things, mindset, um, process, behavioral, but shifting from operator to orchestrator.
00:22:40.640 --> 00:22:41.039
Right.
00:22:41.279 --> 00:22:51.440
So much, I think, of what successful people in AI did uh between 2022 when ChatGPT came out until 2024, 2025, right?
00:22:51.599 --> 00:22:54.000
We were the we were the glue, right?
00:22:54.319 --> 00:22:57.119
A lot of copying and pasting, chaining things together, right?
00:22:57.200 --> 00:22:59.519
N8N workflows, Zapier, right?
00:22:59.680 --> 00:23:00.799
Make.com, right?
00:23:00.960 --> 00:23:02.319
Doing all these things.
00:23:05.039 --> 00:23:08.480
Agents and agentic models are doing all those, right?
00:23:08.640 --> 00:23:09.920
We've seen this explosion.
00:23:10.000 --> 00:23:12.400
I know this is a a cheeky example, right?
00:23:12.559 --> 00:23:19.440
We've seen the explosion of the, you know, clawed bot, molt bot, uh, open claw bot, right?
00:23:19.519 --> 00:23:24.880
But um I do think that's going to lead to something bigger, where we are gonna see very soon, right?
00:23:24.960 --> 00:23:33.440
I don't know if it's gonna be 2026 or early 2027, but we're gonna see enterprise versions of AI agents that work 24-7.
00:23:33.839 --> 00:23:36.480
And all that most many of us are gonna do, right?
00:23:36.559 --> 00:23:38.240
This is what I said in five to 10 years.
00:23:38.319 --> 00:23:42.000
I think our jobs are all gonna be for the most part, many of us are gonna be orchestrating.
00:23:42.160 --> 00:23:43.119
That's all it's gonna be.
00:23:43.279 --> 00:23:44.400
In the same way now, right?
00:23:44.480 --> 00:23:52.480
If you would have told someone 30 years ago, all you're gonna do every day is something that has to do with the internet, and you're just gonna, you know, push buttons on the internet.
00:23:52.799 --> 00:24:00.559
People are like that, that federal government project, the the that thing, no, right?
00:24:01.119 --> 00:24:06.799
Good, good, good work is done with a typewriter and a phone book and walking door to door, right?
00:24:06.960 --> 00:24:09.279
We look back at that now and we're like, no, right?
00:24:09.440 --> 00:24:10.319
The same thing.
00:24:10.480 --> 00:24:15.039
Everything is gonna be orchestrating agents on the front end and on the back end.
00:24:15.119 --> 00:24:21.839
I've I've used the word, you know, tastemaker, like whatever you want to say, but you are gonna be orchestrating agents.
00:24:22.160 --> 00:24:28.559
You are gonna be giving agents the knowledge, the data, the procedural data, right?
00:24:28.640 --> 00:24:30.880
That I talked about in uh step four, step five.
00:24:30.960 --> 00:24:33.119
Uh that's what you're gonna be giving agents.
00:24:33.200 --> 00:24:36.240
And then you're gonna be making sure that they produce their work.
00:24:36.319 --> 00:24:44.559
You're gonna be looking at their work, observing, tracing, making sure they say in your company's guardrails, your department's uh set rules, and that's it, right?
00:24:44.720 --> 00:24:48.240
Don't think human in the loop, expert-driven loops.
00:24:48.319 --> 00:24:52.880
And you need to put your smartest people at those oversight points.
00:24:53.119 --> 00:24:53.680
All right.
00:24:53.839 --> 00:24:55.519
And the biggest thing here, right?
00:24:55.599 --> 00:24:59.200
And this is something to tell you the truth, this is something I'm doing right now, right?
00:24:59.279 --> 00:25:02.880
I'm I'm going through process by process everything I do from a day-to-day basis.
00:25:02.960 --> 00:25:07.200
And I'm saying, how do I take myself out from being the operator of AI, right?
00:25:07.359 --> 00:25:15.680
Oh, I go into, you know, this AI program, you know, I type this in, or I, you know, I have this GPT and, you know, I paste this transcript in, whatever it may be, right?
00:25:15.759 --> 00:25:20.640
That's still me operating, even though I'm using AI and that's an AI first workflow.
00:25:20.799 --> 00:25:22.000
I'm still operating it.
00:25:22.160 --> 00:25:23.680
How do I get myself out of that?
00:25:23.839 --> 00:25:25.519
Just orchestrating it, right?
00:25:25.599 --> 00:25:27.359
It's being done automatically for me.
00:25:27.519 --> 00:25:28.880
It's scheduled, right?
00:25:28.960 --> 00:25:32.880
It's automatically pulling the dynamic data, and I don't have to do anything with it.
00:25:33.200 --> 00:25:36.400
That's the big push, and that is step seven.
00:25:36.559 --> 00:25:37.440
Easier said than done.
00:25:37.599 --> 00:25:41.839
And like I said, y'all, these steps, like AI, they're continually moving.
00:25:41.920 --> 00:25:45.440
But I think if you go through these seven steps, right?
00:25:45.680 --> 00:25:51.839
Go listen to our first uh couple volumes of the Start Here series as well, and go take our free uh prime prompt polish course as well.
00:25:52.000 --> 00:25:59.839
But I think following these seven steps, all right, putting them into practice, easier said than done, staying up to date because everything changes.
00:26:00.000 --> 00:26:02.000
But I wanted to make it easy for you.
00:26:02.240 --> 00:26:07.920
Because if you listen to this show, I want you to be in the haves, not the have nots.
00:26:08.160 --> 00:26:08.480
Okay.
00:26:08.799 --> 00:26:10.720
So I'm making it simple for you.
00:26:10.880 --> 00:26:13.039
If you have questions, let me know.
00:26:13.279 --> 00:26:15.440
But I hope this one was helpful.
00:26:15.599 --> 00:26:25.599
As we wrap up volume six of the Start Here series, now you know how to train your team and the seven essential steps to educate your organization on using large language models.
00:26:25.759 --> 00:26:26.559
Thank you for tuning in.
00:26:26.640 --> 00:26:30.880
If this was helpful, remember please go to startheareseries.com.
00:26:31.039 --> 00:26:35.680
That is going to give you free access to our inner circle community.
00:26:35.759 --> 00:26:42.880
Um, and you can also go take our uh Prime Prompt Polish course and go listen to all the Start Here series in one place.
00:26:43.039 --> 00:26:49.519
Network with other people who are trying to do the same thing you are doing right now, which is doubling down on AI in 2026.
00:26:49.599 --> 00:26:50.319
Thank you for tuning in.
00:26:50.480 --> 00:26:53.119
Hope to see you back tomorrow in every day for more everyday AI.
00:26:53.200 --> 00:26:54.240
Thanks, y'all.
