AI Leadership Q&A wth Noelle

Hello, good to see you all. All right, here we go. I actually don't know exactly how I'm live right now, so we'll see what happens. I think this is our— I think it's right. We'll see what happens. Welcome to the AI Leadership Institute and our I Love AI community call. This is a chance for you to hear all about the things that are happening in the world of AI from my perspective. And in my purview, and also get a chance to hear what's coming up, where I'm gonna be, and what opportunities you have to grow your own experience in the world of AI. So great to see you all here. Thank you for being here. And let's go ahead and get started. I'm gonna close out that window. So just to give you a sense of things that have happened so far this week, really, really cool stuff. I've had a chance to be in a few cities. I'm on my book tour, so I am going to multiple cities every single week. Sometimes I get to speak here in South Florida, but most of the time I have to travel around. Uh, so over the last couple weeks I've been traveling from place to place and talking about how to scale responsible AI. Uh, and of course my book Scaling Responsible AI is, uh, almost out. It comes out April 1st, but it is available on pre-order right now. And as a matter of fact, I will make sure that we throw our link into the chat so that you all have a chance because we're offering something pretty special. We have an opportunity for everyone to learn. And many of you know we have a very successful course. It's called the AI Bot Builder Bootcamp, and it teaches you the first part of Agentic AI. So there's two different parts of it. First, you have to know how to build a single system before you can build a bunch of systems that talk to each other. And so the very first thing we need to know how to do is actually just build a system end to end safely, responsibly, and successfully. And so I'm a big believer. I spent most of last year teaching that. As a matter of fact, the first— and the first version of this class came out during COVID We had 3.4 million people take this class on LinkedIn, and then we updated it for generative AI a few years— a couple of years later. And we had, you know, 500,000 people take that course. And now, like I said, we're in the world of agentic. Agentic AI. And I really realize how important it is people understand kind of the foundations. Agentic is like running or flying, but understanding how to build an AI system with your words— another word for this is prompt engineering or system prompting— this is an incredibly important skill. So as a result, I decided to do something crazy, and I've gotten lots of feedback, good and You know, less good. But I wanted to make sure that you have no reason not to learn this technology, learn this space. And so, as a result of you purchasing the book and getting access to the preorder, you will also get immediate access to my on-demand version of the Bot Builder course. I think it's pretty good. You know me, I can't really watch myself for very long, but I've heard it's really good. People have taken it, they loved it. I also watched like the first 20 minutes and I was like, okay, You know, it's not terrible. We're always harder on ourselves. But I definitely want to make sure that you all know that it is a very unique opportunity between now and the end of next week. You'll have a chance. So tell your friends, all they have to do is buy a $35 book and they get access to a $647 bestselling AI course on building your first agent or building your first bot. And it's really good, really exciting. So I hope people— I hope you will take advantage of that. Now, Yesterday, was it yesterday, Sky? I can't even remember. No, day before yesterday. I can't remember. What's today? Day before yesterday on Wednesday, I had the beautiful, amazing opportunity to join, uh, 2,000 of my new friends, uh, in Atlanta at the Atlanta World Congress. But we also— thank you, Sky. Wednesday. Yes. Uh, but we also— I got a chance to meet with some of our very own I Love AI, AI Leadership members. So Sky is here in the audience. I met with Sky and Verole. It was so cool. Like, we got a chance to hang out in person. Many of you saw and commented on the pictures that I posted, but how awesome is that? Like, that's the world I want to live in where I go to a city and there are people I know there that I don't have to, like, you know, like, it's just so cool. So we, we had a meal, we broke some bread, it was fantastic. But we, I was speaking at a Rural Electric Co-op event, an association that serves rural electric co-ops all over the world. And it was really interesting to me because just, you know, I don't know, sometimes I kind of get maybe in my, my specific way of thinking and specific types of companies that I think can leverage AI. And every time I go and speak to an association, I recognize how big and wide the, the net is cast when it comes to people who need to understand the power and capability of AI in their own industry. And there are met in all these industries. There are people that know, like for example, the work of someone in a rural office, right? Managing the electric co-op. Like there are people that do that work every single day. And if they knew how to leverage AI to make their life better, easier, how to serve their customers more quickly, they would be so grateful for it because it's only them. It's not like they're in a big city where they can hire anyone they want and that they can grow infinitely. Like They're a small, typically pretty small rural organization. And so AI is perfect in that scenario, allows us to massively grow that opportunity. Oh yeah. Oh my gosh, yes, you can become an insider. I'm also gonna put in the chat the funnel that you can access that will take you over to— actually, I don't know if you have that, Sherilyn, but the one that'll actually take you to getting access to the boot camp, we could put that in as well. I will put it in here in case I never sent it to you, which is entirely possible. So sorry about that. But it is www.upskillingai, one of my favorite URLs, upskilling.ai/book. So there you are. There you are. Awesome. Yeah, it was amazing, amazing, amazing event. And it just encourages me to encourage you all that, like, your skill set in whatever domain you have landed yourself in, whether it's finance, rural electric co-ops, legal firms, tax professionals, chiropractors, like whatever your skill set is, that skill set plus your interest in artificial intelligence is going to become a superpower for that industry. It is what will differentiate you. So it's funny, over the years I stopped saying certain things. Sky was there, I think, when this woman said, oh yeah, you were there. And she came up and she goes, you know, the thing I always say, or, or the thing I love that I heard someone say, and it's something of course that I've said for years, but I stopped saying it because I feel like y'all have heard it before, but it's this concept of, you know, when you, um, when you think about the world of AI, you know, people shouldn't be afraid that AI's gonna take their job because AI is not really good at taking their job. It doesn't even know what their job is. What they should be worried about, instead of AI taking your job, you should be worried about someone else who knows AI taking your job. Because I say this all the time, you, plus AI is going to be able to do more, see more, have more, implement more, move faster than your counterpart who has not implemented that technology to help them in their daily life. And the time is now. Like, literally between now and 3 years from now, there will be a huge divide that is created. I'm trying to close it by educating people every single week. But a huge divide is being created between those who are like, oh, I'm not so sure about that, and those of us like you all who are jumping in. And so it's critical that you do that. Exactly. Like, not just the AI skills alone are not good enough. We're starting to see that, right? Like, the, the mortar starts crumbling on these experts that went to market very quickly but don't have the industry experience, don't have the understanding of how it actually— how you scale something in practice. So those of us with actual lived experience in these domains, right, whether it's finance, healthcare, supply chain, user experience, science, product development, project management, like whatever our area is, when you start partnering with AI, you become better than any other PM that is trying to do that without that skill. And it's a natural evolution where we're not saying, hey, come AI, take my job, do all this for me. We are collaborating with this AI system. It's called symbiotic. We're creating a symbiotic relationship with these AI systems. And that is a critical approach you need to take. So today, before I do thing I always ask to collaborate with an AI system. Now, granted, I also think an important point is that you don't just collaborate willy-nilly— is that a word— with an AI system, that you don't just do everything in ChatGPT. I think it's important to realize there are some models that are good at certain things. Even inside OpenAI, there are multiple models that are good at different things. And they try to give us some advice. They try to tell us, like, hey, you know, when you look at the model description, it'll be like, this is for advanced reasoning and for math and for logic. This one's for fast answers to questions. One of the things that we have unfortunately trained the world on and from a software engineering perspective is that the latest number of a model or of anything is often the best, right? iPhone 16 point whatever that we're at, right? 16, 17, 18, 19. That will constantly strive towards getting that latest model. But in the world of AI, that's not actually the case. We saw it first when 4.0 or 4 Omni released and it was very good 'cause it was an omni model, which me— meant that it was— it basically had the ability to pull together a multi-model approach. So you could ask it things that before you had to go to special models to do. So I could go to 4o, um, for Omni. I could go to it for things that DALL·E would've done in the past, or Codex would've done, right? Codex being that model that was specifically for software engineering, DALL·E being that model that was specifically for image generation. The benefit of a 4.0 or an omni model is that it collected all of those models into a single interface. So we could just say one thing and it would figure it out amongst all those capabilities. Agentic, really, right? I ask for a thing and that thing that I'm talking to, ChatGPT 4.0, doesn't actually know how to do that thing, but it has its friend Codex, it has its friend DALL·E, it has its friend Logic Reasoning, and it can delegate to those models effectively. But in doing that, I usually, I always kind of think about what happens kind of in model logic when you are using a large language model. Like, the good news is, is that it's pretty— it's generally pretty good at general things. So ideating on something, it also is trained on the world's information. Okay, not the world's information, but the internet. And being trained on the internet, though it's flawed in many ways, It is very good at understanding things that are on the internet. I don't know how many of you remember the internet when it was first born, but it was born out of academia. So it's born out of academic papers and documentation and research. So a lot of the underlying capability or understanding that these models have come from the huge amount of academic information that was infused to the internet in the beginning. Of course, since then we've got a lot of editorial and opinion content that is woven in, and that's why we can never really trust an open, free model on the internet to be 100% accurate. That's why every one of these models at the bottom will say something to you like, I could be wrong about this, I might make mistakes, um, please double-check for accuracy, right? Those statements are placed there because of this intermingling of truth and opinion. And when you ask a model, it's going to generate you an answer that is often a blend of that. Which means if you're asking for a fact, you might actually get an opinion, but stated as a fact, which is how a lot of humans talk to me sometimes, actually, right? But that is important. And so we're now in a world where we really need to kind of focus on what is my unique skill set. So for me, for example, I real— I mean, of course, I like applied AI. I fell in love with it the moment I got an opportunity to use it, and that was back in Amazon Alexa, right? So when I was an Alexa team member in 2014, as I took on those reins and was like, okay, I'm gonna do this, I didn't know anything about AI, like some of you probably. Like, I didn't know what natural language understanding was. I didn't know what speech recognition was. I didn't know what echo cancellation was or beamforming or any of these things that I'm now very well-versed in. But I didn't know any of that. I didn't know about computer vision. I didn't know about any of it. And, you know, diffusion, like, no, not, not part of my understanding. However, what I did know was that there was a model that I could leverage via an API, and I'm a software engineer by trade. And I was like, wait, I could ask this model, I could give it an image, and it will give me back something. And that's what we did last night. So the members of the AI Leadership Institute, of course, I welcome you all to join us. But the members of the AI Leadership Institute last night. Every month we do what's called a live learning lab where we all get online and we leverage a technology. In this case, it was Microsoft. And we log into Power Builder and we learned how to build and deploy an AI model using a tool called AI Builder. And that's part of the Power Platform. The benefit of it is it's no-code, low-code, like low-code, no-code. Anyway, however you say that. But basically, we didn't have to write any code to do it. I don't think we wrote any code, did we? No, the only thing maybe that hints at code was grabbing a variable. But even then, it was like content assisted. You know, as soon as we typed it in, it showed us the options and we could pick it from a list. So very, very low-code, like, technology requirements. What was necessary was an understanding of the business problem you wanted to solve. So last night in our live learning lab, I went through Power Platform and I demonstrated 3 or 4 different models. And just to give you a sense, these models, when you leverage an AI builder tool, this builder tool is like a WYSIWYG environment, right? What you see is what you get. It's a, a palette. Actually, they call it the canvas. You're building a canvas, and in a canvas, if you're an artist, right, you have a canvas and you have a palette, and the palette is where you grab your colors and make your picture. Well, similarly, in the canvas in AI Builder, you would grab a palette of different tools and widgets to create an application, including AI models, and You could then drop them on a freeform surface on a canvas, and you can craft and create applications. Here's what I loved about what we did last night is we talked about two different ways to do that. One, you can create workflows. So let's say, and I'll give you the example we used last night, let's say I am working at a healthcare facility, and every time someone leaves our healthcare facility, we send them an exit survey and they provide feedback. On our facility. Lots of different kinds of feedback. There's questions for staff, there's questions for, um, how did the— how was the facility, how was the doctors and nurses, right? And so all that feedback gets sent to us in a survey. Now it's my job— typically I'd have to read that survey and I would have to distribute that to everybody who cares about that information. That is something that many medical practices do today. Like, they manually triage this type of information, manually assess kind of the survey results. But now we decided instead we were going to leverage an AI model, a categorization model. And so we use text categorization, but it was what's called a custom model, which meant like, obviously, I don't— how would this model know what categories I have? I'm a medical practice. I called what, you know, I basically broke up my organization into 5 different areas, staff, facilities, care services, you know, customer service, whatever it is. So I broke up my organization. There are different teams that do different part of the healthcare experience— doctors, nurses, all that. And so I want— I now want, when that survey data comes in, I want to automatically— when survey data comes in, I want to grab that data and send it directly to the team that needs it the most, right, that it's geared towards. And so this model The first thing I need to do is I actually need to custom train it on my categories, on my teams, on my structure. And so that's what we did. We went in and we created a custom model. So the benefit of custom models is I actually get to feed it organizational data that tells it, here's what's special about me, so that when the model is used, it's actually customized— customizing the result to my organization. Those custom models do take a little time. They do require you to provide the data. In many cases, these are supervised learning models. So for For those of you not familiar with that term and haven't had a chance to go through the AI Leadership Masterclass yet, that— in the masterclass we talk about all the different types of machine learning models, but supervised learning is that type of machine learning model where you have to give it— you have to supervise its education. So you're teaching a machine, here's what, in this case, here's what my organization looks like. You have to teach it what is a type of message that would go to staff? What's a type of message that would go to facilities? What's a type of message that would go to the care services team. And in order to do that, we call it annotation. We're going to go into a dataset. We're going to carve out some data. And we're going to label all of the survey results. We're going to label it care, facilities, a little bit of both, staff, whatever it is. So we label that data. So what supervised learning means is that we take a little part of the data. We label it. And give— those labels become instructions to the machine to say, hey, machine, when you see a message that's kind of like this, go to staff. When it's kind of like this, go to care. When it's kind of like this, go to facilities. So we're training it by telling it specific to my data, here's how I want you to categorize this. What that means is that in supervised learning, we pass that in for training. At inference, or when we actually want to call on that model for testing, we're going to give a different dataset. So now I'm going to off the top of my head write in, I had a great time. The, Doctors were amazing. Nurses took care of me. Thank you so much. I hand that off to the model. The model's never seen that text before. And now based on all the things I told it in the annotation, it's going to say, based on what I know, I think this model does— matches to this tag. It's going to make that assumption. And that assumption, the way that we value or identify the accuracy level of that assumption is through what's called a confidence interval or a confidence value. And that number is a percentage. So typically, I will give it this new sentence. It's going to go and look at all the other sentences and go, hmm, I think this is staff. And I'm about 90% sure it's staff based on what it knows. So it's going to give me back this confidence. And now, most importantly, we as the humans have to decide what does good look like. This is literally the job of the human. The machine did its job. It went and said, I think this is good. Goes to staff, but we have to say what percentage, what confidence interval are we willing to take in order to make sure that that happens, right? In order to make sure that that decision can happen autonomously. In healthcare, I mean, in survey data, maybe we're okay with like, uh, 80% match. If I'm doing like diagnosis information, maybe I want 99.9%. Confidence, right? And if it's anything under that, a human needs to see it. But these thresholds are often unspoken but incredibly important to the process of building machine learning, supervised machine learning services. And that's what these applied models give us a chance to do, is to custom— customize a model to our organization data, organizational data. And then we get to tell it what the thresholds are on that confidence interval. How confident do I want that machine to be? Before it sends an email to that team. And in this case, maybe I'll say 80%. So if I get a confidence interval of 80%, which I get in the form of a variable, I can now in my code check, is it 80% or higher? If it is, go ahead to the next step. If it isn't, maybe send it to a triage or send it to a human. But we're looking for 80% of those, of course, to continue on through the process. Now, once we've gotten the model to actually identify the confidence interval and hit the threshold, It's not— that's not enough. I don't want the model just to tell me, oh, this text that I just got in this survey, it goes to staff. I actually want it to send it to the staff. So the next part of that workflow is to actually go and execute a request to inside of AI Builder. I can go and say send an email. So now I start with get the information from a survey, which I could use Microsoft Forms to do. Pass that information to the model. The model then identifies the tag, and then based on that tag, I send an email to the appropriate team. So you'll see a little conditional logic in there that says, you know, condition number 1 is, um, is it staff? And if it is staff, then I send it to staff. Condition number 2 is, is it facilities or is it care services? And I create— and again, no code is necessary. I just create this all with palette, you know, I draw from the palette and drop it on my canvas. But I do have to configure it. And I was saying this last night, it is so important to realize that even in no-code AI, you still kind of have to know how this stuff works. And so that's why it's important to really understand. Awesome. Yes. Oh, what's up, Lucy? Good to see you. Good to see you. And oh, good. I'm glad. Oh, good. I'm glad. That's great. I'm glad. Some— yes, oftentimes, you know, and this is just one type. That, that's the only sad thing. This is one type of machine learning. But when you're using AI Builder, most of the models are supervised. So when you go to, let's say for example, you are a produce manufacturer and you wanna leverage a model to help you tell if oranges are ripe or, you know, less ripe. And the reason you wanna know that is because if they're super ripe, you're gonna wanna send them locally versus they're not that ripe, they can go further, right? Because If I'm a produce manufacturer and I'm, let's say I'm in Florida, we're making oranges, which actually I don't know if it's the top produce for us, but anyway, I'm making oranges, making oranges, you know what I'm saying? Uh, growing oranges. When I get an orange off the, off the tree or one, you know, however they do that, shake a tree and they fall out. Some of them are more ripe than others. And if I can tell that, I will know which ones do I send a short distance and which ones can bear a longer distance because they're not that ripe and they'll ripen while they get there. That is 100% right now in most organizations done by a human. Like, where a human's like, ripe, not ripe, ripe. Like, you see these, you can go on YouTube and see these, like, manufacturing lines with humans with the, like, shower caps on and, like, picking stuff out of a line. Turns out AI is very good at this. So good, in fact, right? When an AI looks at an orange, it can tell to the date its level of ripeness. And oftentimes it's something that a human eye cannot perceive. It's like indentations in the pores of a navel orange, like very, very specific, but something that's easy for a machine to see, but very difficult for an, an, a human to see. So in those cases, what can we do? We can actually train an AI model, supervised learning. We can label oranges. This one's ripe, this one's not ripe, right? This one's very, very ripe. This one's not ripe at all. This one should be thrown away, right? Like, we can train a model and then we can create a confidence interval that says, if you're 85% sure that this is gonna take 7 to 10 days to ripen, put it on that truck. If you are less than that, put it on this truck, right? And of course, those decisions, those threshold conditional logic statements can get very complex.. But that is the magic, is that humans don't wanna do that work anyway, right? They'd rather be strategically thinking about how do we create more oranges? How do we sell more oranges? How do we get more trucks? How do we distribute this wider? We don't wanna be stuck looking with our eyes at the pores of a navel orange and deciding if it's ripe or not ripe. Like that is not where humans need to be. So now granted, again, lots of people do this work, but I do want to share with you another story, which goes to another model that, uh, that we deployed last night that I absolutely love, and it's about text recognition. So this is another model. This is actually a prebuilt model, meaning that you don't have to train it at all. You don't have to give it a custom dataset. You don't have to annotate. It's already trained to recognize text on images and inside PDFs. It leverages what's called OCR, which is a model all on its own, optical character recognition, and its job is to look at an image and grab the text from it. Here is how it's being used in practice. One of my very favorite use cases, and you'll know why in a second. But one of my favorite use cases, so Goodwill, a company many of you might know, they take donations and also have stores where you can buy renewable— I don't know how you say that— renewable things. Is that right? But buy used items, consignment, but not consignment. Anyway, a thrift store, right, of sorts. I don't know if that's how they like to characterize themselves, but anyway, Goodwill, you can look them up. But what I love about Goodwill is that they employ people of all abilities, like my son. My son has Down syndrome and he would be someone that could be employed at Goodwill. They, they accept, not only accept, they encourage people of all abilities to join the workforce and then they train them and support them and give them a lifestyle. It's amazing. I'm such a fan. However, in the last 3 years, they started to make some really big changes. They decided, number one, they partnered with Azure AI and Azure in general. So they moved their technology and their infrastructure to the cloud. Then they started leveraging AI services. And one of the things that was costing them the most time and money was that with no additional technology involved, what would happen is people would come and drop off their— and I've done this, I drop off bags. I mean, I have 6 kids, so we go through lots of stuff. So I will drop off bags and bags of clothes and shoes and and backpacks and all of this stuff that we no longer use, and I'll drop it off at Goodwill. And now someone 3 years ago, a human, would have to look at every item and figure out what it is, where it goes, how to price it, how to catalog it, how to put it into the system. Like, that was extremely manual. And so what they did is they ended up leveraging a model just like we did last night, a text recognition model. In this case, it was a object detection and text recognition, but it allowed them to take a picture of this, whatever it is, shoe, shoes, backpack, t-shirt, and it would automatically generate metadata that to describe that item, allowing them to reduce the amount of effort that someone would need to take once they get an item to think, what should I name it? How should it be listed? All of this stuff. Like all of the information. And I really, I love this story because what they found was is that it increased the effectiveness of those people doing that work. Oh, I should move my microphone. Sorry. Hi. It increased the effectiveness of people doing that work by 10x. Like they saw an incredible increase in the amount of items that could be cataloged at any time. But what I loved about it is that it also allowed them to increase the effectiveness of the humans that were doing that work. And many of them were cognitively, you know, had challenges cognitively, just like my son does. But now they can work harder, faster, do more. They ended up growing their retail organization, uh, like incred— I think it was something like 37% in a month, uh, 63% over 3 months. Like it was a massive benefit to the organization and they didn't have to let go of their staff, right? In a, in a company that literally prides itself on building people through careers, like you can't use AI as an excuse to get rid of the core value of your organization. So they didn't. They used AI to help everyone be better at their jobs. And I could cry, like that's good stuff. I want that. I want that for companies, not just Goodwill, who's made it their mission to help people of all abilities, but of every company as they're investigating how do we do more. Like when you have employees that are more productive, happier, delivering more value, you get to be a bigger company, a better company. You actually get to hire more people, not lay them off. So take that, all you companies that don't know what you're doing. But hopefully you found that helpful. These models, like, they are available to you for you to use. If you are not part of the AI Leadership Institute, I welcome you. Please come on over. We do a lot for those people who did the live learning last night. They have an opportunity to actually get a certification called the Power Builder, which is the platform Power Apps AI Builder Power User certification, and it gives you an opportunity to actually demonstrate your skills in this area. And you'll get a guide for how to put that on your LinkedIn, um, so that you can actually be like, look, I'm a doer. I'm not just a talker. I don't just, you know, read about this. I don't just watch it on YouTube. I've put my fingers on a keyboard and I've built something. And that is what will distinguish you in the next 3 years, is that there are a lot of people talking and not as many people doing. As a matter of fact, it's about 10% of people actually do this work, implement, like solve real problems. So be part of that 10%. And join us. So that is really exciting. And that was really fun last night. As you could probably tell by my voice, I'm a little under the weather. And so I was a little bit— I was sick last night and everyone just hung in there with me. So thank you all for that. It was amazing. Yeah, thank you. Thank you. Absolutely. The people-centered approach, 100%. In a world where we have AI that amplifies human ingenuity, literally, Like, what are we doing right now? We're laying off more people than we've ever laid off ever. But that's okay, because I have a vision in my mind that we are in a world where we are moving towards this kind of gig economy, that it's— companies are going to need less employees, but they aren't going to need less work. They're going to need specialists. They're going to need people who understand this problem. And here's the good part. Most of us don't want to be cogs in other people's machines. Most of us actually would like a little bit of freedom, be able to deliver something very quickly, get paid to do that delivery, and then have the freedom to use our extra time. Here's what happens, could happen, you know, in the future. One, uh, one perspective is that like, just imagine we get AI as an employee, we start using it, we start cutting time. Right? Uh, one of the examples, I was just looking up, um, statistics from one of my clients cuz I like data. Um, 97% of their time was saved in meeting, in, in between meeting prep, between prepping to get into a meeting and prepping after a meeting, right? Like taking notes, action items, all of that. 97% of the time they used to spend doing that was freed up when they leveraged an AI system similar to the ones we got here. I don't know how many of you actually use these bots, but we got a few bots here. And when you have a bot that's taking notes for you, some of them are not great, by the way. Some of them are better than others. I leverage a platform called— who I was going to call them their old name, Limitless. And Limitless, I love Limitless because of two things. It doesn't just take notes. Number one, it takes notes on my own machine, in my own heart, on my own hardware. So I own that. Um, it takes notes of the meetings, of what I say in those meetings, of my action items. But what I really like about it is it connects to my Google account and it looks back through all of my email and says, based on who's in this meeting, here are the conversations you've had. So before my meeting, I can get caught up on whatever that thread was that we are continuing in this conversation. And when I go there and there's prep goes and searches and there's nothing in there, I'm like, what are we doing? Why, why are we meeting if we've got no, like, why am I gonna spend this time? It has given me so much more freedom because I now have the information that's necessary to evaluate how I spend my time in those meetings. So that prep is just as important as the meeting summarization and action items that happen at the end. I, I mean, search for that. I use, like I said, Rewind. I think it's like Rewind AI or something like that. That. Oh my gosh, amazing. Yes, absolutely. For those of you, I just got a question in the, in the chat about becoming a builder. And this is at the core of what we do here, right? I love AI. We love AI, but we build AI, right? We're doers, not talkers. We learn by doing. We put our fingers on a keyboard, we build something awesome. And so the first thing we did last year, it became something that was super successful. We ran it every single month until December, um, until I recorded it and I was like, oh, we could just run it on autopilot now. Um, but this course, it's called AI Bot Building and it teaches you how to build a responsible AI system end to end in less than 4 hours. And when you build this system, basically what it teaches you is number one, every AI system needs to have what's called an AI safety system or this harness for creating responsible AI systems. That means that you need to think about the human AI experience. How are humans and AI gonna use this technology together? How are you gonna get the job done cooperatively? And then how are you going to train the model? How are you going to use prompt engineering and system prompting to train the model? How are you gonna pick the right model? Model selection and evaluation. How are you gonna make sure your model isn't gonna accidentally hurt someone or do something it's not supposed to? And now more than ever, What's the infrastructure that that model sits on? How does that infrastructure support the model development? So those 4 things, like most people never even think through these 4 things. But then in this AI Bot Builder course, we end up taking you through the process of actually building your own model. And it's super fun. Those of you who have done it, you can chime into the chat, um, that you are a Bot Builder. Uh, and at the end of the course, you can get a Bot Builder certification. That actually kind of celebrates the fact that you've built a real system end to end. And it takes about 4 hours to do. Some of us have done it in less time. I think I'm now down to about an hour. Um, but, uh, oh yes. So thank you, Sherilyn. But actually, if you want, that is gonna take you to a very expensive version of this course. Um, so now there's a discount happening right now that if you, um, If you— I'll go ahead and put it in the chat. If you buy, if you preorder my book, you get access to the course for free and that'll give you access right away. So that might be a better option. $35 for a $600-something course, like, go for it. So that is www.upskilling.ai/book. There we go. So you could do that. So if you are interested, you can get the book, you preorder my book, which comes out April 1st. You'll be one of the first to get the book and then you'll also get access to the course. And we also, in that process, you'll get invited to join our mentorship, which is, will be starting in April, April 1st, the day my book launches. We have a new mentorship program that's starting. This is much more kind of specific. Every single month, I have a new topic that we'll be diving into. And those of you in the AI Leadership Institute, you'll see all of this. You'll all be part of that program. But this was to go for people and organizations that hadn't heard of me yet but had uncovered my book. And so, we created a special workshop for that. And then of course, if you wanna have one-on-one time with me or with my team, there's an option for you to just get a one-on-one call with us as well. So enjoy. But yeah, I 100% agree with you though. The number one thing you wanna do is build something. And, and I don't mean like go to ChatGPT and be like, hey, you know, help me decide on a meal. I think what we did last night was really powerful in the sense that we were like, how do we solve the, the challenges of businesses regardless of their size? And typically that means we need to niche down into a department like HR or finance or supply chain or delivery or sales or marketing. How do we do that? I recently, actually just, uh, yesterday was working with one of my new marketing clients, a client that wants me to help them with their marketing. And in this effort, right, in this 2-hour call that we had, I showed them a single prompt that we could use to generate 30 emails for them, an email that would go out to their clients every day for 30 days. And they were, they were just blown away. I mean, these are the types of things like most companies don't even know how to do it manually, let alone automate it with AI. So these are the types of things, if you are a marketer, if you're a salesperson, if you're a UX designer, no matter what your personal skill set is, if you partner with AI, you can now deliver even more value to the people that you're serving than ever before. Oh yeah, great, Roxanna, thank you for that. Actually, just ping me. Or actually, Sherilyn, just take down Roxanna's number, or number, Roxanna Barcelona's name, and we'll add you to the course. Absolutely. Yeah, you're going to get an email for all those who bought on Amazon. Typically, uh, you're part of my AI Insiders, so you're all going to get access, of course, to the course as well. Many of you already took it, so, um, but yeah, thank you. Great, great point. Great point. Amazing. Okay, awesome. Um, cool. So let's see, what else did I want to talk to you all about? So that was last night. We did a lot of cool stuff. Um, amazing. Yay! Okay, good. Uh, awesome. Yes, you will see it. So it is Right now, uh, we don't let everyone in, so we kind of, because we've got a bunch of bots in the world, as you probably know. So we manually kind of look at you and go, are you the right person? And we, we put you in the course. So that'll happen right after this, uh, right after this session. That would be amazing. And just email me if you don't see it, but it is in the school. So in I Love AI, you're gonna get access to a private course that only book buyers get access to. And so you'll see, uh, a bunch of people in there. And commenting on the course as you go through it. And we'll also remember these classes and specifically our AI Leadership Institute class. These, these calls are meant to answer your questions about these types of implementations. So my goal is that you will invest in learning how to build these systems, and then you'll come here and ask questions about them if you have any. That way you don't get stuck. One of my biggest concerns often when people start learning new things is that, and this was my challenge as well, is I would learn something and then I would get stuck and no one around me was doing that thing. So I was kind of like literally stuck. Like there was, I'd send a message to the GitHub repo owner and be like, can you help me? They don't respond. And then I would send a message to the product owner and I wouldn't get, I would be just stuck. Now you have a bunch of people, 1,400 people that are on this journey with you. And so we can help you all as you go through. You can comment in the community and we will answer questions that you all have. Amazing. Thank you for sending that. Roxanna, super helpful. Thank you. Thank you. Thank you. Oh my gosh, me too. I'm so excited. So hopefully between now and next week, you will have a chance to play around with that and see it. OK, now I do want to give you a whirlwind tour of where Noelle is going to be in case you happen to be in a city near me and we can get together. Because at the recent event, I got a chance to meet up with a couple of our I Love AI people, as I mentioned earlier, and it was so fun. So if I'm coming to a city near you, let's make it happen. Eventually, I will become so proactive that I'll set up a meetup in that city on the night before. But I'm not, I'm not there yet. But that, that's coming. That's coming this year. Also this summer, we are going to be hosting an AI Leadership Summit in person in the middle of the country. I chose the middle of the country somewhere near Iowa. I chose the middle of the country because I wanted to create equity for all of us. Like, I'm coming from Miami. Some people are coming from California, Chicago, Boston. So It'll give us a chance to all get together. Those details are coming soon. I just want you to know that that'll be happening over the summer. It would be awesome. I am coming to New York. I just got an opportunity. Someone sent me a message for an event in New York. I think it's in April. I don't have it locked in, but as soon as they give me confirmation, I will let you know, um, and we can meet. So that would be amazing. Uh, so to— on Sunday, which is my birthday, yay! Uh, on Sunday I will be headed over to Lisbon, Portugal for a conference. I'll be speaking at Boston Consulting Group's company event. Then after that, I'll be going to Vegas for an event called GrowthCon where I will meet with a bunch of people and a bunch of my clients will be attending and that'll be fun. And then after that, I'm going to be flying to Seattle to speak at the MVP Summit, Microsoft MVP Summit. It's exclusively for MVPs. It's where we get to go and share what's working for us, but it's It's also where we get to go and talk about, and hear from Microsoft's AI product teams. And that's my favorite part. I do like presenting, but I love hearing like what they're thinking, what's next on the list. A lot of it I'm under NDA of everything that I get shown there. However, I am first, like as soon as we have a whole— it's called a book of business or book of events, book of engagement. And that book of engagement, it lets us know the dates that we can talk to you about it. And so I try to be first, like as soon as something comes out, I try to put it on the calendar and talk to you all about it. That way you're getting the latest and greatest information from Microsoft. I don't have that same relationship with Amazon, though I do have a lot of good friends there, so they tend to tell me new things as well. Oh, what? Yes, my sweater. I will tell you how to get it. Yay. Thank you for asking. You absolutely can. And you could get new I Love AI. I will put that link in the chat. I'll have to find the link in just a second, but thank you for asking. Yes, this is one of the hoodies. We have the big— the ones without the hoodie, the zipper. I love zippered hoodies. But also, of course, being part of the AI Leadership Institute, it's, you know, it's super fun and I like to wear swag. Oh yeah, and it's tax deductible. So that's why I almost always wear swag because it's tax deductible for me. So that's something to— thank you. Thank you, Tracy. Thank you very much. Yes, so it'll be fun. A birthday weekend. Uh, after Vegas next week, I will— oh yeah, I'll be at MVP Summit, and then that's on Monday. Then at the evening of the 24th, I'm headed to Chicago. And so if you're in Chicago, I will be headed there for the Responsible AI Forum, uh, hosted by AnitaBee, the organization that also hosts the Grace Hopper Conference. It is one of their first Responsible AI Forums. It's a huge event. You're welcome to join me there. Um, we are— I'm gonna be doing the keynote on Tuesday the 25th. And then I'm also hosting a workshop for part— for attendees, um, on building responsible AI at scale. So I will be taking baby tigers, like little baby tig— not literally, um, but we'll be talking about baby tiger use cases, like use cases that started off innocently and then ended up hurting people. Um, and they're real use cases, but we're gonna walk through the ideation process. And my hope is that it will generate empathy for people to understand that it is a very hard job to build AI systems. Like, it's not just clicking a magic button or typing in words into a prompt. Like, when you start building it, unless you're really thinking about, like, a huge holistic view of all the things that could possibly happen, it's often the case that these models will grow up and do bad things, things you never intended. You see it all the time on the news, right? Like airlines making refund policies or, um, uh, auto dealership, uh, selling cars for a dollar. Like there's all of these AI systems that grew up and did bad things. Um, the nutritionist, right? App that told someone with a, uh, an eating disorder to diet. Like all of these, the data scientists that created these algorithms, the AI engineers that implemented these applications, none of them intended anyone to ever get hurt. Or for the business to lose money, but it happens. And so I like to, in these workshops, I like to present these use cases, but on day 1, I call it day 1 versus day 2. On day 1, here's what we're going to do. And then as we ideate on how to build a safe and responsible system for that use case, and I have education and autonomous driving, as we think about that, then I, I throw in like, oh, you just got 500 million users and, uh, now you are getting users you never considered. What happens then? How do you mitigate the risk of that situation? And so it's really exciting, really fun. I hope if you are in Chicago or near Chicago, I hope you'll come and visit me. So that closes out the week. And then finally, last thing on LinkedIn. Oh great. Oh yes. Oh yes. Awesome, Kirk. Amazing. Of course you absolutely can. You just have to let us know, Kirk. So yeah, if you already took the course and you want to give it away, you're welcome to do that. You'll just have to send a note to me, uh, noelle@noelle.ai, um, or send a message to Sherilyn on school and just give the, the email address you wanna, uh, give access to. So absolutely. Um, but yeah, it's good, good point. So on the, the website, uh, for the book, um, when you preorder the book, it also has an opportunity to get really, um, it was the genesis of the book itself. Now I'll tell you how I wrote the book. I don't really like writing. I'm just very candid about that. It's not a skill set that I am accelerating in. I don't enjoy it very much. It was a chore. Lots of things I love, like this, hanging out with you all. I love this part. Writing is not one of those things. But I got an idea. And actually, I think I heard another author say it that said, the only way to get people who will never meet me to know that I do things that I think will impact them, the only way to impact people that will never meet me is by writing a book. And so that's what I ended up doing. But the good news is, is that I did a keynote speech and Wiley, someone at Wiley saw me do that speech and they said, just turn your speech into a keynote. And I was like, wait, wait, wait, turn my talking into text? I can do that. It turned out not to be that easy, but that's how it started. Then I was during that time, so I'm writing this book. During that time I was implementing a lot of responsible AI, like enterprise scale projects. And it started off really with the foundation, like how to build a center of excellence, how to build a responsible AI board. And so these things became very specific checklists that I kind of went through. And that's what that guide is. It's a series of all the checklists that I use to create inclusive innovation teams inside enterprise companies over the last 3 to 5 years. So some of it's like new AI and some of it's kind of old AI, like how do you just AI in general as opposed to GenAI specifically. But it is, I mean, I, I love it. I mean, I wrote it, so I guess it's a little bit biased, but I've gotten lots of great feedback on it. And people had to pay like $25K to get that because I did, came in and did it personally. So I just threw it in to give you something else that's a little bit secretive and that I don't have available publicly on the internet anywhere because it's part of what I sell to my clients. And it was like, like, I think I put it on for $15 or something crazy. Um, and so hopefully you'll find it valuable, but thank you for that, and thank you for asking. And that definitely— Kirk, let me know, I would love that. Um, that's cool. Uh, and that's a great idea. I should do like give one, get one, right? If you buy a book, um, maybe you get to gift a course too. That's a great idea. Good funnel idea. I love it. Um, amazing. All right, let's see, what else did I want to share with you all? Uh, oh yeah, Yeah, we're at time, I think. Oh no, not yet. We got like 9 minutes. Okay. So one of the things I wanted to share with you is something I think I mentioned it earlier this week or last week, but I want to reiterate it and I'm going to go and show it to you because I'm passionate about it. Obviously, I want you all to learn by doing, implement, and some of these organizations are not implementers, they're talkers, and that's okay. Like, it's a healthy balance, right? You know, you can come to me if you want to build stuff. And I will give you— so one of the, the advantages of joining the AI Leadership Institute, which is like $147 a month, which seems like a lot until you realize like what you get. So you get access to my tenant and free access to Power Builder, Power Apps, AI Builder, Power Platform, Fabric, and Copilot Studio. So I give you a license to those platforms so you can start building on my tenant. Planet. And it's like that alone is worth $100 a month. But, but on top of that, of course, we do hand, you know, hands-on activities every single month. And then we also do these boot camps. So everything that I do, literally everything, like a class on Maven, which I'll share with you in just a second, workshop for a community event. If you are in the AI Leadership Institute, you get access to all of those things. So I taught a class on Maven. On how to be an AI solutions architect. I taught a workshop on how to build an AI consultancy in 2025. I, I have a bunch of content. And so all of that is accessible to you when you become a member. So it's worthwhile if you're on a journey to actually become a doer in this technology. And you don't have to be. Lots of us are listeners and kind of talkers and like, you know, we're, we're kind of adjacent to the tech, but we've got some incredible people in this community that have dove in in and they maybe didn't think of themselves as practitioners before and now they're all in. Like, now they're like, oh my gosh, I could build this. It's no code. You don't have to learn engineering. You just have to learn how to ideate and come up with a good idea. So let me share my screen really quick and share with you. I'm on Google, so I'm going to go there. Yeah, there we go. Share. Do do do do do. Uh oh, where is— oh, there we go. We got it. All right, so So here's what we did last night. I'm just gonna— might log me in. Okay. So these are all the models that we actually did hands-on activities with last night, and that was super fun. So we went in, we built custom model, we leveraged prebuilt models, we consume models. Really fun. What I wanted to share with you now is what's called the Global AI Community. Oh, I'll— oh, look, it came right up. Someone else is searching for this too. So this Global AI Community, it hosts every year. It's during a certain period. You can't do it before or after, during a certain period, which is right now through the first week of April. They launch what's called a Global AI Bootcamp, and I host one every year in Miami. However, because all of you might not have a Global AI Bootcamp where you are— oh, let me go ahead and copy this link for you. Oh dear. Control+C. Of course, yes, sure, you can pause whenever you want. You let me know, just send me a message. And it's, it's, uh, the, the AI Leadership Institute. And I will tell you, we've had people who joined and then, um, we're in a period of transition. And many of us know when you're in transition, like, money is, you know, like, you're trying to do everything you can to, like, be responsible with that. And while I think it's an incredible investment, like, you, you do you. But then they, they, they like stop their subscription and then they got a job based on the stuff that they learned. And then, and it's super cool to know you are surrounded by people on the same path as you, right? So it's exciting and motivating. So that puts you in a good mind space to get a new job. Then you get a new job and then you come back. So absolutely, absolutely. We're very flexible from that perspective. And we've had a lot of people leave and come back for exactly that reason. Things got tough, which it does. Life happens. But then they came back because they really enjoy being here and learning with us, which is fun. All right. Did I put that in there? Yeah, there I did. Okay. So that link will take you here. I am hosting one. So I'm hosting the Miami chapter, Global AI Bootcamp, but it's virtual. So you'll see it on our calendar. It's the 28th. It's actually going to be in the afternoon. Right now, I think I blocked the whole day from like 10 to 4 Eastern time, but it'll be in the afternoon, probably from 1 to 4. We are— our AI Leadership Institute members are going to, they've submitted talks. So I just gave them an opportunity to submit, call for papers, right? So they're going to submit talks to teach that day. I'll be teaching, of course, myself. I'm focused a lot right now on how to help small business owners use AI to grow a business and generate enough revenue so they don't have to go back to a corporate job if they don't want to. Not— I have nothing against corporate America. I was there for a very long time. But if you have it in your heart to serve smaller companies as a consultant or build a, a business or create an agency, like, I wanna help you do that because I think it's equally gratifying. And for some of us, it's even more so. So I wanna make sure, that, that I'm gonna be teaching about that for those of you who are interested in it. But we'll have, conversations about all of it. But I just wanted to introduce you to this because the Global AI Community has local chapters all over the place. There's 80 local chapters— 81 technically, because mine for some reason is not listed. Um, but I showed everyone on the previous call it is there, but I don't know why it's not on the list. So I'll talk to them. But Paris— I mean, there's a lot, uh, globally, um, but definitely some, uh, locally as well. There's one in Houston, uh, hopefully one soon in Atlanta. Um, so we'll see how that goes. Uh, but if you're— you're welcome to either, number one, you can join a community near you, so you can go in person and hang out with people. Or— and, um, oh, let me see if I can find them really quick. Oh, I don't know if I can. I'm gonna try, try and find my— what people get. Let's see if they're in here. It might be. I don't know. I don't think I have them, but I have like all these stickers that they gave me for the boot camp. I have. They're around here somewhere, but I can't find them right now. But anyway, so you got like a sticker when you come in person, you get like, you know, access to swag and all that for joining. But like I said, it's only held during a certain period of time. You can see it here. It's the big end of March, beginning of April. It's happening right now across the world. Mine is April 28th. I have another one happening in person only on Monday the 31st. So, and you can see all the local areas. So find one near you, hang out with people. These are good people. These are my people. These are your people. So it's really fun. I love going in person and hanging out with people that are also doers and builders. And that's what this community represents. All right. So I wanted to show you that. And then the last thing is our new Maven course. So while— oh, I don't know if I could share it with you here. Maven.com. Let's see what happens. Oh good, oh good, analog man. Awesome. So we created a new course based on content that Microsoft launched. Ooh, what? We have a new platform. I'm going to take you all to the landing page so you can see what it looks like. But 8 of us, or 7 of us, got together and created a new course based on Microsoft's content, Agentic AI. So it's an Agentic AI for Beginners course, and we are working our our tails off to understand the framework, the technology, so that we can help you all go through the process of learning this material. So we're hosting it on Maven because most people had not done that before, and I thought it was a rite of passage. So host on Maven, but you'll see there's a bunch of us teaching it. It will have recorded training, and then weekly, every Saturday starting next week, every single Saturday, we're going to have content either next week or a week after, we'll see. But as a result, I created a book called The 16 Laws of Responsible Agentic Architectures. And it's, it's pretty good. I keep adding to it cuz I can't let it go. I keep adding more and more stuff. So I'm eventually, I'm just gonna call it done. Done is better than perfect and get it out there. But it is a, an opportunity for you to learn a little bit about what does, how does agentic apply to the knowledge that you're learning both with me and in this community, but also just on your own. And how do you make sure that you're aligning yourself with this agentic world that is fast becoming the norm, in across organizations, around the world, really. So it's going to have, yeah, like 30 lessons because all of the— every week there's 5 or 6 lessons. You'll get lifetime access once you sign up. It's going to be about $100 because Maven doesn't allow us to do things for free. I think the price is set at $197 or something like that. But we're going to provide a coupon. And again, if you're an AI Leadership Institute member, you get it for free. It's just part of your membership. Membership. But if you just want to take this and you don't want to be a membership member at all, you can just take it for like $100. I always tell people, like, you know, eat out less, maybe twice this month, um, or maybe 3 times depending on what you eat, uh, but eat out less and take an Agentic AI for Beginners course. This will be really, really cool. I'm very excited about the content, um, and just so excited that it is a culmination of not just me— like, I'm definitely doing my part, but It's all members of the AI Leadership Institute that have come together to create this, and we all met through the institute. So it's just really exciting, very empowering. And, uh, I just encourage you to, to know that you are not alone on this journey. Learning AI is way more fun when you learn it with people who are excited about this technology, not just for what it can do for businesses, what it can do for our careers, but also for what it can do for humanity and how we can actually make the world better. Be purple people first. Right? Be people-forward. Be interested in augmenting the ingenuity of humanity and making humanity better. Goodness knows we can use it these days. So I think it'll be amazing. So thank you. I'm going to stop sharing. And there is a question. If there's any questions, throw them in the Q&A. I'll stay for a few minutes. I'm going to just check to see if there's a question. Oh my gosh, this is a great question. Thank you, Art. I think— I hope I'm saying that correctly. But yes, great question. So the question here is, which tool would you recommend to build PowerPoint presentations for business purposes, like building strategy presentations? I'm actually going to share my screen again and show you what I use. All right, so I think it's here. Share. Where is that? Oh yeah, that's here. Okay, so this I'm gonna go to a new tab. That is a Loom. So here's a fun thing while I'm pulling this up. There's a couple tools that I use. One is Beautiful AI. Beautiful AI, and I'll just show you why I like it so much, is that it creates PowerPoint specifically. You asked about PowerPoint. It creates PowerPoint presentations. You can create it with Gen AI. So I'll take you into my deck that I use for customer presentations. And this is what I use like right now for customers. And it allows you to talk, like, use words in generative AI to draft out these guidelines. Where we lose our lives in PowerPoint is like moving things around and organizing things. So it has lots of smart tools. For example, let's move this over. If I add a new slide, like, there's all these smart slides, right? Like lists. Oh, I shouldn't have done that. But there's lists and text boxes, and you can see, like, all of these templates. There's tons of ideas. Like, this is where I think the huge value is, is I can use my words to take a template like this and then customize it to what I need it for, where that is a much longer journey in PowerPoint. Point. Even with Designer, the Designer did help me out yesterday on PowerPoint. That was pretty good. So this is what I personally use. I'm gonna go back. And so I can just give you a sense of what it looks like. So this is just an image. But like, this was created. You could see these are all the different layouts that I use. This was automatic, like this, you know, kind of like tiered. I mean, all of this, it just took so much less time than fiddling around with like smart graphics and all of that, and like orientation and spacing between things. Like, I don't have to do none of that. This is our course, all of our courses that we talk about when we pitch clients, our process. Yeah, I think this process, I manually built that, but I could have probably had AI do it for me better. But yeah, so this is like the structure that I used, and it was very simple, very easy. But what I liked most about it is that it gave me the ability to export it, right? So when I choose to export the presentation, I can export it as a PowerPoint that's editable, which means all these pieces become movable. I don't ever wanna do that. I've done it once. I'm like, you know what? I never need to do that again. But I could also export the entire thing as a PowerPoint where the thing it created for me becomes an image. And that I have done, and I send it to, like, when I do a keynote, I'll send them a PowerPoint version of my deck. Google Slides, which is new actually, a PDF, or present it online with an offline player. So that's one option. The second option is Gamma. And I use Gamma, probably before I only use Beautiful. Now I use Gamma a little bit more. So here you can see like the 16 Laws of Responsible Agentic AI. Let's see what I did here. There. Yeah. So I asked it a question and it created for me. In this case, I think I asked it to create me a white paper. So it formatted it this way. But its job, Gamma's job, is to create artifacts for you. Let me just say add new and I'll show you. Like, you can generate either— let me go, let's paste in some text. Oh boy, I actually have to paste in text. That's not a good idea. Let's just generate. From one line prompt. Here we go. So I can generate a presentation, a website, pretty cool, a document, which is what I just showed you, or social carousels. Now carousels take a long time. So to create, take content that you have in your mind and turn it into a social carousel, like, I'm all about that. I ended up investing in this product. Like, I was like, I'm in, I'm in. This, that alone was worth the investment. But this is also what I end up using. Oh yeah, exactly. So that's a good question. And Andrea, yeah, I agree with you. I, I actually now have pivoted towards using Gamma more often. But that's— I don't know, no reason. Like, I could just use this. I have found a little bit of friction in the creation of these pages. But what I've under— what I now realize about myself is I've evolved my prompting strategy where I want one-shot prompts, right? I wanna say it and you build it. Like, I don't want to go back and forth and edit things and add slides. I want it to be done the first time. And I find Gamma is much better at my one-shot prompting. Zero-shot learning is what it's called in machine learning. Like, I just wanna ask and it show up. Thank you, Sherilyn, for throwing those in. However, I built this presentation a while back and I like it. And so I, I come in here every so often. And no, Beautiful AI is evolving too. Like, they're leveraging their own models. And so they're probably getting better. I only give you 2 options for those of you who wanna kind of play. A, because I think it's important to have options. But honestly, I feel like they both do very similar work for me. Now, I guess the other reason why I would go to Beautiful as opposed to Gamma is sometimes I have trouble getting Gamma to put it in a really good PowerPoint format. And that PowerPoint format is a struggle in Gamma. It is for me. Some of you might have some special tips and I will take them. But it's been a struggle to, like, turn it into a PowerPoint where I have to hand it to another person, like, if I'm doing a keynote for an event and I have to give it to the AV team. It's just too much friction. So with Gamma, it's— or with Beautiful AI, it's very simple. With this, it's a little bit more nuanced. Let me go back and see if I can show you any of the presentations that I created. Here. Oh yeah. Actually, I don't even have to show you it here. This is the presentation in Gamma that I created for our challenge that we did the other day. Right. So you can see here, this is a Gamma presentation. When you're on the free version, it only lets you do 9 cards. I'm not anymore. But so that was the limitation. But when you hit present and again, if you don't, if you're just using it for business presentations, this is pretty cool. All of these images were generated by Gamma. Beautiful doesn't do as good a job identifying, like, doing that work. But I'll just flip through it really quick. Like, it's got a little animation in it. That's not automatic in Beautiful. Like, I kind of liked it. It was pretty cool, right? These are all the topics that I went through during that challenge that we did. Here are the creativity tools. I mentioned Gamma there too. Here are the tools about AI-driven decision making. This is good. I remember this. Do you all remember the challenge? I do another challenge. Maybe I should do an agentic AI challenge anyway. How I imagine a future with AI stress-free, creative. Yes, I see it. And then, of course, I promoted my own book at the end. Let me see. Oh, yeah, that's it. That's it. So kind of neat. Yeah, I'm not sure. Non, that's a good question. I don't know if I can. I would— I don't think so. I have no idea. You maybe could. Oh, yeah. Let me see. Let me see if I go into one of these. So this is that presentation. Let me go in and see. Do, do, do share. Export. Look, export PowerPoint. But I haven't done this. So, yeah, and you can see Google Slides is new. So the Gamma is just a little bit behind where Beautiful was from a feature perspective. But I do like that— remember, it could do social cards, those social care cells. It can post directly. From the, those social care, care, or social carousel cards post directly to LinkedIn. Like, let's go. Like, let's do that. Anyway, you could also embed it right on your website. Here is another fun fact, a reason why I like this. If you embed it on your website, and this is what I encourage you to do if you're doing sales presentations and you're doing, of course, if you're your own company and you're not battling with like a marketing team and what they're doing, But if you have your own website, embed your presentation on your website. So when you're showing it to your customers, you're going to your website and you're flipping through your presentation on your website. And that way you're like, and if you have any questions or wanna see this, just head to my website. You create this stickiness factor of them coming back. And this is a huge announcement for me. You guys are probably gonna be like, really? That's all? But I hit 10,000 monthly people coming to my website this week. So that was really big for me. Like 10,000 people come to my website every single month. Huge. Anyway, I was super excited. Thank you, Cynthia. Yes, exactly. Yes, put it on your website. So now I drive traffic to my website because I do my presentations there. And, and I'll have a different— from an SEO perspective, or what we're calling AEO, AI Engine Optimization. So in order to show up on ChatGPT, you want to create kind of like every business presentation you do. Let's say you do one for healthcare, one for finance, one for retail. You create a different page for that. And that way SEO or an AI system that's crawling your site will go, oh, Noelle's a specialist in this area. And it's very, very powerful from, from that perspective. All these little extra tips, digital marketing at its finest. All right, let me see. I think I answered the question. Yes, hopefully that was a good— that was great, great, great question. Thank you. Art, you are not alone in asking. Yeah, exactly. Thank you, Kirk. Yeah, if you have— if you don't— 100%. If you don't have your own website, you don't actually even need to build a website, right? That's kind of like circa 2006 or whatever. You absolutely could just build a YouTube channel and generate all of your traffic to YouTube. 100%. That's a great idea. I guess it kind of depends if you're, you know, a video content person. I do think today more— that's why I have you all build videos to get certifications at AI Leadership Institute. You, you really have to kind of get over the, the whole I don't wanna be seen on social thing. Padmee, one of our AI Leadership Institute members, she had to get over it, and it has like her, like, everything's bigger, better, brighter as a result. But that's if you wanna, you know, connect with people and have people transact with you. Like they want to know who you are now more than ever. You can't just be a faceless company. People are going to go in first, they won't know if you're real. They'll think you're an AI system. Like, ask me how I know. People all the time, they're like, are you real? Are you really here? Are you really responding to this? I'm like, no, no, for real, it's me. Like, it's, it's me. Um, but the other thing is that you could also create a deepfake of yourself. Leverage a tool like HeyGen, create a deepfake of yourself, and create generate content. I have found out that users know the difference. And so it's better than nothing. It's better than you not being in the world. But I would just, I don't know, figure, find a problem you care enough about to get over your concern about showing your face. Like, there is a problem out there that you care enough about that you're gonna be like, you know what, I need to help people solve this problem. That's how I am. Like, I went, I, I tried to go dark and not be like a social presence. And I realized people will not do business with people who are dark like that, who they can't see, that they can't interact with, that they can't look up. And that changed everything about my business and my success. So I encourage you to do the same. Oh, cool. Yeah. Kirk also mentioned converting PowerPoint to videos. 100%. That's a great idea. Now, do you do that in PowerPoint? Like, do you like turn it into— I think there was an option in there to like turn it into automated Somehow. Yeah, yeah, yeah. Save it as MP4. What? That's gold right there. 100%. Save it as an MP4. Do you talk over it? Like, do you add your own narration? Because I know it has that option too. That's so good. Pretty cool. I love it. Amazing. And of course, if you do this and you want to figure out ways to repurpose it, I would like to give you more tools. But, you know, that's what we're here for. But it's called, maybe I'll pull it up really quick. Repurpose.io. What? I love it. Yes, TTS, my friend. So TTS is text-to-speech. So leveraging, you know, you use ChatGPT to write the script and then you use text-to-speech to generate the audio. And like I said, I like HeyGen because you can actually create a neural voice of yourself. But also inside AI Foundry at Microsoft, you can use Voice Studio and create neural TTS model that is your voice. Also very— I mean, that's really cool because you can now do unlimited podcasts without ever having to say a word, which is pretty awesome. So Repurpose.io, they are— what, what? Repurpose TikToks and Instagram Reels? What? Let's go. So this is an opportunity for you to take whatever content you're building and repurpose it across multiple multiple platforms, giving you the impression that you are omnipresent when actually you only built one thing for one audience. It's a great, great tool for those of us looking to go big and wide as far as our audience is concerned. Yeah, I, I mean, you never know though, right? Like, Nan, I think you, you probably know as well, like, over time it's a, it's one of these waves, right? Like where we, yeah, we start off with like 200,000 of these little apps, then they kind of get aggregated into a single one, and then the world's like, no, no, no, I want them separate again. So we kind of go through these waves, but we're right now in that early wave of lots of different AI models. And yes, you're gonna see, I mean, even now PowerPoint, right? PowerPoint is adding more features to Designer so that we don't have to leave PowerPoint to get these types of capabilities. So I think you're right. We're moving towards, hey, get one platform that does a bunch of these things as opposed to you know, like death by a thousand cuts, like literally $20 here, $30 here, $99 here, like to aggregate all this together. I'm willing to do that because it's my business and my clients pay me to kind of push the edge. But what I do for my marketing clients is I give them a platform that embeds these AI services inside of it, and that way they're using a model to help with email generation and copywriting and image creation and all of that. That's inside the tool. They also have a voice-enabled AI concierge that answers their phones and responds in SMS. Like, all of that is inside a single tool. That makes it much simpler. So I think you're right. The tech— we will definitely get there. But what I found is like we go through this like wave where we, we like it all little pieces and then we want it all together. And then when we get it all together, we're like, no, no, I really wanna break it apart. Kinda like cable, right? Like, I remember the days where I was like, I want cable à la carte. I only want these 5 channels. Why do I have to pay for 200? And now I'm paying for more than 5 channels independently, and it's more than I was paying for cable. So now I'm like, actually, can we just go back to the other thing that I was paying? You know, anyway, it's really funny. Oh my gosh, yeah, exactly, exactly. Bluetooth is a great example, a great example where, yeah, it used to be special, and now it's like, Like, your car doesn't have Bluetooth. My car, I don't think my— my— yeah, my car doesn't have Bluetooth. I have one of those old-fashioned, like, what are they called? Like the AV, you know, where you put the microphone in. Anyway, it's like a cable. What is that? Auxiliary cord. I have an auxiliary cord that connects my phone to my car. No Bluetooth. Don't me about it. My husband is like, we need a new car. And I'm like, it works fine. I'm that person. I'm like, it works fine. It's 2017. It works great. Wait, only got 100,000 miles on it. Like, I could drive this for another 100,000 miles. Anyway, he's ready for me to get out of it. Anyway, well, it's been my pleasure to be with you today. Thank you all so much for coming. I hope it was entertaining, also informational, gave you a sense of some new tools to try out, and also encouraged you to join some of these classes so that you can uplevel your skills in the world of AI. We are here every Friday. If you want to join us live, for those of you who are connected and seeing us on LinkedIn or YouTube or Facebook, Facebook, you can join us live anytime you want. You can be in this room, the room where it happens, on Zoom, where you can ask questions, comment, and see all these links and get access to them live. You can do that by simply joining the I Love AI community, and you can do that by just going to Noelle, Noelle.ai, and it'll give you easy access to that community. If you buy my book anytime recently, you'll also get invited to join the community as well. But if you're you're out there and you're like, how do I get in the Zoom where all those cool people are? That— the best way to do that is just go to Noelle Russell Russell AI and, uh, head on over to us. So thank you all so much for being here. It's gonna be a crazy couple weeks. We are in short final, uh, as I like to say, for our book launch. Um, that happens on April 1st. And so it's all hands on deck. Um, if you know somebody who really wants to get educated and responsible AI at Scale, please share with them the link. And I could not be more grateful for you. It's so good to know I'm not on this journey alone, and we are all really improving the world together with the knowledge of this technology. So thanks for being here. Hope you enjoyed it, and I'll see you next week. Bye, everybody. Let me stop sharing.