RISE UP AI Challenge -Day 1 AI Adoption Strategies

Hello, hello, what's up everybody? I'm so excited for this. Welcome in, welcome, welcome. It's gonna take a minute for people to get here. So excited to see you all. Hello, hello, hello from Washington, DC. I'm coming to you live from here. I don't know how many of you are here from DC, but welcome in, welcome. Hello, my people. Afa is in the house. Good to see you, Sandrine. Nice to see you as well. Don't forget, now's your chance, um, go ahead and throw into the chat if you were here, join us live. What's up? Good to see you. Um, what's up, Tara? Oh, Sherry's in the house. Tony's here. What?

Jacob.

Oh my gosh. If you want to, let's just find out where we're all coming from. In addition to saying hello, why don't you say where you're from? We have over 500 people registered for this challenge, so it is amazing. What's up, Joe? Wow, welcome, Robert. Mr. Perkins in the house. Tristam, nice to see you. John. Amazing. Wow, we got Portugal, Greenville. Good, good to see you all. What up? Oh yes, and of course DC's in the house. New York City, of course. We got a couple New Yorkers here as well. Iowa, I know, is in the house as well. I mean, uh, good to see you, my friends. Oh my God, it's so good to see everybody. Amanda, Rochelle, so cool. Thank you all for being here. Obviously there's going to be over 500 people here, so I won't be able to shout you all out, but I'm so glad you're here. Number 1, most importantly, take a look at the people that are here. These are going to be— maybe not today and maybe not even tomorrow, but these are your people. Um, we are all on the same journey together. Many of us are much further along than others. Some of us are just getting started, but we are all on the same path, and that is what this challenge is all about. I'm so excited to have you here. What's up, Kendra? Yes, Midwest Con is coming up. Can't wait. Nebraska's in the house. Awesome. Toronto, Maryland. What's up, Caleb? I'm in Maryland. I'm waving to you. We will maybe not get to meet in person, but we'll definitely be on the phone soon. Caleb is one of our amazing mentor mentees that are in our mentorship program. Awesome. Amazing. Kajal, good to see you. Githika, it's so cool. Yeah, way from far away in South Florida. I am here in DC today and it is my pleasure to bring you this challenge. I always like to give you all a little bit of background on how this all happened. I woke up one morning and I was like, wouldn't it be cool if we all got together? And I basically pieced together all the things that I have been doing and working on and teaching over the last, let's call it, 3 years, um, and condensed it into a single day. So you're going to get a glimpse into 5 of the main things that I work on as a business. Today we're going to focus on AI adoption. It is the number one question people ask. How do I know where I'm at? How do I get where I'm going? Many of you, um, I don't think I brought one of my books with me, but many of you, if you are in this challenge and signed up for VIP, thank you to the 50 of you who said yes to that experience. You are getting a copy of my book, Scaling Responsible AI: Going from Enthusiasm to Execution. And that is the question, right? How do we get started? How do we not only get started, but many people already have all these little you know, baby tigers. You all know I, I carry one around with me everywhere, but everyone's got a little baby tiger, maybe 100 of them running around their organization, all these little AI models, and people just don't know how to rein them in. So that is exactly what we're going to focus on today. For those of you who did not have a chance, please check your inbox. You were here because we're on LinkedIn Live, but please check your inbox. If you're in VIP, you got an email with the Zoom link. That is a chance to do the workshop live. It's hands-on, fingers on keyboard, minds open, and sharing experiences. And I am so excited for what the next hour is going to bring, but it gets even better in the second hour of this challenge. So if you're up for that, we have so many cool things we're going to do this week. Today is about adoption. Tomorrow is my second favorite— well, maybe they're all my favorites. They're like my children. I love them all, these topics. The first Today is adoption. Tomorrow we're going to talk about AI solutions architecture, which is how we all need to really be thinking and building AI solutions. Today on Wednesday, we are going to build our own agent, fingers on keyboard, build an agent so that you understand what it takes to build the groundwork for building responsible agentic solutions. On Thursday, we're going to talk about what it takes to have a thriving startup company. So if you are a consultant, if you're in a company building an entrepreneurship or an entrepreneur, this day is gonna be really exciting for you. And on Friday, probably the most important thing is not only how do you do the things, which we're gonna spend Monday, Tuesday, Wednesday, and Thursday talking about, but on Friday we're gonna talk about how do you share what you're doing with the world? How do you create thought leadership and a brand presence? And I cannot stress to you enough, the people that are in this chat listening to this right now, these are your friends. You may not know it yet. You certainly are my friends. I'm so glad that you are here. It's going to be a jam-packed week, but I do want to make sure that we talk to you. I try to only, you know, I don't want to overstay my presence in your LinkedIn feed, and I want to give you all of the value that we can today on day one. So we're going to jump right into AI adoption. And I've brought two people who have delivered this workshop for the Generate conference that just happened over the last couple of weeks, one in New York City, another one in Chicago. They were extremely successful. And I really wanted to bring that very experience to you all here virtually. So I'm going to let them take it away and they'll have a chance to talk to you all about that experience, but also what it takes to truly adopt responsible AI at scale. So let me bring them in and introduce them one by one. First, I have Yuvika here, Yuvika Sharma. She is a consultant as well. I'm going to let them both introduce themselves to you also. And my good friend and member— you both of these are AI leaders in the AI Leadership Institute. I could not be more proud. Um, Padmini, so thank you for being here. I'm so excited to have you, and I just want to go ahead and let everyone, you know, learn from you, learn from your adoption strategies that you've already been proving all over the United States. Um, so go ahead and take it away.

Hi everybody. First of all, thank you so much, Noelle Russell Russell, for having us and, you know, conduct this workshop. We are super excited. My name is Yuvika Sharma. As Noelle mentioned that we are part of the, you know, AI Leadership Institute, it's been such a fulfilling journey. Besides, you know, being an active member of the institute, I also run a consulting organization where I focus on AI literacy adoption strategy. Most importantly, governance, risk, and compliance, making sure that people are not ending up in the news for the wrong reasons. I'll pass it over to Padmini. She's my partner in crime and a lot of other initiatives that we have going on. With that, over to you, Padmini.

Hey everyone. I'm Padmini Soni and I'm the founder of Resonance AI and also the head of responsible AI growth at the AI Leadership Institute. Thank you all so much for being here today, and thank you, Noelle Russell Russell, for creating the space for this conversation. I have 25 years of experience in tech across Fortune 500s and small and medium-sized businesses. Yuvika and I have started a community called Asian Women Advancing AI, and along with Noelle Russell Russell, we are also co-authors of an upcoming book on OpenAI GPT builders. I work in, of course, AI ethics, AI adoption, and business transformation using AI. So with that said, I'll hand it over to Yuvika to go over the workshop details or the session details.

Awesome, sounds good. Thank you so much. Let us know if you're able to see the screen.

So far I can't see it. Noelle, are you sharing the screen or Yuvika, are you sharing?

I have it, I think. Okay, there it is. Now we see it. All right, so welcome Welcome today. As Noelle Russell Russell said, that today's workshop and lesson is all going to be about AI readiness and AI adoption. How do we actually get on that journey together? So of course you already heard about our wonderful bios, so we are just gonna quickly skip through that. So with respect to how the workshop is gonna work today, it's going to be a two-part series. So the first part is gonna be this first hour where we are gonna be talking to you about you know, when we talk about AI adoption, what does that really mean, right? What are the core components and how could you actually start implementing some of those strategies to ensure that you have a successful adoption? And with that, we actually have a quick poll that we would like you to, you know, just take. So if I can kindly ask Noelle to just pop in the QR code or the link for our folks to just, you know, take on that quick poll that we have just to understand where folks are on the AI adoption journey.

You can see it, right? It's right there on the screen. So get out your phones, take a snapshot. If you're on your phone right now, we'll put the link in the LinkedIn unless someone nice wants to go ahead and do that for me who's in the community. You all can Help a girl out. But yeah, right there on the screen, that's the QR code. It's going to take you to the Slido that will facilitate this. So there you go.

Do you want me to— familiar faces or familiar names?

Yeah, so excited to see so many of our friends from the AI Leadership Institute. Okay, people are popping in. Amazing.

And right now we're just seeing your slides, but if you want to share the Slido results, you're welcome.

Yeah, yes, yes, actually let me do that. If we can actually just kind of see that.

Um, yeah, thank you for sharing the link, Jay.

Okay, so let's—

we've had 50 people scan it, so it's pretty good.

Amazing.

So it's growing, 55, 60, so it's going pretty fast.

Thank you.

Okay, people are still taking amazing.

I'm gonna go ahead and hide it because it's now on this, the screen itself.

Ah, okay. Yeah. Awesome. So I think people are still taking it.

Um, we have a tie.

Uh, yeah, I think so. So far we've had— oh, it's still going on. Okay, all right, so it seems okay. So I think, um, as you all can see, okay, so we have quite a few people. So the good thing is that, um, there are, you know, there are some folks, uh, oh wow, we have some expert users too! That is pretty amazing. I would love to hear from the folks who are expert and, you know, who are using at the organization and really at that mature level. But it seems that there are folks who are, you know, it's a nice mix, but it seems that there's a good amount of people like 30%. Well, okay. All right. I think maybe we need to stop so you guys don't change your mind. No flopping around going on. So, all right guys, look, you know, we are not getting any award or anything, but thank you so much. So just to kind of take a look, this is nice, right? We see that there are folks who are occasional users, right? They are doing, they're experimenting, there's proof of concept, pilots are in progress. That's fabulous, right? And then there are some folks who are regularly using it and they've actually started implementing it in various departments. This is good. I think this is really good. Like almost like good 50-60% folks are experimenting or have started implementing. However, there are a few folks, you know, about 24% who have just tried it and who have not done anything, right? Or there are some folks who have not even tried it, nothing. So I think this gives us a very good lay of the land. There are a lot of folks who need to understand, you know, how the adoption needs to happen, what are some of the things that need to be executed in order to make sure that we have a successful adoption. So with that, we are going to move over to our PowerPoint. So can you all see the— our screen back to our PowerPoint? Yes. Okay, awesome. So let's dive into what today is going to be. So during the course of this presentation, we are going to be walking through some 3 core pillars that as Padmini and I were conducting these workshops and as Noelle said, right, that we identified that we essentially need to have a real conversation about, right, that okay, what are the respective barriers, right? Because when we talk about any kind of change or transformation and, you know, working with a technology like AI, there are certain barriers, right, to adoption. Then we are going to talk about that, okay, what does it mean to be ready at a personal level as well as from an organizational standpoint? Then we are going to talk a little bit about what are some of the risks. There are some apprehensions, there are some risks that people are actually concerned about when we are thinking about jumping on the AI bandwagon or experimenting with AI. We'll dive into what are some personal risks, where could be some risks from an organizational perspective, what are the ethical risks, and then we'll talk a little bit about some of the risk mitigation strategies, right, that will help you become a little bit more comfortable. And then the third pillar we are going to be talking about is the lessons learned, or what are some strategies that we have seen or heard from folks executing. We've seen with organizations and what are some best practices. So hopefully you enjoy the session and you find it engaging and meaningful. And with that, I am going to pass it over to Padmani.

Thanks, Yuvika. So yes, let's get into the first pillar of this session. Based on the responses that we saw, almost 65% either have not used it or use it infrequently or occasionally. So these are just about 35% are regular users or expert users. So we do see that there is some sort of a barrier, whether it is personal or organizational. Let's dive into that. So I'll start with the personal. What I have seen in my experience in just conversations that I've had in my community here or at events and workshops that we've conducted or just the work that I'm doing with you, Vicka and Noelle, that there is some sort of, and I'm going to talk about just 4 of them, there are several, but there are these barriers. We sometimes come in our own way in order to adopt AI. So the very first one is this skill anxiety. Many of us think, oh, it's technical. I'm not a data scientist, I'm not a PhD, where do I even begin? This sense of inadequacy freezes us, you know, it doesn't let us move forward. And nothing can be further from the truth. AI literacy isn't just for the technical folks, just for the PhDs. Today there are tools that offer low-code, no-code solutions, so it can let anyone— is not necessarily only for the technical folks. So this we have to give up up on this skill anxiety and we have to push through this barrier and come out on the other side where we say, okay, whether it is personal, whether myself or I enable someone to leverage AI, adopt AI. That is the first one. The next one, job security fears. Yuvika and I were actually in New York early in June and when we asked the folks there, what is the one word that comes to your mind when you think of AI, the biggest aha moment was when one of the, a few of the people said job loss. That is sort of top of mind for everyone. You know, will AI replace me? You know, we absolutely need to reframe that narrative. We know AI is excellent for automating repetitive tasks. So we should use AI as a strategic partner, something that can augment us, not replace us. Yes, it will replace repetitive tasks. And I think that's what we want. We don't want to be doing the same thing over and over again. In fact, just this morning, when I was unloading the dishwasher, I was like, and the dishes were clean, I was like, thank God, the dishes are clean. I don't want to be doing this over and over again, you know, let the machine do it. So that's exactly what it is, you know, we do not want to be doing, struggling with these. So we should adopt AI in order to help us hand over some of these tasks to that. And when we develop AI responsibly, that is exactly what we do. We ensure that we have the guardrails in place. We ensure that this is going to just be an extension of us and not going to replace us. So if you are concerned, and then of course, We've all heard AI will not replace you, but people who know AI will replace you. So if you are concerned about job security, then definitely adopt AI because again, you know, people who know AI can replace you. So number 2. So, and then 3, privacy concerns as a personal barrier. If you're worrying about data privacy or black box solutions or opaque algorithms, then you're not just being reasonable, you're being responsible. In fact, we, when it comes to models, we need people who are thinking about these. We need people. So if we have the technical folks building this, we need governance and oversight. So this is a great space to be in. We need tools that will sort of help us make more ethical decisions, drive transparency and mitigate risks. And we— and that way we earn our stakeholders' confidence. So rather than not adopting AI because of privacy concerns, because you feel that, oh my God, I don't understand how this— it, you know, it came to this decision, or I don't understand because I'm not sure if my data is secure there, then I think it is incumbent upon us to make those, you know, the due diligence of, you know, go ahead, make sure that, you know, uncheck the box that says we'll you use your data for training, but don't stay away from AI just because, or not adopt AI just because you're concerned about privacy. In fact, if you are concerned about privacy, flip the narrative again and say that that is your being a responsible person. Get into this, understand, uh, you know, how this works, and, um, be a responsible AI user. And the last one, um, sometimes I call it my favorite one, is this illusion of AI adoption. I've seen this over and over and over again where people are like, oh, I, I I know AI, I have adopted AI, just because they are recrafting their emails in ChatGPT. That is not all of AI. That, that, that you are limiting yourself, you know, by saying that you're already using AI. So push past that. And there are so many more, you know, cost concerns, unclear ROIs, but these are the 4 that I wanted to talk about. And in the interest of time, we'll go on to organizational barriers. So in terms of organizational barriers, what we've seen are these legacy systems. We have this monolithic architectures, fragmented data silos, and outdated tech stacks. So instead of looking at this whole thing and saying, oh my God, we are unable to replace this, we can't do it, we should like— we should go away from that rip and replace, and we should incrementally sort of modernize it, you know, take a pilot approach, identify spaces which are like data strangleholds, and, and then we, we look at those and we can deliver maximum value by identifying those, you know, very focused problems and replacing those with AI. So instead of looking at it like, you know, a legacy system being a barrier, look at it as an opportunity to identify, and then to identify that one you know, pilot case, that pilot use case that you can go ahead with. Then the other is a lack of AI strategy, no unified vision for AI decisions. They are either ad hoc, so different departments are doing different pilots and there is no framework in place. That could be a barrier for AI adoption. So if you are struggling with that in, in your organization, then it is time to look into AI governance. In fact, just I think a week or two back at the Chicago AI Week, that was a roundtable that I was hosting, was exactly that: is AI governance the missing link between AI innovation and trust? And in fact, it is. We need these governance, AI governance in place so that we avoid these missteps and are able to successfully deliver an AI project. Then of course, again, the skill gap. We need to upskill, upskill in the organization. We need to have, um, a, a, like an AI task force or something that, that, that is in charge of this. We need to have some sort of, you know, um, like a T-shaped talent ecosystem where we enable— there's a breadth of AI fluency across the enterprise. Everyone understands what models can do, what models cannot do. But then we paired that with like deep specialists who can architect or maintain these complex systems, who understand the ethics of AI. So it is very important to have this, this T-shaped talent ecosystem. And then, of course, budget constraints and the unclear ROI. So yes, I do see a lot of teams, organizations, you know, not adopt AI because they are concerned about how much money is this going to cost us. We don't even know whether, you know, after spending all that money, whether we'll get a return on our investment. So we have to frame every initiative with a very tight business case, and we have to move AI from being a cost center to a value accelerator and do these time-boxed pilots and report the dollars that we've saved on this. So sort of go from a pilot, start with a pilot, and then finally go on to a portfolio. And with that, I'll hand it over to you, Vika.

Thanks, Bhadmani. All right, yeah, so we already heard about, uh, you know, the personal and the organizational barriers, but, uh, one of the problems, uh, which is also quite key, um, you know, it's almost like, you know, think about the risk, right? People think about innovation, they think about all the benefits, but there are risks associated with it as well, right? And where there is risk there is apprehension, right? Because people think about, oh, because so-and-so technology does so-and-so things, this is going to, you know, it's gonna bring a bad name to me. So we all know that the AI technology is not perfect. Of course, it's quite powerful. We all know that the AI hallucinates no matter what, right? Like, I've had countless times where I'll, generate something with AI, and if I'm not reading it or vetting the content that it's generating or doing some cross-checking, I know, like, you know, I could be setting myself up for a failure, right? And it has a habit, um, you know, to generate inaccuracies because in reality it's not a magic wand, right? It hallucinates, yes, every now and then, and it can produce inaccurate details. So that is looked upon as a risk. But that does not mean, right, that because it's a risk that it should be a deterrent, right? There are ways to better manage that risk, right? Then on top of that, there are data quality issues. As Bhadwani said, right, that there are data siloed across different departments, but there are also quality issues with the data. I cannot even count or point out one organization. I've been at some of the best companies where I have not heard anybody say that, hey, my data is perfect, right? So the data quality is an issue. And we always say, right, that if you're gonna put a bad quality data in the model, the output is gonna be bad, right? Like, what do you say, like, it's garbage in versus garbage out, right? So that's another problem. Then there are like these system integration issues because, as Padmani had said, that a lot of companies have like these legacy systems which are not going to be talking to each other. So as it is, that's a problem. And when you start talking about implementing AI and when you're thinking about scaling responsibly, right, and that's something that Noelle Russell talks about in her book, I think that tight integration and that system integration is extremely important. So that is another technical challenge and a risk that needs to be managed. So that's one side, right? Then you also have the ethical ones. We all know that AI outputs are biased. So I'll tell you an example. Just very recently, I was trying to generate an image, and without fail, anytime I say that, hey, generate me an image of a professional, by default it's a man. And no offense to my, you know, man friends out there, you all are awesome, but why can't a professional be a female, right? And just very recently I was having a conversation with my son, right? And we had another one of our friend's daughter and we were like, yeah, you know, what do you want to be when you grow up? And we are taking some images and I think we talked about, okay, show us an image of a doctor. And the first image that comes is like, okay, there is a man, right? Because, but it's not the AI's fault, it's basically the data it was trained on, right? So these outputs can be biased because humans are biased. Right? Our data is biased and those biases are actually baked in, but they say that yes, it's bad, you know, so there are the ethical concerns and the risk that who's going to use the technology and how do we better manage that. Then we have privacy issues as well, right? That's a risk, as we all know, that if we input private data, anything that's, you know, private identifier information, and if it happens to be used by the AI models to train, that's not a good thing, right? So those are, you know, some other risks that need to be managed. And the most important thing is that some of these models, especially like the deep learning models, these are black box models, right? So if you were to talk about machine learning, if you have like regression models, those you have an idea, right, how they actually came to it. But when you're talking about generative AI or deep learning models, where the calculations are so deep embedded, right? You don't even know what's happening under the hood. So especially when you have these black box models being applied to folks, right? And it could have a bad impact, right? So there are these risks, right, that better need to be better managed, right? So yes, there are risks, but we feel that there are ways to manage it, right? But the main thing is being able to identify that there are risks associated with it But if we manage the risk well, we can actually have successful adoption. Okay, so with that, I am going to talk about— yes, so now how are we going to be able to manage these risks, right? So the first thing is being able to identify the key risk, right? So as Padmini had mentioned, that the ROI happens to be, you know, a big one, especially from a business standpoint. Because businesses are thinking that, oh, what if this project is not successful, right? And what if I implement X million dollars and if I don't get the benefit? That happens to be a risk that businesses need to think about. And just very recently, Gartner made a claim that by 2027, 40% of the GenTech AI projects are going to be scrapped. And one of the reasons is because people are jumping on the AI bandwagon without knowing, right, like what they're signing up for. So we need to be educated. The technology is powerful, it can reap great benefits, but it needs to be managed well, right? So there is this ROI risk and uncertainty, but the one thing to keep in mind is that when people are planning these AI projects, we tend to think that, oh, I need to have immediate return on investment. That's the thing with AI projects. You may not see an immediate return on investment, but you also have to think about what qualitative benefits it brings. And we can have a separate conversation about it. But the other risk that happens to be there in businesses, they take the spender lock-in. So when companies are deciding to implement AI, they're either thinking about buy or rent. And Noelle Russell Russell does this in one of her workshops that you actually have to make a choice, right? Like, either you're going to buy the, you know, the solution, or you're going to rent the solution, or you're going to build it. And there are pros and cons to all of those, right? But if you make strategic decisions, there are ways to manage that. And then, of course, if you are not implementing AI, there is this thing that people think that, oh my God, I'm actually going to be at a competitive disadvantage. And because my so-and-so competitor is doing this, I'm not going to be able to compete with them and I'm going to become obsolete. And that's a very valid, you know, risk and concern that companies are facing right now. We all know what happened with Blockbuster, right? When Blockbuster did not didn't pivot quickly, sadly, you know, they went out of business. I used to love going to that store and renting the DVDs. So, you know, we have to think about those things. And then, of course, with implementing any kind of new solution and change, one thing that we highly recommend is that you actually need to start small, right? You don't try to do everything all at once. So some of the common mistakes that companies try to make is that they'll want to do everything all at once. The best thing is that you start small, you see if it works, right? And of course, you need to make sure that you are starting small and you're designing properly. And when you're designing and implementing, you have this vision that how is the solution going to scale, right? You have to keep that in mind. And then most importantly, you also need to make sure that you have over human oversight. Just this morning I was talking to an executive and I was explaining that when you're implementing AI, it's not a macro, right? That, okay, it's an automation and this is just one and done. No, you have to make sure that how do I make sure I check the work? How do I make sure that I have human oversight built in? How do I make sure that my AI does not go rogue? Right. So those are some things that we need to make sure that we can monitor what's happening with the AI in the course correct, as we mentioned, that these technologies are not perfect, right? Then we also need to have a clear governance structure in place, as Padmini mentioned, right, that AI governance is becoming more and more central to it. When we work with clients, my recommendation is always when you're thinking about strategy, you need to have the AI governance like right there. You need to think from the get-go that, okay, how am I going to, you know, govern this technology? Do I have the right processes in place? Do I have the right frameworks in place? How am I going to monitor? How am I going to do my risk assessment and all that good stuff? And also, while you're establishing the governance structure, that's where your monitoring and adjustment is going to come in place, going to be the golden thing that's going to make sure that things are functioning as a well-oiled machine. And if things break, you are actually able to course correct and get back up quickly. So regular audits are going to be extremely important. Having right feedback loops and continuous improvement. As we said, that this technology is not something that is just one and done. Things are going to evolve. Your situations are going to change. The data is going to change, right? You're going to have new scenarios. So continuous monitoring and being like a watchdog. And you need to make sure that you can actually automate some of these monitoring things in place. Microsoft, which we all are huge fans of, has great frameworks in place which can actually help monitor and automate some of these things, which can make sure that your systems are secure and things actually don't break. With that, let me hand it over to Padmini.

Thanks, Yuvika. I'm seeing some questions in the chat, and one of that was the T-shaped reference that I made. So by that I meant that having a T-shaped talent kind of ecosystem where you have a breadth of AI fluency across the enterprise. So everyone knows, you know, what models are available, what a model can or cannot do. So that is the kind of fluency we need across all teams. Everybody needs to know that. But then we need the vertical line of the T, like in-depth knowledge for certain specialists, like who can architect it, who can maintain complex solutions. So that was the T-shaped analogy that I referred to. Then there was one more about governance. The roundtable that I had was, is governance the missing link between innovation and trust? That is what I had referred to. I do agree, Joe, we need fewer leaders who think AI is a silver bullet because in reality it is not a silver bullet. It has to be adopted responsibly and on a case-by-case basis. Like what Tristram said, we should not use AI where AI is not a solution. So yeah, thank you for these conversations in the chat. With that, so some of the success patterns that I have seen and all of us working in this field have seen for individuals to adopt AI. So the group that is in the, you know, I don't want to use AI, I am barely using it, or oh, ChatGPT is all of AI, So for that, you know, you need to get past that by starting certain specific tasks. So number one step for any AI journey is AI literacy. So I think as a first step or a next step is to educate oneself. So join all these, you know, communities like the I Heart AI or the AI leadership communities, great communities. To start off on this journey of AI literacy. And then focus on augmenting. So, you know, these are the tasks that I keep doing all the time, and I really don't like to do it. So those are the ones that you want to hand over to AI. And then practice prompt engineering, definitely. You know, what you get out of AI is as good as what you feed it in. But these models have been trained on a across the internet. So you have to have good prompting skills. So, uh, practice that, understand that, use frameworks that are available. But what feedback you get, please, like Yuvika said, it hallucinates. So use a fine-tooth comb and understand what the feedback is. And embed ethical practices from day one, in even in your personal— AI governance is not just for the big organizations. AI governance also at an individual level. All the pillars of, you know, sort of, you know, fairness, transparency, it applies to us at an individual level too. So please, please embed those practices from day one. It is not a window dressing, it is not an add-on for later, it has to be foundational. Yuvika, if you can go to the next slide. Thank you. And organizationally, we have to start small. We have to identify those pilot projects where we can, like I said in earlier on, these data strangleholds, maybe that's a good place to start. You have to identify what the pain points are, which are good use cases for, for your organization. So, so start with those pilot projects. Then definitely upskill employees, upskill your teams, because they need to understand what they are dealing with, what technology they are using. And this is— when I say technology, it's just not for the technical folks, but it is for everybody. So especially even execs and people, there has to be adoption at the top. So definitely start with employee training. Ethical guidelines. I'll never get tired of saying that. I'll never get— I'll never stop saying that. So I may sound like a broken record, but you absolutely have to have that all the time. And then, like Yuvika said, feedback. It's not a one-and-done. It's, it's all the time. AI itself is adapting and improving, so we have to enable that, help that, and, you know, be part of just building this whole hold the AI workspace together. Thanks, Yuvika, and I'll hand it over to you.

All right, thank you. So, um, what are some common pitfalls, um, that people can actually avoid, right? Uh, it may seem a little simple, and as I had mentioned, right, that, um, people try to go at, you know, at 180 miles per hour. You don't have to do that, right? Um, you have to think about What's the smallest thing that you can actually see where AI would be able to add value? Of course, you don't want to implement everything with AI. That's what happens when you have a technology like this. Every nail looks like something that you can actually hammer. What is that saying? That when you have a hammer, everything looks like a nail. You want to be very, very mindful about that as well. But with that, you want to make sure that you start small, right? That's the best way— that what is the problem that is repetitive, and that's a problem that I would like to automate because I work on it so many times, but it's small enough that I can actually, uh, get meaningful results from it, right? So, um, you want to make sure that you're avoiding that too much all at once. So no. The second biggest thing that I have seen so many times, and this is not just even related to AI, change management, right? So when we talk about change management, um, change management can be very, very broad, right? So some of the core things that I usually like to emphasize with clients is First, identifying that, you know, the change that you're trying to bring in, right, with AI implementing, it actually aligns with your business values. But understanding, right, that what is going to be changing, what process, what systems, and in order to go from A to Z, right, like doing a proper mapping and figuring out what is the change going to look like, Who's gonna be impacted by the change, right? And then the stakeholder alignment. I cannot emphasize enough how many times people would be in the rooms and they would be making decisions and the actual people who are, you know, in the weeds and who are gonna be impacted, they're not there in the rooms, right? So that stakeholder management and involving the right people is extremely important. You want to make sure the folks who are going to be impacted, whether it is internally, you need to get their voice heard and you need to make sure that they have a seat at the table and you're actually understanding how the change is going to impact them. They say people, process, systems. These are quite important that people essentially tend to miss. Who's the people? Who's going to be impacted? Who needs to be at the table? What process are going to be impacted and what systems, right? Do I even have the infrastructure in place? What systems are going to be changing? So basic things that people tend to miss. And then of course, training. Oh my God, training is something that people do not invest enough, right? So when we talk about AI literacy, I think people really need to understand that, okay, how the technology is going to impact them, but but also like how is it going to enable them, right? So it's very important to invest into training. So even when you are thinking about enabling people or helping them understand like what the technology is, but also when the change happens, we need to ensure that the folks know how to work with AI technology or AI-infused systems, right? So there is not enough time spent on training, but I think it's so important, right? Like, you need to make sure that people are comfortable on how the technology is going to impact them, how they're able to use it, and make the best use of it. So, um, I'm assuming that based on the survey that we had, a lot of people know how to do prompt. But if you've actually worked enough with prompt and you've looked at some of the frameworks and the techniques, I'm hoping that people had better luck with the output that they're getting, uh, with, uh, the training, right? So, uh, it's very important to invest into training, right? Even like pre-implementation, but also like post-implementation. And yes, we all know that, um, if the data is not good, uh, your output is not going to be good. So, uh, what is that saying? That, uh, garbage in, garbage out. But even when you are designing the systems, you need to make sure that the data is good, right? It's good quality data. But when we talk about poor data preparation, it's not only about the quality, right? It's also about the representation. So we need to think about that, okay, who is my system going to be impacted, right? And then do I have the right data set in the mix, right? Do I have everything represented in the data, right? Like whether it is the gender, whether it is, you know, a specific race, and what the impact is going to be, right? You also want to be able to identify any kind of biases or, you know, proxy variables, right? So when we talk about poor data preparation, it encompasses a lot of things, right? Like, how are you being mindful about all those things? So, uh, these are some of the common pitfalls, you know, pitfalls to avoid. Of course, there are many more, and, you know, we can have like a more targeted discussion going into each one of these. And with that, uh, okay, so let's talk about like some of the key success metrics, right? Um, I always like, uh, this thing, and I think I've heard Noelle Russell Russell say it as well, right? That, uh, if you are trying to implement something, you also want to make sure that how are you going to measure the impact of these things, right? So, um, when you are trying to implement things at a personal level, uh, you want to be able to see that, hey, How much time is it helping me save on routine tasks? So I'll give you an example. As a consultant, I have to prepare a lot of PowerPoint presentations, and of course I am a PowerPoint nerd and I really, really love it. But oh my God, the time you have to put in. And how many of you have been in situations when you're trying to work with the SmartArt and you're like, oh my God, this graphic is not exactly what I want. To look like, and this is not the best one. So, um, I can definitely raise my hand because there have been countless times where I used to spend hours and hours, you know, trying to perfect that SmartArt, um, and, and still I used to not be happy, right? So, but with the advent of these new technologies, um, I've been blown away. Like, you know, I use tools like Canva, and there's like Napkin AI, and so many which allow me to go from hours to like minutes. So I know like, okay, my, you know, return on investment or, you know, my time on my time saving, it's like phenomenal, right? And then also, how do you also measure the quality, right? Like the output quality improvement. So if you're preparing any kind of content, of course you do it well, but how can AI actually help improve it?, right? And then you also want to measure how confident, you know, has it made you? Like, has that increased your confidence in preparing or executing certain tasks? So these would be things at an individual level. But when we are talking about implementing AI and figuring out what is the success metrics on an organizational level, you would actually think of looking at, okay, what are the process efficiency gains, right? How much time was it taking before? How much time saving did we do? Then people also take a look at that, okay, how much money did it save? How much dollars did go up? What was the revenue increase? Then the other things that we also look at that, okay, what was the employee adoption rate? You could build the most amazing dashboard or something that is so amazing, so cool, but what if people are not using it? These are just some high-level things to measure as you're thinking about implementing AI in real really help, you know, drive your personal and professional growth and also take your business to a new level. And with that, okay, Padmani.

All right, so, um, I think we have about 10 minutes left, and, um, we— so you're here today, you know, which itself means that, you know, you're headed in the right direction with AI. Definitely start addressing the barriers, whatever you want to push past, reach out to us, connect with us on LinkedIn. We're here to help with conversations. Manage your risks, and I'm going to run through this slide, apply best practices and get started on your AI journey. There are some questions in the chat and we can get to it., you know, later, maybe, Noelle. So I'll hand it over to you because I know—

Yeah, absolutely.

Amazing.

Thank you so much. Um, there were so many nuggets people were commenting on. Um, we are, you know, live on 6 different channels, so me trying to manage all the chat happening in all of the rooms was a little challenging, but I tried to get to all of you. But I did want to highlight a couple things that came up that I thought really interesting. The first, and this is kind of a question to everyone, this one came from Maria, who is interested to understand, and I think this is huge. Every executive organization I talk to often struggles, and anyone who started really building production AI at scale often will talk about change management and how difficult, like The technology is, yes, its own kind of learning curve, but really change management is this long tail that just gets more and more complicated over time. And so I definitely would love— there's a few, I mean, we have about less than 10%, but about 10% of the audience said that they're pretty advanced. I know that there's Tony and Joe and Nitin, like there are some pretty good examples of people that are using this in production. Would love your perspective. Please throw it in the chat. I know this question is a more deep question, but maybe you could be like Joe and be like, hey, I have an article on this or a tool on this. One of the most valuable things that we have uncovered is the value of our community and of knowing each other. Like, we all can't be experts on all the things. So I always really appreciate these kinds of comments. But I definitely know that— and you all can look up this term, ADKAR. There are many different frameworks for managing, uh, the adoption of AI into an organization from a change management perspective. I have found my own personal, uh, perspective is what we're going to talk about tomorrow in AI Solutions Architecture. I do believe that AI Solutions Architects are the bridge for that change management and adoption, and I'll share with you my thoughts on that tomorrow. Um, but really appreciated this. So if you have any ideas on this, please throw it in the chat. So that we can all kind of learn from each other. Another thing Tony mentioned, you had so many good comments in here, Tony. I just picked a couple that you had mentioned, but you had talked about, you know, I believe AI adoption starts with solving the right business problems. And I think this sums up what Yuvika and Padmini talked about in a couple different sections, but it is so right. And you actually also mentioned that one of the most common problems organizations have is around bad data. But I would go back to what you originally said, it's kind of a combination of not only not really knowing the right problem to solve, but also not really understanding the data they have to solve that problem. And that combination of things is really, I think, pretty interesting. So I just really appreciate you, Tony, always. Tony is, um, as you saw probably throughout the chat, I don't know which chat y'all were in, um, but he is another practitioner, owns a business on this. He actually Many of us in AI started with data, right? Data was the emphasis pre-generative AI taking the world by storm, and rightfully so. I think a lot of us get excited about the advancements that are coming with these models and the ability for us to build really interesting experiences with large language models. But for those of us that are a little closer to the production side of things, I think we're starting to notice this shift away from large language models that are uber knowledgeable about all the things and this refocus on retrieval augmented generation and the specialization of models. These smaller language models that are fine-tuned and focused on specific problems that they solve. We're going to talk about that throughout the week, but I know we talked about a lot, so I just wanted to reiterate that from both of what you all were sharing as well. And then thank you to Joe for also throwing in this link. So he had mentioned, we mentioned at the beginning this concept of kind of like investigating the barriers that we ourselves have to adopting new technology. And you would all, you both had mentioned kind of Blockbuster and kind of companies that really struggled. And that was a change management problem, right? It wasn't because they didn't know. I've actually talked to executives that were in the room with Blockbuster that were like, we need to change. It's not because— there were people in the company being like, we need to make a diff— we need to choose a different path. And just because that organizational shift was so difficult to manage, they just couldn't do it. Um, so I, I really appreciate this article, so I wanted to throw it up here for those of you who wanted to see it. You could also go up and see the hyperlink in the chat. Um, and, uh, and I wanted to thank you, Joe. And any of these resources, if you have a link that you like, a book you like— oh my gosh, I'm gonna have to share with you. Um, today I'm over in cybersecurity, I'm talking to the Cybersecurity Initiative in Arlington, Virginia, and, uh, one of the speakers brought up this really cool book called Threats. Um, and you can look it up. I think you'll have to look up like Threats and Star Wars. I don't know how many of you know this, but it's basically um, talking about threat detection and threat mitigation through the lens of Star Wars analogies. And it was so cool. So I, I bought the book, like, when he said it, I went and looked it up. It was so cool. He seems like Scott— either it's the first name or last name— but just look up threats and Star Wars. And if that speaks to y'all, um, you all can go and take a look at it. Um, yes, and don't forget, um, to connect connect not only with the speakers, Yuvika and Padmni. Please don't forget to do that. I'm also going to share my screen very quickly to give you all— oh dear, just kidding, I won't be sharing my screen— to give you all— because it's saying I have to like open system settings or something. I guess you all did this before.

Escape.

But escape.

Okay, let's see.

Oh, allow. Let's see, let's try that. But really cool opportunity to get to know each other. Everyone in the chat if you've seen people say things that resonate with you, become— the first step is just being connected. That's it. Just connect, because then you'll start to see content and articles. And all of us are pretty good at posting, but if you're not good at sharing what you're learning, that's what we're going to do on Friday. The last one, Nitin, I wanted to give you a shout out, um, as well, because you were talking about startups, and we do have a day dedicated to startup initiatives and launching a startup. But I wanted to make sure in case for whatever reason you come a little later, you miss it. But how can we pilot implementations where data is new and clean? Actually think you have the best scenario, right? In a startup, you have an opportunity.

Exactly.

Yeah, right? Like to not only grab, use new data, but you can also manufacture data. Like you have the ability to synthesize data. So you have a much better opportunity. Do not be bound by the data you get. And as a startup, you have the opportunity to create, reinvent an entirely new dataset to solve your problems. But we're going to talk a lot more about that on Friday as well. Let me go back to the comments because those are just the ones I starred as we went through. I just want you all to give Padmini and Yuvika some shoutouts, some love in the chat. Let them know how much you appreciated their thoughtfulness in going through this. Here's what we do next. For those of you who are in our VIP, and it's never too late to join VIP. Um, but for those of you in VIP, if you do not have the link to Zoom, meaning it was stuck in your spam, or you— we sent out all the emails twice, once yesterday, once today. But if it's in spam and you haven't found it, just send me a message on LinkedIn, like just DM me and be like, hey, I'm supposed to be there, and I will check. And Sherilyn will also be monitoring my inbox, and we'll send you the link so you can join. We'll wait a few minutes before we get started, but I want to let you know why these after sessions are so critical and so important. Number one, um, this work that Padmini and Yuvika just showed you, it was kind of like the TED Talk version of the most exciting part of this workshop, which is we're gonna actually break out— yeah, into small groups, and we're going to work together. We're actually going to answer the questions that are in— if you're VIP, you've got access to a very specific Roadmap. If you're in our I Love AI community, you're going to get an assessment that you can take on your own if you're not VIP, but— and it's in the email if y'all go back and look at it. There's a little assessment you can go take a look at. We're going to actually take a look at some of the answers that came in from this afternoon's work and kick off tomorrow with it. But for those of you coming to VIP, it's a hands-on workshop. We're gonna work together to talk through how you adopt successfully in your organization. What are the common challenges? What are the common successful solutions? And most importantly, what are the questions you can ask yourself to encourage a smooth and successful adoption of AI in your organization? So thank you. Just want to share some of the love. Great info. Good job, you all. So cool. Thanks, Brian, for being here. Thank you, Addison. Super awesome.

There's— Noelle, are you able to throw my LinkedIn profile into the chat? For some reason I am not able to do that.

Um, that's a great question.

Thank you.

And would love to connect with all of you. You've been such a great audience. Thank you so much.

Um, so I'll stay here for a second. Oh, there, there it is.

Yay!

All right, so I just tagged Admini in the chat, and I'm gonna do Yuvika now as well. Just click and connect Connect with all of us. I'll tell you, you never know. Today, where we all are will be different than where we are a year from now. And we are the doers, right? We're the people who said yes to a challenge. We're learning new technology. By the end of this week, you're going to know more than you ever thought you wanted to know about these technologies. But more importantly, you're going to be enthusiastic and excited. And that is going to be an anomaly of people you're going to want to be around. So, oh dear. Trying it again.

There we go.

Uh, Yuvika, uh, let's see. Oh, I wonder if it ended. No, it's not, it's still there. Yuvika. Oh yeah, I forgot to put an symbol, so I'm tagging Yuvika in here as well. Somebody else might have beat me to it. Um, there we go. There. So be friends, hang out. This is like the coolest people I know are in the AI Leadership Institute. We are all together, all learning together. As you know, there's so much to learn, new stuff every day. The best way to learn is to learn together. So thank you all for being here on day 1. We are about to head over to the Zoom for live in-person. We get to open up our mics, ask each other questions, and then get a facilitated adoption workshop from these amazing leaders. So thank you so much for your time today. Thank you, Yuvika. Thank you, Padmini. Amazing. And yes, please connect with each other.

Thank you.

We'll see you all Tomorrow, same time, same place. Check the LinkedIn, you'll see it. Um, and also if you're in the I Love AI community, which all of you should be, that is where the links to all of the calls are. So you have to go there to get your calendar. If you're not there yet, just go to noelle.russell.ai. That's the website. I love AI. It's free. You just— that's where all the links are. So go there. It's also where all the assets will be, all of Downloads, all the PDFs, the checklist, all of it for the rest of the week are going to be in the I Love AI community. So join us there. All right, thanks everybody. Y'all are amazing. Thank you.

Bye. It's great seeing you all.

Thanks. Bye. See you later.

There's someone asking for the Zoom link. The Zoom link is only if you've signed up for VIP.

Yes, if you're in VIP, send me a message, we'll add you to the Zoom. And anyone who's in AI Leadership Institute member, you automatically get access. So just go into the AI Leadership Institute and you will see a pinned post with the Zoom link in that post. It's already there. So I'll see you all there. Super excited for this. All right, Yuvika, Padmini, are you ready? Are you ready to do this?

I'm ready.

Yes, let's go.

We're heading over. Okay, bye everybody.

Bye.