Thank you for your time today. I'd like to start with a story. I wanna tell you about 11 year old Monique.She's the youngest of six kids, and she attended a small school in New Orleans, Louisiana. And one of her schoolmates came to her one day and offered to, teach her how to crochet because this schoolmate is a guru with crocheting. And it sparked, a lot of attention at Monique School because the school is small enough that even small things become really, really big. So the principal even gets involved with this crocheting and, and actually launches that school wide competition. So all the kids got involved and they're gonna actually get involved with, with making a granny square. So Monique runs home.

She tells her mom that she's gonna participate by, making this granny square. And she asked her mom to go to the store and they buy a blue yarn and a gold yarn. And somehow Monique creates this awkward looking, granny square, which she submits to the competition and, and, you know, anticipates what's going to happen. Well, she didn't win the or even the prize. But in fact, though, Monique, she won the grand prize. She she she actually won this huge gold trophy. The principal presented to Monique a huge gold trophy. And I'd like to ask you ladies today, in the chat, if you'll let me know in one word, if you can describe a time when your intuition or your data led you to outperform even your expectations.

Let me know in the chat. My name is Desiree Lemons, and I've spent, my entire career at the intersection of data communication infrastructures and innovation. And most recently, I've been a global product leader for an industry leading data center that is shaping requirements for AI ready, solutions and platforms across the global industry. Now I'd like to talk to you about this agenda because AI today is only as powerful as the data that's behind it, and it is good data. So to lead in this era of AI, we have to confront as as women leaders, we have to confront three truths about the surging AI demand. And that is if you're in data center like I am, Data center and infrastructure growth forecasts are very challenging. If you're in an enterprise, if your company should or should not get ready or if it is ready for AI readiness, that's even unclear.

And the area, which is gonna be fun, as consumers, we've seen so much AI magic happening around us. How do we trust the data that's empowering this AI? What can we do to trust it? And then I'm gonna conclude with the blue and gold. Now for housekeeping, let's talk about what do I mean when I say, trusted data. Well, this data is accurate. It's reliable. It's verifiable, and it actually is essential for helping us to make key decisions. It helps us to make informed decisions and even to reduce the risks around the, products and services that we deliver. And as far as the successful outcomes, it'll be across our industries, and it'll help us to be strategic.

Let's go back to that point I mentioned to you about, the challenge with, with AI and its demand. It makes it very difficult for, those of us who are in a data center space who are building out infrastructure to properly forecast how to build out that infrastructure. I mean, from space, power, and even cooling capacity. Let me draw out some numbers to you. Right now, the reports are showing that in this year alone, that we're gonna be spending $300,000,000,000 on infrastructure all the way up to over a trillion dollars in infrastructure by 2029. This is by network world. Now how do we relate to that, women? Well, $300,000,000,000 would be comparable to every woman in The US buying a designer handbag every week for an entire year, and that's a huge number.

And $1,000,000,000,000 is comparable to, the industry's beauty, fashion, and personal care spending all combined, all that revenue for the next ten years. That's the trillion dollar number. Now that's big. Now when we look at some of the the noise that's in this space, we have some of the players like golden Goldman Sachs. They say that, power for AI, it will increase by 165% by 2030. And McKinsey weighs in by saying something that's slightly different, that 33% of CAGR is going to is going to increase by 33% for AI ready capacity. When we look at our utility stakeholders, they are speculating that a lot of this demand is five to 10 times higher than what actually the growth will be. I spoke with an executive just this past week around and you utilities, and she said something similar.

She said that most of what we're seeing for this expectation is redundancy. So what really is going on, and how can we know what we should do as women leaders and AI technology? This is where I step in to talk about one of the two models I'm gonna mention to you. One is my three c model. And this three c model might be something similar to something that you have, done yourself, where you wanna reach some type of clarity around what's going on. So we're looking at the clarity from that perspective. We're looking at the context, and we're also going on to look at the credibility. Let's step back and look at, the clarity. What's really being said here, and what units are is the speaker speaking on? Is it the megawatts?

Is it AI and workloads, and is it being expressed in FLOPS? Are we talking about training compute, or are we talking about real time inference? And for context, who's actually the player here? Who's the stakeholder doing the forecast? So when I spoke with the exec and utility space, she said oftentimes, you have one hyperscaler, and you have multiple data centers that's actually trying to service this hyperscaler. So you can see how there can be a redundancy in forecasting and an expectation that's far higher than it's projected. And lastly, the credibility. Is it possible for that forecast to be triangulated across other areas like rack pipelines, edge deployment realities, or GPU availability? The takeaway is this. AI infrastructure demand is in fact very real, but there is also noise.

When we use, these the simple model here, like the three c model, to help us to see what's going on and the environment around forecasting. It helps to remove the hype, hope, or the hard data. Separate that out. And when Goldman Sachs says things like the demand is gonna surge or McKinsey says something about gaps, or utilities say, oh, this is overestimation. We can actually not see that as contradictions. We can see this as insight. And it's our job now, yours and mine, to look at this data and not just consume it, but help to interpret it so that we can build trust through its insights. At this time, I'd like for you ladies to drop in the chat if you also have seen conflicting AI infrastructure predictions or anything around AI in terms of expectations.

Did you believe it and why? Now the next area I would like to talk about is enterprise AI readiness in your company. Or are you ready for AI? Have you seen AI being implemented inside of your company? Or you've seen maybe a straying away from getting involved with AI. I'm sharing with you here, how is it possible for us to know, or how are we as leaders to know if our enterprise is ready for AI integration? I'm sharing with you here on the right of this slide, a model that I learned from Northwestern Kellogg School of Management. This is their, AI, radar. This is a model that they use to help, businesses to know where it is they should start with AI integration. And I found this model to be completely fascinating. The AI radar is a 360 degree view of the enterprise functions and helps to prioritize the targets for opportunity.

Now there are two dimensions that are are shown here, and those two dimensions help us to see where we should start. Data intensity is one dimension and process complexity. Because when it comes to if our companies are ready for, AI, if there is not a tremendous amount of data density around a specific area where we want to integrate AI. It would make the prediction the predictions, the trust according to those predictions. It will make it very difficult and uncertain. So the data density has to exist there. And that means that there's a tremendous amount of data, not just, any type of data, but trusted quality data. And when it comes to the processes, the process itself will have to be a a process that, is seen as being, sluggish. It's a necessary part of the business, but it's really manual.

And you see this as an area where if you could, automate it with AI, it would be perfect for your business. Now how do you make this work? You make it work by saying, those 12 dimensions that are shown on the AI radar, each of those dimensions can be scored between one and ten. And that score, again, is based on the data density and the process complexity. So score it between one and ten. And if you'll notice, there are three dimensions per quadrant. So each quadrant can actually have a score between three and thirty. So this whole 360 degree can have a score between twelve and one twenty. Now that's how that model works. I actually use it against my, my company, and I was able to, actually offer suggestions on where we can start.

But, again, we can see that, with the lack of data or lack of trusted data, even this model would not work. So inside of our companies, it's very important that we produce that data. So please put in the chat, whether or not you believe your company is ready, AI ready. Put one if it's yes and put two if it's no. The next area I wanna talk about is AI magic. Is there a lot of AI magic around you? Are you seeing lots of products and services where we're seeing AI being, pushed at us through these products and services? I'm gonna share with you just two products. And one is a product who's the founder of this product. It's called ReSpin. It's by Halle Berry. Now this product, ReSpin, actually is a personalized menopause care, and it what it does is it analyzes 150 data points per user.

And it looks at looks at things like our sleep, our mood, our hormone patterns, and our lifestyle. And it builds a tailored wellness plan using AI. Now that kind of personalization, it requires a tremendous amount of AI infrastructure, and it looks at a lot of, sensitive health data. Now do we trust that? Would we do that? What model can we use here? I will still use the three c, model here to scrutinize this data. I wanna know whether or not Hallie has done all the work that she could do to make these predictions accurate. Is this something that's just done in a lab or done with people for clarity? For context, is this a diverse set of people? Is Halley also looking at, a diverse range for menopause? Is it beginning? Is it later stages of menopause? For credibility, are there any real experts?

In this case, Halle, has taken on, doctor Maki as a leader in women's cognitive health. So we know we that she's done her work. The bottom line here for this product is menopause, in fact, does deserve innovation and AI can help. But we know our health is very important and is very sensitive to us as women. So we would want to know if we can trust this, and Hallie has done all she can do. So in the chat, would you trust an AI platform to guide your wellness journey for yourself, for your mom, or your sisters? Why and why not? The one I wanna share with you is another interesting one. And this is, Alice Chang, and she's doing something really innovative, and it's around both AI and AR.

And that means and what she's doing here with, Perfect Core is it's a mix of proprietary datasets, confusion computer vision, and deep learning models to deliver what is, again, similar to Halle's personalized skin care and beauty insights all done across the screen. It analyzes from pictures your pore size, your hydration levels, wrinkles, and even your hair texture and curl pattern. And then it makes product recommendations. She has partnered with dermatologists and other credible folks. It seems like to me that Alice has really put in the work, but I would still put my three c model across it to determine whether or not, this product is for me? Are the features and outcomes explained in a way so that, as a user, I truly understand? Can I tell what the specific data points are around the wrinkle depth or the moisture level that's being mentioned here?

And for context, did Alice look at all women, all skin tones, and lighting conditions? Are there cultural differences? And for credibility, are the AI outputs, verified by dermatologists? Did they train the models accurately across a diverse set dataset? The bottom line is Perfect Core is a leading example of AI and consumer beauty. But if we're gonna trust these tools with our faces and even our feelings, we do want transparency. We wanna believe that this data is in fact accurate. So in the chat, would you trust this AI to recommend skin care based on a selfie? Why or why not? And I wanna conclude here. I wanna return to that 11 year old girl, Monique. She actually is me. I'm Desiree Monique. And I'll tell you I was completely stunned when I won that huge trophy. I did expect, Stacy to win because her granny square, it was perfect.

Mine was very awkward, but I actually did not win because I had the bet I was the best crocheter. I won because I picked the right data, trusted, relevant, and emotionally resonant data. The blue and gold weren't just colors. They were actually context. They were customer empathy. They represented brand alignment. That was actually the data that sealed the deal. They were the school colors, and the principal was drawn to that. So in today's AI landscape, we face the same challenge. Will we chase performance benchmarks or trust indicators? Will we fine tune for accuracy or optimize for inclusion, security, and human impact? What's your blue and gold? And what trusted signal and what data do you need to embed in your work to make your AI solutions not just intelligent, but trustworthy?

When women lead with trusted data, we don't just build twos. We build belief. We build trust, and we win grand prizes. Thank you for your time, and please put in the chat any questions and comments you might have. And feel free to reach out to me and connect with me. Thank you for your time.