AI Mastery: Navigating the Next Wave with Dell Technologies

Courtney

Hello and welcome back to the Dell Technologies Forum Podcast Series, where ideas and innovation meet acceleration. In this series, we bring together Dell's brightest minds and industry pioneers to help you navigate the ever evolving landscape of technology. I'm Courtney Hughes, your host on this journey of discovery, and today we're taking a different look at artificial intelligence, a game changer that's reshaping our world, we'll delve into its transformative impact, examining how it's not just augmenting, but revolutionizing how we approach technology adoption.

00:00:41:10 - 00:01:09:22

Courtney

With the Dell AI factory, we've already laid the groundwork for organizations to turn complex challenges into streamlined solutions. And joining me today is John Siegel, senior vice president of Dell Portfolio Marketing, whose team has been instrumental in demystifying AI and making it accessible to all. John, it's so great to have you here to share your insights on how Dell is laying a path for customers to embark on their own AI journey with confidence and clarity.

00:01:09:23 - 00:01:11:07

Courtney

Welcome to the show.

00:01:11:09 - 00:01:12:20

Jon

Great to be here, Courtney.

00:01:12:22 - 00:01:35:11

Courtney

I'm excited to have you here. Like so John, in previous episodes, we defined the Dell AI factory with Matt Baker and Doug Schmitt covering what it is and how it works. So today, I'd really like to ask you about its impact so far. From your perspective, can you share how the Dell AI factory helps customers overcome their biggest challenges with AI?

00:01:35:13 - 00:02:02:05

Jon

Absolutely. So looking first of all, just for a quick recap from folks, the Dell factory itself is is really our approach to helping customers accelerate their AI adoption. And the the biggest challenges that we consistently hear from customers around AI option. Boil down to three things really complexity, risk and cost. And so I, let me tackle each of them a bit first, and then I can talk about how we're helping customers today through each of them.

00:02:02:07 - 00:02:21:17

Jon

First is complexity. And we've talked about in the past AI projects can be challenging, especially given the gap in AI skills and talent that most of our customers have. And it also data readiness is a big hurdle for them as well. It's much more complex than they didn't, they expected to access data that is relevant, that is accurate.

00:02:21:17 - 00:02:55:00

Jon

And then be able to feed that into their AI projects. And as we often say, right, good data equals good AI. And yep. And unfortunately bad data also equals it goes bad AI. The second is really around risk. And our customers are concerned about the risks. You know, everything from data and IP leakage today, privacy issues to bias issues and you know, we see time and time again, examples of this, as reported recently in the in the Boston Globe, I just read a local Asian college student asked Gen AI to make her headshot more professional, and unfortunately it changed her race.

00:02:55:02 - 00:03:05:03

Jon

And look, the this is just an example. There's many like this of the consequences of AI gone wrong, and they're pretty well publicized. And like as we both know too, no one wants to make the headlines.

00:03:05:06 - 00:03:05:24

Courtney

Exactly.

00:03:06:00 - 00:03:22:19

Jon

And third is cost. It can be really costly to move data to the cloud and run AI workloads there. We ourselves, we support a hybrid cloud model where we really support bringing AI to your data. So to look, when it comes to, you know, I'm going to cover each of them now. And when it comes to addressing complexity, the goal is to just first of all, make it easier.

00:03:22:21 - 00:03:45:10

Jon

we are helping customers today get started with expert services that guide customers all the way through the process, from building a strategy to preparing their data, etc. and we also really aim to provide that easy button for infrastructure, where we've pulled together the the full stack solutions geared for specific use cases, and we've fully validated them. We pretest them as well.

00:03:45:12 - 00:03:48:19

Jon

A great example of that, Courtney, is the is the AI factory with NVIDIA.

00:03:48:22 - 00:03:49:14

Courtney

Yeah, right.

00:03:49:14 - 00:04:06:21

Jon

Well we have one that that helps customers build their own digital assistants. So we really take it on the hard work so our customers don't have to. And then you know, when it comes to mitigating risk, we're really about ensuring our customer's most valuable data is protected. We have a history of doing this right with our trusted infrastructure and services.

00:04:06:21 - 00:04:27:00

Jon

And I think a great example there that we're doing specifically in AI is is our work with Meta, where we recently helped customers deploy the latest Llama 3 LLM on premises and showed customers how to do it securely through a white paper. And this ultimately helped reduce the risk of data breaches and hyper leakage, etc.. So, really, really important there.

00:04:27:00 - 00:04:45:17

Jon

And then finally on the challenge of reducing cost, you know, look, there's a lot we could spend a time on there. I will say for a recent study, I know if you saw, but a recent study found that inferencing for LLMs can be 75% more cost effective on premises with the Dell AI factory versus the public cloud. And that's that was done over a three year period.

00:04:45:17 - 00:04:53:14

Jon

And this is because it's simply more cost effective for customers when building their AI factories to bring AI to their data than the other way around.

00:04:53:16 - 00:05:03:03

Courtney

Right. So on the topic of the customer's own AI factories, how does Dell help them prioritize the outputs like their own use cases?

00:05:03:05 - 00:05:09:02

Jon

Yeah, another great question. Like we believe it's critical to start with the output or the outcome in mind first.

00:05:09:02 - 00:05:09:21

Courtney

And okay.

00:05:09:21 - 00:05:29:07

Jon

And you know, and their use cases are going to inform that. The use cases we're seeing, by the way, or just across the spectrum, you know, from code generation where developers are using natural language now to, to automate routine tasks and actually write code, to digital assistants, as I just mentioned, to transform self-service experiences to digital twins, you name it.

00:05:29:07 - 00:05:51:03

Jon

So the customers are often coming to us with lots of use cases and we've done it ourselves. Inside Dell. Yeah, we do it for customers as well with our expert services. In fact, we recently helped a customer in six short weeks take what was essentially 200 use case ideas and narrow them down to five. And we did that by helping ensure that they could drive the greatest return for those five.

00:05:51:03 - 00:06:12:05

Jon

And just once we help them gain consensus. So through our services around high value use cases, and we really wanted to make sure that those use cases themselves were focusing on ones that really had common requirements and that delivered really, really good value for them. And by taking this approach, they could avoid waste. They could essentially deliver a much better return on investment and do it much faster.

00:06:12:07 - 00:06:20:18

Courtney

Okay, John. So speaking of use cases, you know, we recently announced our Gen AI solutions for digital assistants. Can you talk to me about that use case?

00:06:20:20 - 00:06:48:01

Jon

Yeah. Look there's a there's a lot of excitement right now about the potential for self-service, you know, improving that self-service experiences for customers. We have, for example, customers like the city of Amarillo who came to us looking for a way to better provide information to their constituents, that essentially speak 60 different languages across the city. And so they were looking for a simple way to provide information through a self-service experience.

00:06:48:01 - 00:07:08:23

Jon

And so what we've done with the new solution, and that we've collaborated on with NVIDIA, is essentially pull together an easy button for the solution. We've essentially done all the hard work providing a joint solution with NVIDIA that our customers can deploy and in in a few short months, help see the impact. And in the case of City of Amarillo, yeah, the impact has been substantial.

00:07:08:23 - 00:07:22:05

Jon

They are able to now communicate information as needed to their constituents. Constituents themselves can can access government services that they couldn't before. And this is a game changer and one that they're going to build on for some time.

00:07:22:07 - 00:07:44:23

Courtney

Yeah I agree with you. It's a really good example of how Dell and NVIDIA are helping customers, you know, accelerate their ideas. Definitely a great example. And so customers are telling us about the importance of reaching both their AI and their sustainability goals. Right. How can Dell, in particular the Dell AI factory help with that?

00:07:45:00 - 00:08:02:22

Jon

Yeah. Look, we have, the Dell has a history, of taking an end to end approach when it comes to sustainability. And the Dell AI factory is a really good example of that. It focuses on delivering AI results and reducing environmental impact. So we really are focused on both. And it does the, the latter in a couple key ways.

00:08:02:22 - 00:08:14:02

Jon

First, it enables actionable insights around sustainability. So our AI ops, monitoring capabilities can analyze energy usage across a customer's IT estate.

00:08:14:04 - 00:08:14:20

Courtney

Oh, wow.

00:08:14:22 - 00:08:36:19

Jon

Yeah, yeah. And actually make recommendations on how they can reduce their data center footprint, if you will. Physical and carbon footprint. So that's something new that not everyone knows. Second is, we help lower energy and cooling costs through our own technologies. We have such as what we announced that Dell tech world's recently, our new power edge 9680 L, which is based on NVIDIA's Blackwell.

00:08:36:19 - 00:08:59:11

Jon

And that's designed for liquid cooling, really helping maximize compute power while maintaining thermal efficiency. So, you know, we have, we are constantly looking at the latest and greatest in technologies to help their and and third of all, we're about helping customers transition to AI with circularity in mind. That means ensuring our customers can upgrade to new and more sustainable, you know, technologies.

00:08:59:11 - 00:09:18:22

Jon

You know, we are certainly on the verge of a major refresh right now in technology, with data centers moving towards AI factories and of course, the the roll out of AI PCs as well. And in the past, the refresh cycles have come typically with environmental cost. Right. Okay. Actually, this time around we have the experience and the opportunity to break that cycle.

00:09:18:22 - 00:09:30:13

Jon

Yeah. And we're hoping to do so by using circular practices like our asset recovery service for laptops. So where we can keep products and materials and use much, much longer and reduce waste.

00:09:30:15 - 00:09:38:00

Courtney

Finally, John, how has Dell's our own experience with AI helped shape that Dell AI factory?

00:09:38:02 - 00:09:57:17

Jon

I would say yeah, a lot. How's that? But let's just say we've we have skinned our knees and elbows a bit on this journey ourselves. And, you know, we're about passing along those key learnings to our customers and I got to tell you, our customers love to hear about our own experience in this area. You know, we break our own AI journey down to a couple phases.

00:09:57:17 - 00:10:19:00

Jon

First is our strategy itself. We've established a chief AI officer position at the company, and we've centralized all the AI initiatives under that at that position and under that office. That office itself is really responsible for narrowing down the use cases, which they did from 800 down to four. Those four happened to be pretty common ones across our customers as well.

00:10:19:02 - 00:10:39:17

Jon

Software development. Customer service, sales, content generation, those are those are the big four right now that we're focused. So the second big phase of ours is implementation itself. This meant, for example, getting our data ready for AI. And that's that's not an easy task, particularly and a large, large an enterprise like ours. So listen, we had to identify where the data was.

00:10:39:17 - 00:10:53:09

Jon

We had to clean it, prepare it, you know, worry about access controls and making sure we had the right security and governance in place. And, you know, but once we did so, we were ready to get started. And it's, it's that that experience.

00:10:53:11 - 00:10:53:19

Courtney

Yeah.

00:10:53:22 - 00:11:11:17

Jon

That's really contributed to be, first of all, our advice to customers around being either data because we've lived it. And we see that whether your data lives on prem or at the edge or in the cloud or even on your own PC, we've had success ourselves in bringing out our data, and we're advising customers that, hey, you know, you can benefit as well, you know?

00:11:11:17 - 00:11:28:14

Jon

And then the last thing was just around the adopt and scaling phase, if you will, that was the third big phase that we had. And this is about putting our our use cases to work and also investing in broad training and skilling across our company. We found that this has really helped quite a bit, and it's been a best practice that we've share with our customers.

00:11:28:14 - 00:11:45:01

Jon

So all these experiences that we've had in really treating ourselves as customer zero, yes, all those lessons learned, those are really baked into our data factory. And this is really ultimately what it's about, which is helping our customers realize their own potential with AI.

00:11:45:06 - 00:12:02:08

Courtney

Yeah. And I'm sure also from your perspective too, that also builds a lot of trust and credibility with the customers. When you've gone through the journey yourself and you can talk about the pitfalls and the lessons that we've learned versus somebody who hasn't done it, trying to teach somebody else how to do it.

00:12:02:10 - 00:12:12:04

Jon

That's exactly right. And I get to tell you, that is probably the most popular topic that we have with customers today, and one that will certainly be talking about a lot at the Dell forums, coming up soon.

00:12:12:06 - 00:12:40:18

Courtney

Thank you so much on some great insights into navigating complexities of AI adoption, how innovative solutions enabled by the Dell AI factory addresses common challenges, streamlines processes, and empowers organizations to leverage AI effectively and economically. And thank you to our listeners for your company on this journey into the future of technology. It's our hope that the insights we've shared have given you a clearer path forward in your own AI initiatives.

00:12:40:20 - 00:13:01:00

Courtney

And for those of you seeking further knowledge, the Dell Technologies Forum roadshow is coming, so look out for that on our website, or you'll also find our online resources delving deeper into these topics. John, thanks again for the expertise shared in this conversation today, which has definitely enriched our understanding of AI's transformative potential.

00:13:01:02 - 00:13:03:09

Jon

And thank you, Courtney. Thanks so much for having me.

00:13:03:11 - 00:13:18:03

Courtney

And to our dedicated audience, your engagement is the heartbeat of this series. We look forward to continuing this exploration with you in our next episodes. Dell Technologies Forum Podcast series is brought to you in association with Intel, Microsoft and NVIDIA.