Please welcome President, Digital Experience, Anil Chakravarthy.
Morning, everybody, good morning. Hope everybody's doing great. And welcome to Day 2 of Summit. It's great to have all of you back. And I really hope yesterday left you inspired about the opportunity in front of all of us when it comes to driving growth from experiences. I'm really truly excited by the innovation we discussed around the Adobe Experience Cloud, and how we'll really power experience-led growth over the next few years working with all of you. We also heard some great examples yesterday of what this looks like from forward thinking leaders, how they differentiate their brands, how they build trust and loyalty with their customers, and how they maximize their business outcomes by really meeting their own customers where they are. We heard from Eli Lilly, we heard from T-Mobile, we heard from Prudential Financial. I'm sure in the Innovation Super Sessions, you had the opportunity to listen to a lot more of your peers and your customers. A huge round of thanks to all of them for doing this. I really appreciate it.
Today we'll shine a light on leaders in other fields, their journeys, their lessons are equally instructive to us as we look forward and think about the next few years ahead in this dynamic environment. But before we dive into those sessions, we want to dive deeper into generative AI. Yesterday our announcements in generative AI generated a lot of buzz. And today we want to go deeper and talk about generative AI.
When we talked about generative AI yesterday, we said, look, along with data and with content we know that AI and generative AI is foundational for achieving experience-led growth.
It will reshape every aspect of what marketers do and creatives do in marketing, whether it's marketing planning, execution, analysis. It's not just marketing content, that's a big piece of it. But journeys, audiences, campaigns, experiences. Generative AI has the potential to transform every aspect of marketing.
So now at Adobe, we have a decade plus track record of pioneering AI innovation. We call it Adobe Sensei. And here's some of the features that we have done. Hundreds of features introduced across our product delivered to you via the products and applications of the Adobe Experience Cloud. Each of them focused on helping you work and collaborate and transform what you do in new ways.
We see generative AI as the next natural evolution of what we've been doing with Sensei. We made significant announcements in Summit. Just to recap a few of those. We talked about Firefly yesterday with Adobe Creative Cloud. David introduced Firefly. We also talked about the Sensei generative AI services in the Adobe Experience Cloud, including our comprehensive vision to really help you build the copilot for you as marketers and creatives to increase your productivity, to amplify what you do from the first spark of creativity all the way through your customer journey.
And we are putting it to work across all of our clouds today.
From Adobe Firefly, which got a lot of buzz for the precision, the power, the speed, the ease to generate high quality images and stunning text effects. We saw a lot of those visuals yesterday. And we just say for commercial use the new Sensei GenAI services that are natively integrated in the Experience Cloud and into the workflows that are in the applications that you already use today. Whether it's marketing copy generation, conversational experiences, audience and journey creation, caption generation. This were all the services that we announced that are already going to be available in the next couple of quarters. But that is the framework for the next few years. So what we want to do today is really go into how we're going to do this, but I wanted to put it in the framework of what we call our AI ethics. How do we responsibly bring the AI to you so that when you use our AI, Sensei AI, generative AI built into our applications, built into our workflows, it really reflects the trust and data governance and the overall principles that Adobe carries? So there are three key areas we focus on. We call it accountability, responsibility and transparency - ART. The ART of AI ethics. And let me touch on each one of them.
With accountability, we're committed to taking ownership of the impact of our work by having processes in place, and making sure that there are resources dedicated to receiving and responding to your feedback or concerns and feedback and concerns from your customers. Ultimately, we are committed to advancing AI in an ethical, responsible and inclusive manner.
With responsibility, we will be responsible during every phase of designing, developing, deploying, and maintaining our AI systems through thoughtful evaluation and careful due diligence. You will see that as you go through any of our sessions, as you go through all of these demos. And we are committed to making these experiences with Adobe's AI productive and efficient for you with control and trusted governance capabilities in the hands of marketers and creatives.
Finally, transparency. We are committed to being open about how we use AI. We talked yesterday during the announcement about how we built it to be commercially safe. And we want to bring our customers into the journey. That was the idea of opening this up with beta and really work with you, our community, to design and implement AI that respects our customers’ preferences, and your interest. And this is exactly how we view the future of generative AI when it comes to the entire marketing process.
So I'm excited for us to give you a deeper look at all of that. So what I'm going to do now is to turn it over to Gina Casagrande for a great demo of the potential of where this could go for marketers over the next couple of years. Gina, take it away. Thanks, Anil.
I'm excited to share the future of Adobe Experience Cloud, where Adobe Sensei GenAI is integrated into each stage of the customer experience workflow. Actionable insights, audiences, content, campaigns and journeys are all automatically generated, helping teams to deliver personalized experiences. I'm going to share this through the lens of a travel and hospitality company. Let's get started.
Adobe Sensei GenAI identifies and alerts the team to a new travel trend. It's always on sifting through mounds of internal and external data. And it's trained on the context of the company's industry.
It surfaces relevant trends and insights in the Experience Cloud as they become predictive.
This trend summary indicates that one of the core audiences is beginning to take more trips alone. And these data stories are automatically created by Sensei GenAI, detailing insights around the audience and making the data easy to understand.
The marketer is guided along as they ask how they can best leverage this emerging trend. Sensei GenAI is like a copilot, highlighting opportunities that drive more bookings, acquire new customers and even guides product decisions based on usage.
The factors driving these predictions take into account the company's context and key business goals. And once a selection is made, a new campaign plan is generated. It leverages the breadth of Adobe Experience Cloud end-to-end, turning insights into action.
This Bird's Eye campaign view is created in the moment along with the content and journey opportunities for this audience. It includes a tagline and a look at the interconnected programs for the booking on trip and post-trip experiences. Sensei GenAI provides an inspirational foundation for the marketing and creative teams, and it accelerates the overall campaign process going from months to minutes.
The marketer is guided and has a great start and wants to explore what other audiences travel solo.
In response, they see these overlapping audiences and digging deeper they want to understand how likely they are to book. And the data is right at their fingertips, so they can make an informed decision quickly. Wanting to extend their marketing reach with these audiences, they make these selections and generate a new plan.
Now, all of this is happening on the fly with AI. A newly generated campaign plan, content and journeys for these audiences. I just want to pause for a moment here, and let's take this all in, because this is going to be a game changer. Adobe Sensei GenAI will fundamentally transform how we do work, and for the better. It will help make teams more productive, by giving them a head start.
Now that we have our campaign plan, let's dig into the creative production and have a little fun. Here in Adobe Creative Cloud, this AI-generated content with Adobe Firefly gives the creatives an inspirational starting point. And it's served up with compliance checks for brand, ethics, accessibility and intellectual property. With Adobe, generative AI is powerful, and it's also reliable and on-brand, and accessible to all audiences with enterprise-grade trust. The marketer can play around with the designs and explore different remixes for additional inspiration.
This is truly unlocking creativity for all here. The designer can better capture the spirit of this campaign and make it more inspirational. And based on this change, new headlines and content are generated.
The designer can make the necessary edits and send it for approval. And then we head to the next stage of the workflow. Now the marketer wants to understand how this is all going to come together for their customers and prospects. Sensei GenAI creates a snapshot of the recommended journey, and clicking into More Details, the marketer can visualize the engagement touch points, and see where the content will be mapped across channels. They can even simulate a journey, and fine-tune it based on predicted performance. This gives marketers the confidence to know that this campaign is going to reach their targeted business goals. Once they've made their tweaks, they can set it live, and once the campaign is launched, it will be delivered across touch points and personalized to each individual as they engage with the brand. And real-time data will start flowing in. And based on that real-time data of the campaign coming in, the results will fuel the model for continuous training and to fuel future optimizations and insights. The team is super excited about the positive results of this campaign, but they're always on the hunt for the next big thing. So, when they see a new alert around sustainability, they're ready to dig right in. What I've just shared is how Adobe Sensei GenAI will fundamentally transform the customer experience workflow: By making recommendations that are accessible and actionable, by providing an inspirational starting point for creative, campaigns, journeys and experiences, and by increasing productivity and efficiencies for teams throughout the enterprise. Thank you. Awesome. Thank you, Gina.
Thanks for a great demo. Really good representation of what would be possible across the Adobe Experience Cloud, leveraging Sensei GenAI services as well as Firefly. So, I want to quickly build on what Gina shared. To accomplish an end-to-end customer experience, like you just saw, you need reliable scale and accuracy of data that fuels the generative AI. So, with Adobe, across the Experience Cloud, with the applications you're using today, with the Adobe Experience Platform, you have access to the signal-rich data sets that are required for generative AI. And as a customer of the Experience Cloud, you cannot only leverage all the goodness that Gina shared, but you'll also benefit from everything that we will build, as the potential of generative AI takes hold over the next couple of years.
To go deeper on Adobe and generative AI, and to take a look behind the scenes, how does this all work today, wanted to learn a little bit more about the technology, I'm happy to welcome Amit Ahuja, who is Senior Vice President of Products for Adobe Experience Cloud. And Amit will be in conversation with two of our foremost technical experts: Anjul Bhambhri, who is the Senior Vice President of Engineering for Adobe Experience Cloud, and Ely Greenfield, who is the Chief Technology Officer for our digital media business. Amit, over to you.
This is awesome. Thank you, Anil, and thank you, Gina. Hopefully everyone enjoyed that demo. It was absolutely fantastic. So, thank you. I'm privileged, I'm honored to be with the two of you, who are largely leading a lot of the technical efforts around what we've been discussing. So, super excited to dive into how we're thinking about this, go a little bit deeper on what's been shared. Ely, maybe I'll start with you. We talked a lot about Firefly with David, Shantanu and Anil. This room is full of marketers, as you look around. I think they are very interested to hear what does Firefly mean to them. How do you think about that? Sure. Thanks, Amit. In short, it comes down to - from the conversations we've had it's around content supply chain. What word we've been using. Yes.
To go a little bit deeper than that, because we're the technical people, from the conversations we've had with marketers, as we've been developing this technology and looking what's out there in the market, what we hear, and you all know this, of course, is it's all about more granular approach to audience segmentation and personalization. That's the name of the game right now. But from what we hear, everyone is limited, essentially, by the speed at which you can create the content to support that work. And so, what's exciting about GenAI for everyone that we've talked to is it promises to be a massive productivity boost to creatives and to marketers, in both content creation, but also content personalization, and adaptation. That's the excitement of it. But the fear, or at least the problem that we've heard from all the conversations is: Great, but everybody out there has concerns right now about the state of the technology today, the state of the market, and whether it's an IP concern, or a licensing concern, or a legal concern, or you name it. So, everybody's excited about it, but they can't really dig in yet. So, when we look at the way the market's going, we think that this is how it has to go. It has to evolve to address these concerns for all of these people to be able to use it and deliver on that promise. That's essentially what we've done with Firefly, is primarily focus on this question of designing it to be safe for commercial use. I love that. And I love how you linked it. We spent a lot of time yesterday talking about content supply chain. I think the fit there is very, very clear. So, thank you for that. Anjul, maybe I'll come to you. Ely talked a bit about Firefly and the marketing aspect. We've also highlighted so much around the broader experience side, which Gina just showed. Maybe you talk a little bit about how you think about GenAI more broadly in the context of experience. Sure, thank you. Obviously, GenAI is a very transformational technology, and we are going to see the impact of these innovations everywhere, including, like you said, in how customers are going to interact with brands. So, maybe I'll start with a personal anecdote. This is my experience with ChatGPT, which I'm sharing here because it highlights the benefits of GenAI, as well as where we still have work to do and things have to be beefed up.
So, my husband and I are trying to remodel our kitchen, and when I was given a task to come up with a plan, I thought I'm going to use ChatGPT as my virtual assistant. So, I asked ChatGPT to pretend to be a design consultant. And, strangely, artificial intelligence is very good about pretending. There's an inception kind of a thing going on here. Sorry for the drift. Let me get back to my example. So, we had very good back-and-forth conversation, and for the most part, ChatGPT really understood my preferences. I wanted a two-tone kitchen. so it recommended the right cabinet colors, the lighting on the island, the flooring that goes with it, appliances, where should they be placed. So, it got me part of the way there. I think, in the ideation part it helped. But then I also realized during this experience that there were parts where GenAI was missing that depth. And what I mean by that is that even though I had maybe twenty such conversations and interactions, it never really connected every interaction about me, or any of my touchpoints, any of those transactions. So the personal context about me was missing. Yes. And it recommended appliances and things like that, but the brand specific products and recommendations were not weaved into that conversation. So the brand specific context was missing. And the entire conversation was pretty textural. So, I couldn't really upload the picture of my kitchen, there were no 2D, 3D maps, no colors. It was all very textural. And I think that's where lays the opportunity that we can make GenAI really very context rich and multimodal. And that, I think, will really help supercharge these experiences even more. And I know, Ely is on the multimodal thing. He is on the multimodal. I think, Anjul, your comments on, hey, it's providing a huge amount of value, but there's still the notion of personalization, about making it more contextually aware. I think that resonates very, very well. And thank you for the personal story. We've all gone through remodels, and we know how that is.
Ely, maybe coming back to you. Again, going back to Firefly place a little bit, we discussed yesterday a little bit around differentiation and how we're thinking about it, and how we're unique. It was interesting, following up that session yesterday, I was lucky enough to be with a lot of customers and that question keeps coming up a little bit. Just around: Hey, how are you Adobe and unique in what you're doing, given there's so many other models, etc. Maybe you can spend just another couple of minutes on that. Sure, yes. There're a couple of points on there. I'll start with the one that I think we've said a couple of times, and I referenced it earlier which is it starts with what we're training on. And that's where a lot of the questions come in about these models which is: Where does the training data come from? What's in there? And that affects what is this thing going to spit out? So, because we've trained on Adobe Stock, on open license content on public domain content where the license expired, we have a high degree of confidence of what it's going to be spitting out and what it's not going to be spitting out, because of the moderation process that content goes through. So if you're a consumer goods company, you don't want your generated content that's coming through a fast generation pipeline, accidentally spitting out content from your competitor, or somebody else using and accidentally spitting out your content. We've designed Firefly so it won't do that. But of course, when you're using it, you do want it to be able to generate content with your brand, your IP, your look, your product in it. And so one of the things that we'll be doing as we roll out Firefly, getting deeper into the products, is we'll be focusing on customization. And on taking Firefly and using your own data, your own content, to be able to customize it for generating exactly the content you want. Now, obviously, that's exciting, it's also a little scary, because now you're putting your data into this stuff, and you're wondering where that's going.
We're your trusted partner already through the Digital Experience Cloud. We already have strong principles of data governance. And that will go into how we allow customization to make sure that when you customize Firefly, it'll be for your use, and your use only. And there's a bunch of different technologies that we're working on to be able to drive that. The third thing I'll say, I'll touch back to what you were saying Anjul, which is: these technologies are amazing. There're a lot of great versions of it out there. We think the benefit is going to come in when we bring it directly into the workflows. So we'll take the core models that we're building, the core technology we're building, and we will be integrating that directly into our creative products, our digital experience products, to be able to get away from the rolling the dice process that you have to go through today of try a prompt, try a prompt, try a prompt. to really allow multimodal inputs to allow iterative, higher control processes that allow you to really get the content you want. And the other thing I'll say is we'll look beyond images. We're starting on images here, but we've already got technology that we're developing in house to take this beyond images to other content types, also to entire templates. So imagine you're creating a social campaign using Express, and you want to use generative AI to personalize and to segment and adapt entire assemblies of content. Not just the image, but also the text, also the content that goes in there. Also the layout. We think there's a lot of runway for this technology to be able to drive a wide range of the types of content you're creating. It makes perfect sense. And that actually was the perfect segue because I think you alluded to: Hey, one of the things we're really focused on is bringing this power directly into the workflows. And you mentioned the Creative Cloud side, but you also mentioned the Experience Cloud side. Anjul, maybe back to you. Gina showed this demo, showcasing how we're thinking about GenAI, much more broadly, to your point, Ely, than just text, copy. We're thinking about audience, etcetera. I think the question that's probably top of mind for a lot of folks: When does that show up? How does that show up? And again, I know, you're leading a lot of that, maybe you can just speak to that for a couple minutes. Sure.
I'm a denier, obviously, partnering on all of this. So you heard from Anil and Shantanu yesterday, that we are in Experience Cloud going to be lining up, releasing a series of copilots. And what I mean by that is that think of it as a large language model. It's a psycho virtual assistant, a productivity tool, which is really pre-trained on all the customer data that you have about your customers. So it is very rich from a context standpoint. And Sensei GenAI, as well as Firefly will definitely be the backbone of this.
And it'll be pre-trained, like I said, on the data that you already have collected about your customers. And it could also be pre-trained, if you like on your brand-specific information. It could be products, it could be documentation, something that is unique. It could be brand content which is specific to your brand. So with that, we are going to deploy this in kind of two modes, if you mean. There is an external facing copilot which is what is going to help with how customers will interact with brands. So it's going to kind of sit in the middle of that interaction.
Then there are going to be internal copilots, which, like you mentioned, Amit, and like Ely, you said, that they are really going to guide the marketing team. They'll be the assistants for the marketers, their teams, to help them really design experiences, target the right audiences, and really amplify those personalized experiences.
Maybe if we click down one level, specifically in the products, I can't resist saying that, in real-time customer data platform, think of this marketer as, it's going to leverage as well as extend the 360-degree view of the customer via the unified profiles. And the conversations will be powered by the profile.
And, when you have the conversation, the insights, the attributes, what was learned during that conversation will feed it back into the profile. So remember, in my example, I was saying that the personal context was missing. Now, the personal context comes back into the picture with that unified profile. And if brand specific information was added to this, then that profile plus that brand information is making that conversation even richer. So that's the copilot that's helping with the conversation. Now for the marketers, it's going to help generate audiences. Because from the conversation, you may learn that - I stick with my example, the kitchen remodeler. Now, there's a new segment, a new audience that gets generated, which I didn't have before. So that's in the real-time customer data platform. In Journey Optimizer, it's going to help generate journeys, emails, campaigns, channel content very quickly, so that the marketers really can iterate, test, optimize. This didn't have to approve all of that. So there is always a human in the middle to make sure that everything that is being generated the brand ultimately has to put their seal of approval, that this is good. But since everything is generated, it's easy to hit that Approve button.
In Customer Journey Analytics, the copilot will really provide that natural language. It's a very intuitive, natural language interface to Analytics for both visual and text. And if you look at everything here for Marketo, we pre-train the copilot with the B2B context, with account information. And you get everything now with the context of when a business is talking to another business. And then of course, like, partnering with Ely and team, in Adobe Experience Manager for Firefly is just a natural... I think that's where we showcase the integration with Express. Exactly. Leveraging Firefly, I think we showed that as well, which I think there's a lot of excitement there. Yes, I think you... For content generation, layout generation. And then last but not least, and that's where I would invite all of you to partner with us, that working with you we can really generate brand-specific copilots. I mean, I think that's the game changer. Because otherwise, the conversations are very generic. So those deep insights, that deep understanding is going to come from the brand-specific information. And I think that's how we can collectively wow all of you and wow all of our customers. Awesome. No, thank you. I honestly wish we could go on and on. The light is blinking at me that we cannot go on and on.
So we will not. But Ely, thank you very much. Anjul, thank you very much. Very exciting. And back to stage. Thanks so much. - Thank you. - Thank you, guys.
Awesome.
Thank you, Amit. Thanks, Anjul and Ely. Excited for the possibilities ahead as you build this generative AI technology, integrate them into our applications and workflows. And I love the idea of this brand-specific copilot or business-specific copilots for your business that are really aware of the context of your customers, your principles of how you want to apply the AI, and make sure that it's used in a very responsible and transparent way, specific to your brand, and specific to your business.
So next up, I'm really excited about the next discussion. I want to welcome our chairman and CEO Shantanu Narayen to the stage for a conversation with Lisa Su, the chair and CEO of AMD. I've been really looking forward to this. Shantanu, welcome on stage.
Good morning, it's so good to see all of you back today. But I'm delighted to have Dr. Lisa Su, chair and CEO of AMD, joining us today at Adobe Summit. She's led the company through this absolutely incredible transformation.
Over the course of Lisa's career, she has received numerous honors, including being named a fellow of the Institute of Electronics and Electrical Engineers, elected through the National Academy of Engineering, and I think most impressive, recognized by the IEEE with its highest semiconductor honor, the Robert N. Noyce Medal. And she was recently appointed by President Biden to the President's Council of Advisors on Science and Technology. And from a number's perspective... Yes. Thank you.
From a number's perspective, when you took over as CEO of AMD, the revenue was about 4 billion. It's approaching 24 billion. The market cap was in single digit billion. And today it's 160 billion. So just an amazing, amazing accomplishment. Thank you. Welcome Lisa, thank you for joining us at Summit. It's wonderful to be here, and this looks like a wonderful crowd. Well, since taking over the helm in AMD, you've done this amazing transformation. So maybe you can talk a little bit about the turnaround and AMD for us. Well, absolutely. Again, Shantanu, thank you so much for being here. It's really my honor. And if I just tell you a little bit about AMD. First of all, I'm an engineer like you. So we like to build things. And at AMD we're all about high-performance computing, how can we push the envelope on computing? How can we do more? How can we help all of you, as creators and as businesspeople and as marketers, do more, faster, better? And in technology, it's all about making long-term bets. And so it has been a wonderful journey, and it's all about how do we get more technology out there to solve some of the world's most important problems. Till a few years ago if people were talking about chips, they would probably have baked or fried, as sort of the question. But after what happened with the global chip shortage, everybody is wondering, what is it that caused this global chip shortage? So maybe you can touch on what really happened and how do we address this. I think Shantanu is saying that chips were not sexy. Is that what he was saying? Look, I've been in the chip business for 30 years. It's always been these little things that are under the covers. You didn't have to care about it. And then one day in a flash we decided we need to live and do business and communicate a different way. And so it was: How do you work from anywhere? How do you school from anywhere? How do you communicate when you can't travel? How do you get all of this technology? And we had this incredible surge in demand. And finally everybody understand, wow, chips are not just important. They're actually essential and they can really change the way we do things. But also, they're critical for all the services that we need. So it's been an amazing ride. I think the most important thing now is that I think we're much more aware of how important it is to have access and resiliency to all of critical technologies, including semi-conductors.
What's absolutely fascinating to me, when you are at a company that's creating something that's at the heart of everything that people are doing, how we work, how we live, how we entertain, how do you continuously innovate and how do you think about, how the world is going to continue to engage with chips? I love being in this audience because I know there're so many creators here and we're always thinking about: What's the next big thing? I think, in our industry it's about really long-term bets. So, somehow, we have to look into the crystal ball and say: What do we need five years from now? What does the world need? How can technology really change the way we do things? And then make some of those bets. And frankly, the technology that we're building today, we decided the elements of it three to five years ago. And I'll just give you one example, Shantanu, of some of the work that we're doing together. When you think about what our creative professionals have to work on and the technology that they're working with: If our computing technology with your software can take the time to render something down from 30 minutes to 5 minutes, just how much more productive can we be? And those are the types of technology bets that we make with the idea of, we can really just change the rate and pace of innovation. This is Adobe Summit and Adobe Summit for us has been, Lisa, all about customer experience management and how people are engaging with their customers. AMD is at the heart of, as we said, this entire computing revolution. But maybe a little bit about how you think about marketing, what AMD does, you're an ingredient of every computing experience. And how do you think about B2B marketing? And how is that evolving? It's a great point actually. I think about it all the time because we want to reach broader and broader groups to understand the power of technology and what it can do. The way I think about marketing, Shantanu, is, it's all about authenticity. We have to have product truth like what are we really trying to do with the company? What are we trying to do with the products? And then that's what you do with the brand. We just recently, actually, given how broad our portfolio has become over the last four or five years, we actually just recently introduced a new brand campaign. And our campaign is around the thought of together we advance Because there're two key principles for us. One is innovation. So as I said, it's about technology innovation. And then the other is collaboration. We believe that you can't do anything yourself in this world. Everything is too complicated. You have experts in each field, whether hardware, software, market professionals, business analytics, all of that. And you need to figure out how you bring those together. I think the authenticity of the brand is, how do we innovate and collaborate together. And by the way, I use a lot of your products in my marketing backend. Certainly we appreciate the deep partnership. But I think the overarching thing is, how do we translate what is perhaps an esoteric concept into something that everyone can digest. You know, we've been spending some time talking about some of the new things that Adobe announced, generative AI, but for the audience there is clearly so many buzzwords that are out there right now, whether it's Crypto or Metaverse or AI, or generative AI or ChatGPT, autonomous cars, just to name maybe a few. I'd love to get your perspective on how you think about, how these trends are going to impact the future? And maybe a second question: Is there one that we're not talking about enough that you feel like is going to become the next buzzword? Well, I think this is the week, the month, maybe the year of AI. I think you can't go anywhere and not talk about AI, and congratulations on the Firefly announcement, the Sensei announcement. When I think about AI, I think about it in kind of two pieces. One is the AI that's under the covers, which is the data analytics. How do we become smarter with all of this tons of data that we are generating? How do we make most use of it? But I think the piece that has really unlocked potential, and I know you guys have talked about it here over the last day or two as well, the generative AI concept of it's sort of like your little personal assistant or your copilot helping you solve whatever your problem set is. I mean, that's just incredibly powerful. Now the technology that we're doing around AI is training these incredibly large data sets and then allowing people to access that data. You need an incredible amount of computing power, tens of thousands of GPUs and CPUs. But that capability, I think, just unlocks new things. And when I think about, even my dad is talking about ChatGPT. Everywhere you go, it's like: What can you do with that? I think you can see that it's capturing people's imaginations. And the key is how are we going to use it for business advantage? How are each of us in our world going to use it for business advantage to make us a bit more productive? I think we were talking a bit about it's so early. I mean, you're in beta phase. There's a lot of work to really see how AI can unlock the power, but you can see how it can make us 50% more productive or 80% more productive and think about how we can use those cycles now to do so many other things. So, you're second question: What has it not done? I think we're just at the very beginning. I think we're at the very beginning. Now for my team, we build chips that, our largest chip now is like 100 billion transistors. and it takes us about three years from start to finish. Now, if I could cut that in half, or if I could make sure that it's right on day one, it has just tremendous business value in that. So all of us are learning together. It's a complete race. Would you agree, Shantanu, that this is the one place where I see everyone kind of leaning forward and leaning in and willing to experiment? I would. And I think we're all also tackling with the responsibility of what AI ethics means. Just maybe your thoughts, because one of the concerns is that I did want to check on whether your father, dad's checking up on how Lisa is doing on ChatGPT? Maybe a little bit on the ethics and the concerns about does this replace humans versus augmented? Just your perspective. Yes, I think there's definitely a need for let's call it safe AI. And again, I think, you guys have been very thoughtful in how you've thought about it. We're also thinking about it: How do you ensure that the data that's being trained is not biased in some way or is not leading us down the wrong path? But my view of the world is it's not a replacement at all. The creative genius of everyone in this room can actually just be even a little bit faster, a little bit better, a little bit more capable. We're not talking about replacement, but we're really talking about augmentation and productivity. So, it's going to be a fascinating era over the next ten years, I would say. If there are two themes here Lisa, one is about how technology is going to change, the other one is about leadership and culture. And so I know you've talked a lot throughout your career and even after you took over the helm of AMD about this adage of run towards problems. So maybe you can talk a little bit about change and how do you accomplish change in organizations? Well, it's a wonderful thing. I mean, we all work so hard. By the way, I think we wake up every day and we say we love what we do, we might as well work on something that's really important. And so running towards problems, using technology for good, finding a way to partner, to solve something that you never thought possible. Those are the things that have guided me. When I think about what we do at AMD and how we bring people together under this notion of joint collaboration, it's how can we take one plus one and make it greater than three and really use that to motivate? So it's not about what are you doing with the technology? It's what problem are we solving for society with the technology? You've been an incredible role model for so many folks, Lisa, and maybe just a little on championing, championing diversity, being a role model, and how you think about that in today's work culture? Well, the one thing that I would say is, as we go through life, it's all about the opportunities that people give us. I was very, very fortunate. I grew up in an environment where people were very open. It didn't matter how old you are, how young you were, where did you come from, male, female, all of that. You wanted the best ideas. And at the end of the day, that's the most important thing about diversity, is how do you get the best diversity of ideas so you can make that one plus one be greater than three. And what I try to do, and I think what we're all trying to do as CEOs of large companies, is to create the opportunities and the pipelines so that people get a chance to learn and experiment and fail and make mistakes, and really have that diversity of ideas come together. One of the things that struck me about what you talk about the purpose of AMD is together we advance. So two questions there. Maybe one about purpose and the role that purpose plays in an organization? And maybe second about partnerships, because you use the word together and I assume it's employees and partnerships. So a little bit on both. So when you think about purpose, and let's talk about technology for good, people don't realize how much you can accelerate if you really take technology to its limits. And I'll give you another example. During the pandemic, I think those first few months we were all figuring out how are we going to run our businesses? And then we were also going to figure out how do we help? How do we come and help the global ecosystem kind of solve this problem? And we did things like we came together across the industry ecosystem and said, let's donate a ton of computing power such that we can accelerate the rate and pace of vaccine research and all of the therapeutics and the learning around that. We did that in a week. In normal rate and pace that might take like a year, with all of these various things. But it was the purpose that we had a common goal of trying to solve problems. I mean, some of my most interesting technology projects are things like how do we use tech in healthcare? And I know you're also very personally involved in healthcare, but there's so much more we can do, Shantanu, if we really took the best of technology capability, the best of medical professionals, the best of systems and software folks, and put them together and say: How can I do this much faster and much more efficiently? And that's what I believe what the purpose is. There're lots of things that we do for our business, but there's so much more we can do when we come together in partnership across various disciplines to learn and make the world a better place. As a giant technology company... I don't know about giant. Maybe you're giant.
How do you think about when you're both partnering as well as potentially competing with some of the companies, just maybe how that's evolving for you? It's a great question. I think, more than ever, partnership is king. And partnership is king meaning we're all specialists in certain areas. Like when we can put the power of Adobe engineers and AMD engineers together, we will create better content creation capability. We'll create better marketing capability. We'll create better platforms. If we have our teams operating together versus operating in silos. Like, we each do our things. And sure, we're also large enough. There's a lot of competition out there. But more than anything, I think, open platforms will really help us really take the best of breed. And we're getting to a place where what we're trying to do is bend the curve of what's capable. And I'd like to say that because people used to say: Why do you even need the next generation anything? Like, isn't it good enough? And the truth is: no: it can be so much better, it can be 10 times better, it can be 100 times better. I mean, that's the power of partnership, and deep, deep collaboration. I'm sure I speak for everybody when I say that nobody's ever said: My PC is fast enough. Or: My networks fast enough. And so we're all willing to consume whatever power you're going to provide. And maybe one question on: We all have our pet projects at the company where we feel like there's going to be something. Anything exciting that you want to share with us in terms of what's going to happen in computing? Probably the most exciting, we all have our favorite things. But one of my most favorite projects is we're in the middle of building the fastest supercomputer in the world. Can I wear it on my hand? It will take a very large building right now. So the current fastest supercomputer in the world is at Oak Ridge National Labs. It's based on AMD technology. There's another one that's coming in this coming year that we're in the middle of building. And why are these things fascinating? Because you get to break every problem. You break everything, when you have to kind of aim at the top. And what it's being used for is it's going to be used for significant government, as well as private research. And it's one of those fun things that you get to do. And now, my goal is to take that and make sure you have it in a few years, and run your stuff on it, too. Well, that's awesome. We can't wait. As I've always told you, we'll consume as much power, and as much capacity, as you give us. I'm counting on that. Yes. I like to end sometimes with a word association. So I'll just give you a word and whatever one word comes to mind, Lisa, maybe you can respond with that. I have no idea what words you're going to tell me. - No. - So okay, let's see. Semiconductors.
Essential. New York. Home. MIT.
Lots of geeky people.
For those of you who don't know: Lisa has three degrees from MIT, including a PhD, which is... Very geeky, high geek factor. Yes.
Culture. Humble.
Artificial intelligence.
The most important thing over the next 10 years. Impact. Impact? I think it's personal. What is our personal impact? And lastly: Lisa Su. Your friend. Thank you so much for being here. Please, thank Lisa.
Thank you!
Thank you, Lisa and Shantanu! We really appreciate what Lisa shared about authentic brands, the importance of partnerships, and collaboration, and for her, personally, about running towards problems throughout her career and the success that it has brought for her, and for AMD. Lots of lessons for all of us. That's wishing you the rest of an incredible day today. And make sure you're back here at 5:30 for Sneaks, where you will get a peek into what's coming from Adobe's labs with our fabulous hosts, the comedian Tig Notaro, and Adobe's own Eric Matisoff. Then we'll head to the bash with music from Macklemore and Rev Run. Thank you again for sharing your time with us. Have a wonderful day. Thank you. -