Transcription
[music] Hi, I'm Illy from SCSP.AI. For my next episode of Memos to the President, I just hosted Jensen Wong for the second conversation in less than 12 months here in our office space in Crystal City. We talked about layers of AI. We talked about China and expert controls. We talked about AI adoption and open source. Take a look at it. It's an incredible conversation ahead of our biggest event, AI Plus Expo, which takes place next week, May 7th through 9, at DC's Convention Center. If you haven't registered yet, please do so. We expect 20,000 people. Incredible programming. Register. But first, watch the episode of my conversation with the CEO of Nvidia, Jensen Wong. All right. Uh, welcome back. Uh, I'm here sitting down for the second time in less than a year with the founder and CEO of Nvidia, Jensen Wong. Welcome.
>> Very happy to be here.
>> It's it's an honor to have you, uh, Jensen. And, uh, you know, it's incredible. I was watching our podcast from last July. So many things have happened. So many things have evolved in the AI space. uh but let's start by talking about the state of AI first. You talk about the AI scale versus computer scale uh and how today's world what does that mean and how it's different from the computer age. Can you elaborate that a little bit?
>> Uh we're in a new computing paradigm. uh artificial intelligence if you break it down to the basics of computing. The way we used to do computing is essentially retrieval-based computing. We would pre-record information, take a picture, record a video, uh write a story, do a brief, so on so forth, and we put it online. That's why you would put it in a data center, a center of data. And then based on whatever retrieval method you use, if you're shopping, it's probably recommender system. uh if you're doing search uh it recommended a list and you point at you click at something um or if you use YouTube it just feeds you whatever it feeds you and so uh then once you select it would go and retrieve that file for you and if you're a shopper then it's very likely that based on your previous preferences and whatever context it is where you are and what from what other link did you come from right all the cookies you left behind mind, it might use that information to recommend one of maybe 16 versions of ads that's already created for you. It's all retrieval-based computing. So far, all of the way we largely use computing, we create content and we retrieve content.
Well, the [clears throat] way that we do uh computing now with artificial intelligence is based on your context, based on your request, based on your intentions. It would then generate from some initial seed of information. It could be grounded in some fact. It could be grounded in a piece of report and it generates information for you for the very first time. And every single time it's, you know, largely different. And so it's generative computing. It used to be retrieval based, now it's generative. The benefit of course is that every time you use computing based on your context, based on the changes of ground truth, based on your intention, intelligence would be will be applied and provide you the information that best suits you. And so that's makes sense. Okay, that's the big breakthrough. And the reason for that, we need so much more computation instead of a bunch of storage. We need a lot of computers. That's the fundamental difference. Now of course the the big breakthrough um is because of that we can now we have to learn how to perceive information of all kinds. It could be text, it could be video, it could be right images. We have to reason about what it is that you're asking me the context and your intent. And then I have to plan to answer your question accordingly. It could be uh providing you an answer or providing you a summary or maybe write you a brief based on all the things that I've researched or even do some shopping for you. And so we went from retrieval-based computing to generative and on top of that we now made it possible for us to uh do tasks.
>> Right?
>> And so artificial intelligence is really about that journey.
>> Right? Um I want to ask you a personal question here. When was the first time you saw a chatbot and how did you react to that?
>> The first time I saw a chatbot was shortly after Nvidia built Megatron. uh before the first large language model was announced, Nvidia announced with Microsoft the first large language model. It was called Megatron. And Megatron was about 400 500 billion parameters. The technology that we invented to make it possible is pre-train pre-training the the models and together between us and Microsoft we pre-trained a really large model. at the time this is uh probably late 90 uh mid 91 late 91 okay so this was before chat GPT and the thing that was that was uh uh great about it of course was it it encoded it was able to memorize a lot of knowledge but the thing that we didn't invent that um open AAI invented to make chatbt useful is the concept of alignment reinforcement learning human feedback back and to and as a result, you know, you you give it you give it a prompt, the Megatron, you give Megatron a prompt, it would generate just a it would start spewing off all kinds of things that it was encoded in its memory and it was just nonsense and was really it wasn't very useful and and it was we were waiting for another great invention was reinforcement by human.
>> Uh we have a segment we do every week here in our office called
>> But the first time I saw JPD I was mind-blown. It was so I was so happy to see it.
>> Yeah, we have a segment called AI in the office here. We record every week one of our staffers. What did they do with AI last week? What did you do with AI last week that really was interesting? Uh yesterday I was using it to finish my shareholder letter and um I and so I'll I'll outline something and then I'll say go off and uh read everything that I've already read I've already I've already said and all the things I've already written all the keynotes that I've already done and based on this based on this outline just populate it with things that I've already said just give me a basic framework and then from that uh I might take it and refine refine that or com that wasn't very good. I I'll rewrite it alto together. Um and then I would give it to to the AI again to ask it um to uh make it tone consistent or you know highlight some things that it could it could do a better job on.
>> Right. Um, going back to Nvidia in this town
>> Writing is so hard. I hate writing and but you know with AI it was it was tolerable.
>> Yeah. Um, Nvidia is more than just a chips company. Um, it builds an entire ecosystem and when you look at what you do is you not only invest you know in building the ecosystem that got you here but you also build an ecosystem going forward. So how would you describe in Washington what does Nvidia do? What is that ecosystem that you have built?
>> We build the computing infrastructure for modern AI. And so if if describing modern AI is the new computing paradigm, then we are the computing infrastructure, the computing fabric that makes that possible. A computer consists of the chips, but the systems, the system software, the algorithms where it's computing. It could be in a data center. It could be on prem. It could be uh inside a factory. It could be in a base station like a like the work that we do with telecommunications. Um, with uh uh it could be you know in a car and so it doesn't matter really where computing is done but we create the computing infrastructure necessary to do that. uh each one of these different applications also have have different algorithms and our job is to translate the intention of the computer of the science to best fit into our architecture and so NVIDIA is really this collection of software middleware essentially the operating systems of these domains that um makes it possible for people to do all this computing and and then we we then connect all of that to the world's ecosystem of partners whether it's uh Lily for drug discover discovery or a caterpillar and in um uh energy generation and and construction and uh heavy equip heavy equipment and you know the number of companies we work with United States is basically everybody.
>> Yeah. Um, you have talked recently about AI being a five-layered cake. So I want to get into each one of the layers and I want you to tell us we're in Washington. What are the what where do we stand in each of the layers? Uh what are the things we need to do? What are the things we are ahead?
>> AI is a new computing paradigm. It is also a new industry altogether. And the reason for that is because the way that AI operates is so fundamentally different than the way computers operate. It no longer uh stores the information exactly as you described it and retrieve it from storage from a from a disc drive or SSD. We now generate it and the generation output is numbers and those numbers are essentially tokens. We call them tokens. No different than than well tokens just floating point number or some number. And we take those numbers and we reformulate them into the output that you desire. And the output you desire could be text, it could be video, could be images, it could be sound, you know, kind of like kind of like Tang, you know, we it's in powder form and we add water to it and turns into orange juice and and so so we reformulate it in such a way that you can use well that token generation process requires really large computers and so the first so and computers need energy and so the way that the this industry has formulated is one energy on the bottom whole bunch of chips and systems computers and these computers are are fairly gigantic. I mean, you know, they're they're the size of they're the size of football fields and um and and there's a lot lot of them. And so the next layer up from that is is uh um we call infrastructure. Infrastructure is land power and shell, but also the software for cloud services. And then on top of that is models. And on top of that is really the most important part for the United States and and every single country is the adoption of the technology. And this is in fact the layer that I'm most concerned about. There's one area that that really worthy of us spending time to talk about is making sure that that we are mindful about on the one hand making sure we have proper guard rails and keep people safe in the application of the technology but to ensure that United States is at the front at the pioneering front of applying the technology because it drives so much productivity and prosperity and and technology and as well as economic leadership. We can't afford for another country to leave us behind. We were the front runners of applying technology in the last industrial revolution. We need to be careful not to be the last in this industrial revolution. And so the five five layer cake starts with energy, chips, infrastructure, models, and even models. Uh we we of course think about large language models which is the things that we can engage easiest the most. But remember AI can represent information of any kind. And some of the most important information we represent isn't language and numbers, but biology, chemicals, physics, articulation, animatronics. Um, all of these types of information is represented by these tokens. And it's really important to recognize that when we think about AI, don't just think about the chatbot and the services we're talking about. There's enormous industries that are adjacent to it that don't get the voice that some of the AI companies get, but they're super important to the future of our nation. And so the AI models and the applications on top.
>> Can we talk a little bit about the energy? Um, obviously all these models require require a lot of energy going forward. Um, we have a lot of hurdles in that space from the grid where we are from building the uh necessary you know pathways uh we work a lot on fusion um but you know solar so there there's a variety of way how to get to the energy needed how would you assess our energy situation right now?
>> Well, that's really important question. I think first of all you got to take a step back and ask ourselves do we want to reindustrialize the United States. Do we want to bring back to the United States this entire sector of the economy? This entire sector of labor that we consider essential to having a properly shaped society and a properly shaped economy. We've become we've become a nation and an economy where unless you get a four-year degree, unless you get a master or PhD, you're going to get left behind. And that's that's unfortunate and unnecessary. I think we can all agree that that's unfortunate, unnecessary. We have to re-industrial and re-industrialize this country. For the first time in a generation, we have a market force that's incredibly powerful to drive the reindustrialization of our country. AI will cause us to create several plants. The first plant is chip plants. Um, we're the largest AI company in the world today. We committed half a trillion dollars of consumption so that we can bring the supply chain from the east into the west back to the west so that we can build chip plants and packaging plants, computer plants so that we can build all of the manufacturing necessary for Nvidia's AI supercomputers to be built here and used here. So the first plants are chips. The second plants are associated with the computers themselves. And the third putting these computers into AI factories. Alto together we're talking about trillions of dollars of manufacturing, high-skilled labor jobs. We're going to create enormous amounts of manufacturing opportunity here in the United States. That I think is the first thing we have to confront. Do we want to be that country or not? Do we want to be that kind of society or not? we want to be that economy or not. Now, if we decide the answer to that is yes, then we obviously need energy because you need energy to transform atoms in one form to atoms in another form. You guys understand what I'm talking about? You have to you have to break covalent bonds and you have to make covalent bonds. And in order to do that, enormous amount of energy is necessary to change the phase of matter. And so the whole concept of manufacturing is about energy. And so we need energy in this country in order for us to re-industrialize. And
>> As you build these
>> We are manufacturing.
>> You ask me a question. I'm sorry. And and what is the what is the state of our our capabilities?
>> Right.
>> Clearly we are behind because of all of the policies we've had in the past. All the concerns about climate change caused us to of course underinvest in energy. Now we want to be conscious about the environment. Everything has to be balanced. And so now the question is what do we do going forward? The first thing that we we can do is we can modernize the grid. We can make the grid more efficient. We overprovision in the grid. As you know, we have to we have to make sure that our grid can take care of our society and all of our infrastructure on 12 of the worst days of the year. The rest of the time is overprovisioned by enormous amounts. And so the question is what kind of service level agreements can we come up with so that the power utilities can provide excess energy when they can and during the times when they can't now we have to provide for subsequent backup energy. The backup energy can come from solar. It could come from nuclear you know all kinds of sustainable energy in the future. But we have an opportunity right now with this incredible market-driven force to use take advantage of this opportunity to one to make sure that United States become a manufacturing nation again create enormous amounts of manufacturing labors and jobs and high-skilled and high high paying jobs. Um, and then the second is to use this opportunity to enhance the energy system of our country. Um, in July when we last spoke, you talked about the next wave of AI being Agentic AI. Um, it's May now and we're living it. Um,
>> My goodness.
>> Yeah. Can you talk a little bit about how do you assess the state of Agentic AI right now?
>> There's a lot of conversation yesterday we've had here and and in town about the state of Agentic AI uh you know the successes, the failures and what's happening right now. You know the the the big breakthrough from large language models to chatbots was reinforcement learning human feedback. The large breakthrough from large language models to agentic systems is a system called harnesses. The model itself advanced no doubt the pre-training is improved. The reinforcement learning is improved. Teaching it how to use reason better and tool use all of that improved. However, the big breakthrough is the concept of harnessing these large language models so that they have connection to um ground truth so that they could do research that they could use the web browser so that they could reason and and have memory and improve itself and and communicate with others. And so these agents um in the last six months made enormous breakthroughs. Almost all of them perform incredibly well. Uh we're big fans of Codeex. Codeex and Cloud Code both are incredibly good. Codeex 5.5 just came out >> huge breakthrough big leap and cloud code obviously incredibly good. The the thing that that
>> Um, we can now do almost the you know the vast majority of software tasks are now completely automated and meaning that we don't have to do the programming ourselves. The one the one interesting observation which was the prediction was because of agents coming out all software engineering jobs will be gone that the first thing you should do is don't whatever you do um plan on being a lot of things in the future but don't be a software engineer. Well it turns out the number of software engineers we're hiring is increasing. Every company is increasing. The number of software engineering jobs is increasing. And the reason for that is this. It's a very big idea. This also happened in radiology. uh 10 years ago it was somebody predicted that the first job that's going to go the first industry that's going to radiologists that job is gone forever and the reason for that is computer vision was going to completely transform studying these images the the prediction was 100% right AI has now permeated every aspect of radiology the only thing that was wrong was the prediction radiologists are in short supply and so the same thing is happening in software engineering the reason for that is is this in our jobs. The task that we do in the case of software engineering, the task is programming. You could argue it's kind of a fancy version of typing. And so the task is programming, coding. However, the skill, the purpose of the job is not programming. The purpose of the job is not coding. The purpose of the job is innovate, solve problems, connecting with collaborators, um find problems that exists exist and solve it. Find problems that nobody's even expressed. It's called innovation. Connecting unrelated things, creating something new. That's the purpose of software engineering. And so our our engineers are their purpose in life is to innovate, solve problems, you know, move the company forward. It includes coding, but coding is not their job. Coding is their task that they, you know, some of them do in their jobs.
>> Um, the next wave you've said is going to be
>> AI is creating jobs. Anybody who is saying that AI is wiping out jobs is scaring people and is scaring people out of precisely the jobs that I need. The one thing that I hate for us to do is to tell all of the young people don't be software engineers because it turns out I need them and hospitals don't tell people don't AI researchers should not tell people to stop being radiologists because humanity needs radiologists. The radiologist's purpose in life is to diagnose disease. Reading scans is a task that they do in service of diagnosing disease. So if you if you separate the purpose from the task in a job, it'll help you think through it better.
>> Um, as I was saying, uh the next wave you predicted was the physical AI. Um, 10 months later from where we last spoke about this topic, where do you think we are right now in physical AI space?
>> One of the best breakthroughs in physical AI, the first the easiest one is uh self-driving cars. Robo taxis are here. That that problem, the science is now completely solved. Now it's a lot of engineering and uh and even the engineering is just around the corner. So we're going to have robo taxis.
>> Have you ever used one of them by the way? Just
>> Oh yeah, sure. Sure.
>> Yeah. How was your experience using?
>> Excellent. Nvidia makes robo taxis. We created this software called Alpio. It's the world's first thinking robo taxis. The thinking car meaning it can come up to a circumstance that has never seen before and it would reason about it the way it reasons about it decomposes it into, you know, more mundane things. Oh, I've seen this, I've seen that, I've seen that. In composition, I understand what's going on. And so, you know, I I know exactly what to do. And so, the reasoning system uh has really expanded the capabilities of these cars.
>> So, the robotaxes are here. What about human robots?
>> Right around the corner. Um, and the reason for that is if you could imagine this, you guys know this, if you could prompt a um uh prompt uh a video generator. And so, Chad GBT has a new excellent video generator. And you say, I want to generate uh somebody picking up a coffee cup and taking a sip. And it would generate my hand picking up the coffee cup, knows exactly where the coffee cup is, and and picks it up. And if I can generate a video doing that, why can't I generate the robot to do it?
>> Right?
>> And so so I think you could imagine based on this, the other capability must be around the cap must be around the must be around the the corner.
>> And and so do you think this is like 3 to 5 years? This is one to three years. What's the time frame you envision you know these robots really becoming more mainstream than we're seeing right now in like
>> The big challenge right now is partly the AI model but a lot is associated with the mechatronics the motors the hands um uh the structure you want to be on the one hand light on the other hand it has to be strong and so you want it to be light because if something were to happen it tips over
>> Uh having something that weighs 80 pounds tipping dipping on you is probably okay. Having something that weighs 300 lb dipping on you probably isn't. And so, so there's a very big difference between, you know, the current state-of-the-art versus where we need to be. And we're going to get there. And so material science matters, motors matters,
>> Um, battery technology matters, uh sensors, um and so all of these things matter, and then of course there's the the AI itself.
>> And when do you think uh Olaf will become available at Target? I'm not I'm not hoping for Olaf, but you know, the the uh the other robots that I that I have on stage usually with me from Disney, uh you know, they're they're incredibly cute. Who doesn't want their own R2-D2 growing up with them?
>> Um, let's move to another serious topic. We're in Washington, uh Jensen. Um, we got to talk China and export controls. Um, you have been a proponent of us selling chips to China? Um, there's a big criticism there's a big opposition to that point about us not enabling our main competitor to get ahead of us. How do you address that criticism?
>> Um, the easiest way to think about that is to remember that AI is a five-layer cake and United States America should win in every aspect of AI. We should be the world leader in energy production and this and energy technology production. uh we should uh be the world leader in chips. We should be the world leader in infrastructure. We should be the world leader in models. We should be the world leader and absolutely absolutely cannot afford to lose this one to be the world leader in AI application and AI adoption. Uh the first thing that we should do is we should enable and encourage every single one of the the uh layers of that cake and all the industries associated. Go lead the world. Go lead in technology. Go lead in market share. Go lead in economic share. go lead in in uh diffusion of our technology across the world so that everybody depends on the American American tech stack. And so whether it's energy export or it's uh chip export uh or it's uh infrastructure export or model export or application application technology export. Let's make sure that we're the world leader in every one of them. Let's export like crazy as we know. We would love to have an incredible trade imbalance meaning extraordinary amounts of exporting and so why shouldn't we have AI also uh be in that camp. Now, of course, we also want to make sure that from a technology perspective that we favor the American tech stack, meaning whenever we can, let's favor the American um uh uh companies that are part of that stack. And so, we should always have the best technology first and most to American companies and we do. And so, uh you can find this balance on the one hand uh provide the best and the most to United States. on the other on the other hand uh ensure that American companies win around the world.
>> Yeah. The argument Johnson there is that across the five-layer cake there's one particular layer that it's too important because in the others China can get ahead. They have cheaper energy. They have incredible talent. So they can I mean they build already incredible applications. Adoption is probably much more popular there than here. So the compute becomes the essential nod according to uh you know Washington like or some people in Washington that you have to you can slow them down not completely a zero-sum game but at least they cannot catch up with where we are right now because if they do catch up and if surpasses then they'll become a global leader on probably what we have agreed is the most important technology of our lifetime.
>> Every layer is important. Every layer is important and we should make sure that we can accelerate the United States. Those two conditions I think both we could agree. Every single layer is important. Um, the chip layer obviously is important. Obviously why if the chip layer is not important why are we talking about chips all the time? And so obviously the chip layer is incredibly important. Nvidia had you know call it 90 some odd percent of the world's market share today. In China we have now dropped to zero. um conceding an entire market the size of China probably don't make a lot of strategic sense and so I think that that has already largely backfired maybe it made sense at the time but I think the policy really needs to be dynamic and is needs to stay with the times at this point in time I think it would be fairly safe to say that having American chip companies and and other companies in China makes a lot of sense um I think that with respect to the model layer we have to make sure that we do everything we can to help our model companies have the best, have the most, and have it first. And we do that. And so, you know, this is the largest market in the world. We're the fastest market in the world. And uh uh the tech companies here are agile and they're innovative. I have every confidence that we're providing all of the best here anyways. And so, you know, I have every confidence that that uh AI model companies here in the United States will continue to stay ahead.
>> Yeah. Um, I mean another thing would be that they will pursue self-sufficiency regardless of our policies because they have done it in almost every industry and they're pursuing it also in your industry. Sorry. Well, they let's see they have um self-sufficiency and uh pretty significant leadership in many layers at the energy level. I think we can all acknowledge whether it's the production of energy or the technology that we use or the technology that we use to pro produce our technology our energy they are the world leader okay and so we are the world leader in chips
>> Um, in AI models I would say that that we are ahead uh we're unquestionably ahead they're close behind uh they have just such an extraordinary number of AI researchers you know for whatever reason it's because of interest in science and and um math and uh the encouragement of of um uh you know the the social fabric uh they just have such extraordinary number of science and math experts and and as a result of that the number of AI researchers in China is is quite extraordinary it's one of their national treasures if you will greatest na greatest natural resource and so we have to be mindful that we continue to attract that natural resource to United States we have to make sure that that we welcome here that they want to come here and um um quite frankly kind of concerned about about many of them deciding to stay or not allowing to leave. And so so I think that that um at the model layer our technology is advanced but we do deeply rely also on the talent that we that are around the world. And then and then lastly at the highest level my greatest concern I I remain very concerned today. We are so we are so if you will uh cinematic so incredibly science fiction in the way we describe AI that we're causing so much consternation and so much fear uh around around the United States. Meanwhile, uh the rest of Asia are embracing adopting AI with great enthusiasm. And so this is a this is something we have to be very quite concerned about. This is how we get generally left behind.
>> Yeah. Um, at at the GTC you talked about open source. Um, and you touched about open claw and how popular I I still remember that graph you showed uh the the downloads and everything. Um, open source is really incredibly important for the AI democratization and accessibility and whatnot, but at the same time, it brings safety and security issues. How does Nvidia address these things?
>> Um, so first of all, open source enhances safety and enhances security. Uh the the way that you're going to deal with with a um a super agent cyber security attack is through a massive swarm of models that you that you adopted from open source that you've trained to be uh defense, you know, a massive swarm of defense agents. And so you're not going to defend against a super agent with another super agent. you're going to defend it with massive swarms and so to use asymmetry at your advantage and that they that requires open source technology for us to to develop upon um the crowd strikes the Palo Alto networks the Cisco the Microsofts um all rely on open source technology to do that open source also has the benefit of being completely development in in um in sunlight and so we understand the technology we can uh keep it safe and so safety and security is frankly enhanced with open source with respect the open claw. Um, it was such an incredible phenomenon. The the concern that we had which is the reason why we create we invented two technologies. The first technology is open shell. Open shell puts we call it open shell a shell because it puts a shell around a lobster you know open claw. And so so the the shell keeps it in a safe cage if you will. And that technology was built at NVIDIA as a starting point. we've contributed to uh to uh to uh open source and it's being adopted by so many different companies and the the basic idea is to give it a virtual environment a a sandbox and we're mindful about uh the type of uh information it can access and the policies that uh the policy engines will be able to control it. Uh the information that it can send in and out uh it could u be mindful about the privacy of of personal information that's sent out. It could be uh um uh give you access to certain information that you can't cannot send out. And so the things that all the policies associated with the privacy associated with uh each one of the instances of the claws are now captured in open shell. And so we created an environment that makes it a lot more safe. Uh doing all of this completely in in open source uh enables all of the world's enterprise companies to be able to adopt the technology because they understand what's underneath. and uh working with the US government, we're working with uh cyber security companies around the around the world and and um you know this is how you're going to keep these agents safe.
>> Um, you touched on this really important topic throughout your conversation, AI adoption and the fear out there.
>> Last time when you and I talked in July, we were both upbeat where the trajectory of this uh technology is going. But today there kind of a several elements that I think are boiling out there. You have the political discourse, you have the data center conversations, you have the state-level regulations. Um, you have the layoffs whether or not you can blame it on AI or the other things that have been happening from the tech companies. But I'm afraid that we're entering this phase now where people might build a stigma around AI which I think is not helpful because then you will have as a consequence a lack of AI adoption. How do you assess this? I mean you mentioned this throughout your conversation today that this is a big concern.
>> Um, there's the things that the things that are said are very counterproductive and in fact hurtful. I on the one hand maybe some maybe a scientist thinks that by warning people that AI is going to completely permeate and proliferate across radiology and therefore radiologists are going to get wiped out. Um, on the one hand that might be considered warning and therefore helpful, but in fact the counter would have been hurtful. If we convinced everybody not to be radiologists and we now need radiologists, that actually is hurtful to society. Um, it is hurtful if we convinced all the young college graduates to not be software engineers and it turns out United States need more software engineers than ever. That's hurtful. Um and so we have to be mindful of how we communicate uh the importance of this technology and what it's able to do um to advocate for policy and advocate for guard rails on the one hand. On the other hand, uh scaring people with things like saying nonsensical things which are not going to happen that this is an existential threat to humanity. There's 20% chance that is existential. That's ridiculous. That it's going to wipe out uh 50% of of uh of uh new college grad jobs. Uh that it's going to completely destroy democracy. I mean, these kind of comments are not helpful. They're not based they're they're made by, you know, people who are like me, CEOs, you know, and and um somehow because they became CEOs, you you adopt a god complex and before you know it, you know everything. And so I think we have to be careful and really ground ourselves to talking about the facts. The facts are this. The facts are AI has created uh more than half a million jobs in the last couple years. The facts are AI is our greatest our best opportunity to re-industrialize United States to bring manufacturing jobs back to United States. The facts are that's going to generates hundreds of thousands of jobs, trillions of dollars of new economy back into United States. The fact of the matter is companies that use AI have dis have demonstrated the ability to grow faster. When they grow faster, they hire more people. Apparently, AI creates jobs. And so the question is why is it that on the one hand I'm telling you AI creates jobs. On the other hand, they they destroy jobs. Well, the very simple idea is this. Let's pretend for a second that the total lines of code that we have to write in United States is 1 billion lines of code. And that 1 billion lines of code was supported by 10 million software developers and about a trillion dollars worth of economy. And that one billion lines of code are now going to be automated because most of it is going to be written by AI. Well, if our job is basically typing and 1 billion lines of code are now automated away, you would descri you would come to the conclusion a trillion dollars worth of jobs will be destroyed. But that's fundamentally wrong for two reasons. The first reason is the task of our job and the purpose of our job are related and not the same. If you come if you if you apply that to me, you would come to the conclusion what Jensen does for a living is tap on phones and talk. And tapping on phones and talking, AI has done that just fine. And therefore, my job should be gone, but I'm busier than ever. And so, so there's a fundamental difference between the purpose of the job and the task of the job. And and so that's one. The second thing is it is a fundamental flaw that we only need a billion lines of code written. We need a trillion lines of code written. We need, you know, way more code written than that because we have the imagination of solving problems whether it's in in healthcare or science or, you know, in manufacturing and retail and just luxury living. All of the different fields that we have, we have lots and lots of imaginations of all the things that we can do. If we just didn't have to type anymore, we can go and do those things. And so maybe it's just because in the last 50 years, society has take so this one little device with a keyboard on it consumed all of our lives to the point where we just can't imagine living without typing anymore. And so we just we'll we'll figure it out you guys. I mean, you know, I the idea that that being h being human means to hunch over on this little thing typing all the time. You know, I 50 years before that people didn't do that. and and so in the future we're going to do less of that. We're going to do more of something else.
>> Um, as you know we have the expo next week and I think part of like I as you know SCSP really believes in AI adoption. The reason why
>> It's anybody who loves typing.
>> Yeah. Uh, the
>> Saying there's more to life than typing.
>> Reason why we do
>> Mean something else.
>> The reason why we do the expo is because we want people to see the prospect of this technology. And at the expo, people have the opportunity to build robots and to see, you know, what universities are doing and tech companies with AI. So, unfortunately, we will miss you next week, but I know we worked with your team to have Nvidia presence there. It's a big event for us.
>> We have lots of people coming. We have lots and lots of people coming. It's such an important work. And I love that you're celebrating AI through the lens of healthcare, AI through the lens of energy, AI through the lens of manufacturing, AI through the lens of, you know, retail and all the
>> And quantum. And quantum. Yeah.
>> Different fields of science and different industries and different professions and and all of those lenses. I think the people that have engaged it are so in incredibly excited. Um, all the engineers at NVIDIA today use AI. Uh AI basically does most of our coding and yet you know the engine we're hiring more engineers than ever. We have more challenges than ever. We have bigger dreams than ever. Where our ambitions are greater than ever. And so if you just said this to yourself, suppose we infused AI into this country and as a result of that we are doing things faster than ever before. Our ambition is greater than ever before. Our expectations are greater than ever before. How is that a bad condition for our country? That's exactly what we want to be more ambitious than ever. To be faster than ever, to be better than ever. And so I hope that we adopt AI irrespective of anything else whatever we do help people understand how AI has been transformative into so many people and so many so many industries already in just the last six months
>> In just the last six months since it's been really been useful you know prior to that it's been incredible but not useful now it's useful and incredible and so I really do hope that every industry every company takes advantage of it We we're surely, you know, all in on it and um we're we're supercharged by it. Uh exhilarated by it and um our ambitions are greater greater than ever because of it.
>> Jess, thank you so much for coming by. Uh all the best and uh I'll see you again.
>> Thank you, Elie.
>> Thank you. Thank you so much. Thank you. [applause]