Dwarkesh PodcastSatya Nadella — Microsoft’s AGI plan & quantum breakthrough
EVERY SPOKEN WORD
150 min read · 29,960 words- 0:00 – 5:48
Intro
- SNSatya Nadella
Us self-claiming some AGI milestone, that's just nonsensical benchmark hacking. The real benchmark is-
- DPDwarkesh Patel
(laughs)
- SNSatya Nadella
... is the world growing at 10%? Being very good at understanding what are winner-take-all markets and what are not winner-take-all markets is, in some sense, everything. If this thing is really as powerful as people make it out to be, the state is not gonna sit around and wait for private companies. We like the analogy of thinking of this as the transistor moment of quantum computing. Maybe you will use quantum to generate synthetic data that then gets used by AI to train better models. If intelligence is log of compute, whoever can do lots of compute-
- DPDwarkesh Patel
Right.
- SNSatya Nadella
... is a big winner.
- DPDwarkesh Patel
Right. Satya, thank you so much for coming on the podcast. Um, so just in a second, we're gonna get to the two big breakthroughs that Microsoft has just made. And congratulations, same day in Nature, uh, the Majorana 0 chip, which we have in front of us right here, and also the world human action models. But can we just continue the conversation we were having a second ago? So you were describing the ways in which the things you were seeing in the '80s and '90s, you're seeing them happen again.
- SNSatya Nadella
Yeah, I mean, the thing that, uh, is exciting for me, Dwarkesh, first of all, it's fantastic, uh, to be, uh, on your podcast, I'm a big listener and, um-
- DPDwarkesh Patel
(laughs)
- SNSatya Nadella
... uh, and it's just fun to be, um, you know, I love the, the way that you do these interviews and the broad topics, uh, that you explore. Um, it sort of, to me, reminds me a little bit of my, I would say, first few years even, uh, in the tech industry, starting in the '90s, um, where there was, like, real debate about whether it's gonna be RISC or CISC or, "Hey, are we really gonna be able to build servers using even x86 or..." You know, we, when I joined Microsoft, that was the, you know, the beginning of what was Windows NT.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
Uh, so everything from the core silicon platform to the operating system to the app tier, uh, that full stack approach, I mean, like it's being, the entire thing is being litigated, and that's, I think, perhaps, you know, you could say cloud did a bunch of that and obviously distributed computing, um, and cloud did change client server-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... the web changed, uh, massively. But this does feel a little more like maybe more full stack, um, than even the past that at least I've been involved in.
- DPDwarkesh Patel
When you think about what actually, um, which decisions ended up being the long-term winners in the '80s and '90s and which ones didn't, and especially when you think about, uh, spe- you know, you were at Sun Microsystems, they had an interesting experience with the '90s dot-com bubble. Um, people talk about this data center build-out as being a bubble, but at the same time, we have the internet today as a result of what was built out then. What are the lessons about what will stand the test of time? What is an inherent secular trend? What is just ephemeral? What stands out?
- SNSatya Nadella
Yeah, it's actually, it, it's interesting. I mean, I think the, um, if I sort of go back even at least the four big transformations that I've been part of, right? If you say the client, uh, and the client server, uh, so that's the birth of the graphical user interface and the x86 architecture basically even allowing us to build servers. Um, it was very clear to me, I remember going to P- what was PDC in '91, in fact, I was at Sun at that time, um, and in '91, I went to Moscone, went to basically that's when, uh, Microsoft first described their Win32 interface, and I said it was pretty clear to me, uh, what was gonna happen where the server, uh, was also going to be an x86 thing, right? So that, when you have the scale advantages accruing to something, uh, that's the secular bet you have to place, right? So that, um, and so what happened in the client was gonna happen on the server side-
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
... and then you were able to then actually build client server applications, so the app model and it became clear. Then the web was the big thing for us, which we had to deal with in starting, in fact, as soon as I joined Microsoft, I think, what is it, like, the, uh, you know, the browser, the Netscape browser or the Mosaic browser came out, what, December, November of '93, right, I think is when, uh, you know, Andreessen-
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
... and crew sort of had that, and so that was a big game changer, right? But in an interesting way, just as we were getting going on what was the client server wave, and it was clear that we were going to win it as well, uh, we had the browser moment, and so we had to adjust. And we did a pretty good do- job of adjusting to it, right? Because the browser was a, a new, um, I'd say app model, um, and we were able to embrace it with everything we did, right? Whether it was HTML in Word or build a new thing called a browser ourselves-
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
... and compete for it and then build a web server, uh, on our server stack and sort of, uh, go after it. Except, of course, we missed, uh, what turned out to be the biggest business model on the web, uh, because we all assumed the web is all about being going to be distributed. Who would have thought that search would be the biggest winner in organizing the web?
- DPDwarkesh Patel
Right.
- SNSatya Nadella
Um, and so that's where, uh, we obviously didn't see it, um, and Google saw it and executed super well. Uh, so that's kind of one lesson learned for me is like, hey, you got to really not only get the tech trend right, you also have to get, uh, where is, um, the value going to be created-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... um, with that trend? And, uh, these business model shifts are probably tougher, uh, than even, uh, the tech trend changes.
- 5:48 – 16:02
AI won't be winner-take-all
- SNSatya Nadella
- DPDwarkesh Patel
Wh- where is the value going to be created in AI?
- SNSatya Nadella
That's a great one. So I think the, the, at least in my current, uh, thing is there are two places where I can say with some confidence. One is the hyper-scalers will do well, right? Because the fundamental thing is if you sort of go back to even how Sam and others describe it, I mean, like if-... you know, intelligence is log of compute.
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
Whoever can do lots of compute-
- DPDwarkesh Patel
Right.
- SNSatya Nadella
... is a big winner. (laughs)
- DPDwarkesh Patel
Right.
- SNSatya Nadella
Um, and the other interesting thing is, if you look at underneath even any AI workload, like take ChatGPT, it's not like everybody's excited about what's happening on the GPU side. It's great, but it's like the ratio, like, uh, in fact, I think of my fleet even as a ratio of the AI accelerator to storage to compute. And at scale, you gotta grow it.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
Um, and so that infrastructure need for the world, um, is just going to be exponentially growing.
- DPDwarkesh Patel
Right.
- SNSatya Nadella
Right? So, in fact, it's group mana from heaven to have, uh, um, these AI workloads because guess what? They're more hungry for more compute.
- DPDwarkesh Patel
Right.
- SNSatya Nadella
Right? Not just for training, but we now know for test time. And as I said, test time. Like, here's an interesting thing. When you think of an AI agent, it turns out the AI agents is going to exponentially increase compute usage.
- DPDwarkesh Patel
Yep.
- SNSatya Nadella
Because you now are not even bound by just one human invoking a program. It's but, uh, one human invoking programs that invoke lots more programs.
- DPDwarkesh Patel
Right.
- SNSatya Nadella
And so that's going to create massive, massive demand-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... at scale for compute infrastructure. So our hyperscale business, Azure business, I think that's like, and other hyperscalers, I think that's a big thing.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
Then after that, it becomes a little fuzzy. Uh, because there, you could sort of say, "Hey, there is a w- winner-take-all model." I just don't see it because this, by the way, is the other thing I've learnt, is being very good at understanding (laughs) what are winner-take-all markets and what are not winner-take-all markets is, in some sense, everything. Uh, like I remember in the, even in the early days when I was getting into Azure, I mean, Amazon had a very significant lead and people would come to me and investors (laughs) would come to me and say, "Oh, it's, uh, game over. You'll never make it. Amazon's, it's winner-take-all." And having competed against Oracle and IBM in client server, I knew that, look, the buyers will not tolerate winner-take-all, right? I- structurally, hyperscale will never be a winner-take-all, um, because buyers are smart. Um, consumer market sometimes can be winner-take-all. But anything there, the buyer is a corporation, an enterprise, an IT department, a B-, they will want multiple suppliers. And so you gotta be one of the multiple suppliers, (laughs) not, um... And so that, I think, is what'll happen even in the model side. So there will be open source, there will be a governor, just like on Windows. One of the big lessons learned for me was, um, you, if you have an, uh, closed-source operating system, there will be a complement, uh, to it, which will be open source. Um, and so to some degree, that's a real check on what happens. And so I think in models, there is one dimension of, uh, maybe there'll be a few closed-source. There will be, there will definitely be an open-source alternative. And the open-source alternative will actually make sure the closed-source, uh, winner-take-all is mitigated.
- DPDwarkesh Patel
Mm.
- SNSatya Nadella
So that's kind of at least my feeling on the model side. And oh, oh, and by the way, let's not discount. If this thing is really as powerful as people make it out to be, uh, the state is not gonna sit around and wait for private (laughs) companies to go around and, uh, and, and all over the world. So it's sort of I don't see it as a winner-take-all. Then above that, I think it's going to be the same old stuff, which is in consumer, in some categories, there may be some winner-take-all network effect, right? After all, ChatGPT-
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
... is a great example. Like, I mean, um, it's kind of like, it's an at-scale consumer property that has already, uh, got real escape velocity, right? I go to the App Store and I see, you know, it's always like there in the top five, uh, and I say, "Wow, like, that's pretty un- unbelievable." (laughs) So they were able to use that early advantage and parlay that, uh, into an app advantage. And so in consumer, that could happen. In the enterprise, again, uh, I think there will be, by category, different winners. So that's sort of at least how I analyze it.
- DPDwarkesh Patel
I, I have so many follow-up questions. Um, h- uh, we gotta g- get to quantum in just a second. But, so on the idea that maybe the models get commoditized, um, look, I, I, maybe somebody could make, have made a similar argument a couple of decades ago about the cloud, that fundamentally it's just like a chip in a box. Uh, but in the end, of, of course, you and many others figured out you guys have amazing profit margins in the cloud, and you figured out ways to get economies of scale and add, uh, their value-add. Um, and fundamentally, e- e- even forgetting the jargon, like, if you've got AGI and it's like helping you make better AIs, m- right now, it's synthetic data in RL, maybe in the future it's an automated AI researcher, that seems like a good way to entrench your advantage there. I, I'm curious what you make of that, just the idea that like it really matters what I had there.
- SNSatya Nadella
Um, by the way, uh, uh, at scale, nothing is commodity, right? So to your point about cloud, I mean, everybody would say, "Oh, cloud's a commodity," except, you know, when you scale. That's why the knowhow of running a hyperscaler, right?
- DPDwarkesh Patel
Right.
- 16:02 – 22:23
World economy growing by 10%
- SNSatya Nadella
- DPDwarkesh Patel
Um, you recently reported that your yearly revenue from AI is 13 billion dollars. But if you look at your year on year growth on that, in like four years it'll be, you know, 10X that, you'll have 130 billion dollars in revenue from AI if the trend continues. If it does, what do you anticipate we're doing with all that intelligence? Like, this industrial scale use?
- SNSatya Nadella
It's a great one.
- DPDwarkesh Patel
Is it going to be like through Office? Is it going to be you deploying it for others to host? Is it going to be, uh, you gotta have the AGI's to have 130 billion in revenue? W- what does it look like?
- SNSatya Nadella
Yeah, I see, the way I come at it, Dwarkesh, is I say- it's a great question because at some level if you're going who have this sort of explosion, abundance, whatever, commodity of intelligence available, you know, the first thing we have to observe is GDP growth, right?
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
Before I get to what Microsoft's sort of revenue will look like, I mean, there's only one governor in all of this, right? Which is, this is where a little bit of we get ahead of ourselves with all this AGI hype, which is, hey, you know what? Let's first see if, let's say developed, I mean, like remember, like the developed world is what? 2% growth and if you adjust for inflation it's zero.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
Uh, that, like, so in 2025 as we sit here, call it, I'm, I'm not an economist, at least I look at it and say, "Man, we have a real growth challenge."
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
So the first thing that we all have to do is let con- and, and when we say, "Oh, this is like the Industrial Revolution, blah, blah, blah," oh, let's have that Industrial Revolution type of growth. That means to me 10%, 7%, developed world inflation adjusted growing at 5%. That's the real marker, right? So we, it's not just, it can't just be supply side, right? It has to be out, in fact that's the thing, right? I think there's a, a lot of people are writing about it, I'm glad they are, which is the big winners here are not gonna be tech companies. Uh, the winners are going to be the broader industry that uses this commodity that, by the way, is abundant.
- DPDwarkesh Patel
Right.
- SNSatya Nadella
Um, and suddenly productivity goes up and the economy's is going, you know, growing at a faster rate. When that happens, we'll be fine as an industry. But that's, to me, the moment, right? So it, us self-claiming some AGI milestone, that's just nonsensical benchmark hacking to me.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... the real benchmark is-
- DPDwarkesh Patel
(laughs)
- SNSatya Nadella
... is the world growing-
- DPDwarkesh Patel
Right.
- SNSatya Nadella
... at 10%?
- DPDwarkesh Patel
Okay. So if the world grew at 10% ... The world economy is 100 trillion or something. If the world grew at 10%, that's like extra 10 trillion, uh-
- SNSatya Nadella
That's it.
- DPDwarkesh Patel
... in value produced every single year. If that is the case, um, you as a hyper scaler, it seems like 80 billion is a lot of money. Um, shouldn't you be doing like 800 billion (laughs) if you, if you really think in a couple of years we could be really growing the world economy at this rate? And the key bottleneck would be, do you have the compute necessary to deploy these AIs to do all this work?
- SNSatya Nadella
I mean, that, that is correct. And so therefore ... But by the way, the balance is like, I think a little bit of it is right now is like, hey, let me ... Like the classic supply side is, oh, let me build it and they'll come, right? I mean, that's an argument and-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... you, you know, after all we've done that, we've taken enough risk, uh, to go do it. But at some point, the supply and demand have to map. Uh, and so that's what I think we ... And that's why I'm, I'm tracking both sides of it, right? So that's why I think, you know, you can go off rails completely when you're like all hyping yourself with all the supply side, versus really understanding how to translate that into real value to customers. Um, and so unless ... And that's why I, I look at my inference revenue. That's one of the reasons why even the disclosure on the inference revenue. It's interesting that not many people are talking about their rail revenue. Um, but to me, that I think is important, uh, as a governor for how you think about it, right? And you're not going to say, "Oh, they have to symmetrically meet at any given point in time." But you need to have existence proof that you are able to parlay yesterday's, let's call it capital, into today's, uh, demand so that then you can again invest maybe exponentially even, uh, knowing that you're not gonna be completely rate mismatched.
- DPDwarkesh Patel
Yeah. I wonder if there's a contradiction in these two different viewpoints because look, I mean, one of the things you've done wonderfully is you make these early bets when there's ... You, you invested in OpenAI in 2019 even before there was Copilot and any applications. If you look at the Industrial Revolution, these, um, you know, six, 10%, uh, build outs of railways and whatever things, many of those were not like, we've got revenue from the tickets and now we're gonna, whatever.
- SNSatya Nadella
It was a lot of money lost.
- DPDwarkesh Patel
Um, that's true. (laughs) Um, the, uh ... So if you, if you really think like there's some potential here to 10X the, or 5X the growth rate of the world, and then you're like, "Well, what, what is the revenue from GPT-4?" Um, I mean, like if you really think that that's the possibility from the next level up, shouldn't you just like, "Let, let's go crazy, let's do the hundreds of billions of dollars of compute." I mean, there's like some chance you get that.
- SNSatya Nadella
It makes sense, right? I mean, like the, the, the thing is like, I mean like here's the interesting thing, right? The real question quite frankly to answer is, um, is this just about ... Like that's why even that balanced, uh, approach to the fleet at least is very important to me, right? Which is, it's not about building compute, it's about building compute that can actually help me not only b- train the next big model, but also serve the next big model. Uh, and you understand until you do those two things, you're not gonna be able to really be in a position to take advantage of even your investment, right?
- DPDwarkesh Patel
Right.
- 22:23 – 31:03
Decreasing price of intelligence
- SNSatya Nadella
- DPDwarkesh Patel
Yeah. Um, I mean, speaking of prices coming down, you recently tweeted after the DeepSeek model came out, uh-
- SNSatya Nadella
That's right.
- DPDwarkesh Patel
... about Jevons Paradox. Um, and I'm curious if you can flesh out ... So Jevons Paradox occurs when there's like the demand for something is highly elastic. Um, is intelligence that bottlenecked on prices going down? Because when I think about at least my use cases as a consumer, it's like intelligence is already so cheap. It's like two cents per million tokens. Like do I really need it to go down to .02 cents? I'm just like really bottlenecked on it becoming smarter. And if you need to do, charge me 100X, do 100X bigger training run, um, I'm happy for companies to take that. But maybe you're seeing something different on the enterprise side or something. What, what is the key use case of intelligence that really requires you get a .002 cents per million tokens?
- SNSatya Nadella
I mean, I think the, the real thing is the utility of the tokens, right? So which is in some sense, uh, both are, uh, need to happen. One is intelligence needs to get better and cheaper. And any time there's a breakthrough like even what DeepSeek did or what have you with the efficient frontier of let's say performance per, you know, token changes, uh, and the curve gets bent, um, and the frontier moves-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... that just brings more demand. And so that's sort of how I look at it. And that's what, what happened with cloud, right, by the way. Here's an interesting thing. We used to think, oh my God, we've sold all the servers in the client server era, except once we sort of started putting, uh, servers in the cloud, uh, suddenly people started consuming more, uh, because y- they could buy it cheaper and by, uh, it was elastic, uh, and they could buy it as a meter versus a license. Uh, and it completely expanded. Like I mean, I remember like, you know, going let's say to a country like India and sort of talking about, oh, here is SQL Server. Uh, we sold a little, but man, the cloud in India is so much bigger (laughs) , uh, than anything that we were able to do in the server era. And that I think is gonna be true. Uh, like you take ... I mean, if you think about like if you want to really have in a, in the global south, in a developing country-
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
... uh, if you had these tokens that were in, like available for healthcare, uh, that were really cheap-That'll be like the biggest change ever. (screen whoosh)
- DPDwarkesh Patel
A quick word from our sponsor, Scale AI. Publicly available data is running out, so major labs like Meta and Google DeepMind and OpenAI all partner with Scale to push the boundaries of what's possible. Through Scale's Data Foundry, major labs get access to high-quality data to fuel post-training, including advanced reasoning capabilities. As AI races forward, we must also strengthen human sovereignty. Scale's research team, SEAL, provides practical AI safety frameworks, evaluates frontier AI system safety via public leaderboards, and creates foundations for integrating advanced AI into society. Most recently, in collaboration with the Center for AI Safety, Scale published Humanity's Last Exam, a groundbreaking new AI benchmark for evaluating AI systems' expert-level knowledge and reasoning across a wide range of fields. If you're an AI researcher or engineer and you wanna learn more about how Scale's data foundry and research team can help you go beyond the current frontier of capabilities, go to scale.com/dwarkesh. (screen whoosh) I think it's like quite reasonable for somebody to hear people like me in San Francisco and think like, "Look, look, they're kinda silly. They don't know how - what it's actually like to deploy things in the real world." As somebody who works with these Fortune 500s and is working with them to deploy things for hundreds of millions, billions of people, what's your sense on how fast deployment of these capabilities will be? Even when you have working agents, even when you have things that can do remote work for you and so forth, with all the compliance and with all the inherent bottlenecks, is that gonna be a, a big bottleneck or is that - is, is that gonna move past pretty fast?
- SNSatya Nadella
It is going to be a, a real (laughs) challenge because the, the, the real ch- issue is, um, uh, change management or process change, right? I mean, this is, uh... Here, here's an interesting thing, right? Which is, one of the analogies they use is, uh, just imagine how a multinational corporation like us did forecasts, uh, pre-PC and email and spreadsheets, right? I mean, faxes went around. Uh, somebody then got those faxes and then did an inter-office memo that then went around and people entered numbers and, you know, and, and, and then ultimately a forecast came maybe (laughs) just in time for the next quarter. Um, then somebody said, "Hey, I'm just gonna take an Excel spreadsheet, put it in email, send it around, people will go edit it, and I'll have a forecast." So the entire forecasting business process changed because the work, the work artifact, and the workflow changed. That is what needs to happen-
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
... uh, with AI being introduced, uh, into knowledge work. In fact, the, the - when we think about even all these agents, the fundamental thing is there's a new work and workflow. Like, for example, for me, like even prep- prepping for our sort of podcast, you know, I go to my Copilot and I say, "Hey, I'm gonna talk to Dwarkesh about, you know, our quantum announcement and this new, uh, you know, model, uh, that we built for game generation and just kinda give me like a summary of all the stuff that I should read up before going."
- DPDwarkesh Patel
Right.
- SNSatya Nadella
And it knew like the ti- the Nature papers, it took that. In fact, I even said, "Hey, go give it to me in a podcast format." And so it sort of even did a nice job of two of us chatting, uh, about it. Like so that became... And in fact, then I shared it with my team, right? So I took it and put it into Pages, which is our artifact, uh, and then shared. So I... So the new workflow for me is I think with AI and work with my colleagues, right? So that's a fundamental change management of everyone who's doing knowledge work suddenly figuring out these new patterns of how am I going to get my knowledge work done-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... in new ways. That is gonna take time. It's gonna be something like in sales and in finance, in supply chain. So for an incumbent, I think that this is going to be one of those things where, you know, let's take one of the analogies I like to use is what manufacturers did with Lean. I love that because in some sense if you look at it, Lean became a methodology of how one could take an end-to-end process in manufacturing and become more efficient, right? It's that continuous improvement which is reduce waste and increase value. That's what's gonna come to knowledge... This is like Lean for knowledge work in particular, and that's gonna be the hard work of management teams and individuals who are doing knowledge work, and that's gonna take its time.
- DPDwarkesh Patel
Can, can I ask you just, uh, briefly about that analogy? Um, one, one of the things Lean did is physically transformed what a factory floor looks like. It revealed bottlenecks that people didn't realize-
- SNSatya Nadella
Exactly.
- DPDwarkesh Patel
... until you're really paying attention to, uh, the processes and workflows. Um, you mentioned briefly what your own workflow, how your own workflow has changed as a result of, uh, AIs. Um, I'm, I'm curious if you can add more color to w- what would it be like to run a big company when you have these AI agents that are getting smarter and smarter over time?
- SNSatya Nadella
Yeah. In fact, it's interesting, it's interesting you ask that. Like I was - been thinking about like, for example, (laughs) you know, today if I look at it, uh, I, you know, we are very email heavy. Um, so I get in the morning and I'm like, man, like my inbox is full and I'm responding. And so I can't wait for some of these Copilot agents, uh, to kind of automatically populate my drafts so that (laughs) I-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... can start reviewing and sending. And so that's kind of one. But l- literally, I do feel like I already have in Copilot like at least 10 agents, right? I have which I do at - I, I, because I query them as sort of different things for different tasks. And I feel like there's a new inbox that's going to get created, which is my millions of agents that I'm working with will have to invoke some exceptions to me, notifications to me, ask for instructions. So at least what I'm thinking is that there's a new scaffolding, uh, which is the agent manager is going to be that one. Like it's not just a chat interface. I kinda need a smarter thing than chat interface, uh, to manage all the agents and their dialogue. So that's why I think of this Copilot as the UI for AI is a big, big deal. Um, and each of us is going to have it as, uh, you know... So basically think of it as there is knowledge work and there's a knowledge worker, right?... you, the knowledge work may be done by many, many agents, but you still have knowledge work, worker who is dealing with all (laughs) the knowledge workers. And that, I think, is the c- the, the interface that one has to build.
- DPDwarkesh Patel
Yeah. I mean, I, I'm sort of curious about,
- 31:03 – 43:35
Microsoft's Quantum breakthrough
- DPDwarkesh Patel
like, um, you're one of the few people in the world who can say that you have access to 200,000 ... You have this, like, swarm of intelligence around you in the form of Microsoft the company and all its employees, um, and you have to manage that, and you have to, like, you know, uh, h- how to interface with that, how to make m- best use of that. Um, hopefully more of the world will get to have that experience in the future. Um, I'd be curious about how your inbox w- if that means everybody else's inbox will look like yours in the morning. Okay, before we get to that, I wanna keep asking you more about AI, but we real- I, I really wanna ask you about, uh, the b- the big breakthrough in quantum that Microsoft Research has announced. So can you explain a- uh, can you explain what's going on here?
- SNSatya Nadella
Yeah. This has been ... It's, uh, an another, whatever, 30-year journey, uh, for us. It's unbelievable. Like, I'm the third CEO-
- DPDwarkesh Patel
(laughs)
- SNSatya Nadella
... of Microsoft who's been excited about, uh, quantum. Um, I, I think the, the fundamental breakthrough here, or the vision that we've always had is, um, you need a physics breakthrough, uh, in order to build a utility scale quantum computer that works. Um, and so we took that path, you know, uh, which was a, the path of sort of saying, "Look, the one way for having that, um, less noisy or the more reliable qubit is to bet on a physical property that by definition, um, is more reliable." And that's kinda what led us to those Majorana, uh, zero modes as the thing to go, um, which was theorized in the 1930s.
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
And so, so the question was can we actually physically fabricate these things?
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
Can we actually build them? So the big breakthrough, effectively, and I know you talked to Chethan, was, um, that we now finally have existence proof, uh, and a physics breakthrough of our Majorana zero modes, um, in a new phase of matter, effectively, right? So we did ... This, this is why, I think, we like the analogy of thinking of this as the transistor moment of quantum computing where we effectively have a new, uh, phase, which is the topological phase where, uh, which is more reli- wh- which we, wh- which means we can even now, uh, reliably hide the quantum information-
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
... and measure it, uh, and then ... And we can fabricate it. And so now that we have it, we feel like with that core foundational, um, fabrication technique, uh, out of the way, we can start building a Majorana chip. Uh, that Majorana 1, which I think is going to basically be the first chip that will be capable of a million qubits, physical. And then on that, thousands, mil- thousands of, uh, logical qubits-
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
... are, are corrected. And then i- game on, right? So then you can suddenly have now got the ability to build a real utility scale quantum computer. And that, to me, is now so much more feasible, right? We've been work- Because without something like this, you will still be able to achieve milestones, but you'll never be able to build a utility scale computer. And so that's why we're excited about it.
- DPDwarkesh Patel
Amazing. And by the way, I believe this is it right here.
- SNSatya Nadella
That is it, yeah.
- DPDwarkesh Patel
Yes.
- SNSatya Nadella
Yeah, I forget now. Are we calling it Majorana ... Yeah, that's right.
- DPDwarkesh Patel
(laughs)
- SNSatya Nadella
Majorana 1. And, uh, I'm glad we named it after that. And this is the, uh ... I mean, to think of the, the fact that we are able to build, like, something like a million, uh, qubit quantum computer in a thing of this size is just unbelievable. Like, I mean, that ... And, and that's, I think, the crux of it, right? Which is unless and until we could do that, you can't dream of building a utility scale quantum computer.
- DPDwarkesh Patel
Hmm. And you're saying the eventual million qubits will go on a chip this size?
- SNSatya Nadella
That's it. That's right.
- DPDwarkesh Patel
Okay, amazing.
- SNSatya Nadella
Yeah.
- DPDwarkesh Patel
So other companies have announced 100, uh, physical qubits, Googles, IBMs, others. Um, w- when you say ... And you've announced one, but you're saying that yours is way more scalable in the limit.
- SNSatya Nadella
Yeah. So we're ... By the way, we're t- The one thing that we have also done is we've taken sort of an approach where we sorta separated out our software and our hardware, right? So we're building out our software stack. So which is ... In fact, we now have, uh, uh, with couple of different, with the, uh, neutral atom folks, uh, with the ion trap folks. Uh, we're also working with others who even have, I think, pretty good approaches even with photonics and what have you. So that means there'll be different types of quantum computers. And in fact, we have what? 20 f- I think the last thing that we announced was 24 logical qubits. So we have also got some fantastic breakthroughs-
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
... on error correction, uh, and that's what is allowing us even on a neutral atom, uh, and an ion trap, uh, quantum computers to build, uh, these 20 plus. And that, I think that'll keep going even in throughout the year. You'll see us improve that, uh-
- DPDwarkesh Patel
Hmm.
- SNSatya Nadella
... yardstick. But we also then said, "Let's go to the first principles, uh, and build our own super c- uh, quantum computer," uh, that is betting on the topological, uh, qubit, uh, and that's what this, uh, this breakthrough is about.
- DPDwarkesh Patel
Amazing. The, um, the, the million, uh, topological qubits, thousands of logical qubits, what is the estimated timeline to scale up to that level? What-
- SNSatya Nadella
Yes.
- 43:35 – 50:35
Microsoft's gaming world model
- DPDwarkesh Patel
let's dig into the- the other big breakthrough you've just made, and it's amazing that you have both of them coming out the same day in your gaming world models. I- I'd love if you can tell me a little about that.
- SNSatya Nadella
Yeah, so we're, I think we're gonna call it Muse, is what I learned-
- DPDwarkesh Patel
(laughs)
- SNSatya Nadella
... is they're going to be the model of this world, um, um, Action, uh, or Human Action Model. And, um, it's just very cool. See, one of the things that, uh, you know, obviously DALL-E and Sora have been unbelievable in what they've been able to do in terms of generative models. And so one thing that we, uh, want- wanted to go after was using game play data. Uh, can you actually, uh, generate games that are both consistent and then have the ability to generate the diversity of what that game represents and then are persistent to user mods, right? So, uh, so that's what this is.
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
Um, and so they were, uh, able to work with one of our game studios, um, and, uh, this is the other publication-
- DPDwarkesh Patel
Right.
- SNSatya Nadella
... in Nature. And the cool thing is, what I'm excited about is bringing, you know, so we're going to have a catalog of games, uh, soon that we will start sort of using these models, uh, or we're gonna train these models to generate, um, and then start playing them. Um, and in fact, you know, when Phil Spencer first showed it to me, uh, where, you know, he had an Xbox controller and this model basically took the input-
- DPDwarkesh Patel
Right.
- SNSatya Nadella
... and generated the output (laughs) based on the input. Uh, and it was consistent with the game. Uh, and that, to me, is a massive, massive, uh, you know, moment of wow. It's kind of like, you know, the first time we saw ChatGPT complete sentences or DALL-E draw or Sora. This is kind of one such moment.
- DPDwarkesh Patel
Yeah, and it'll, um, I only j- I got a chan- I s- got a chance to see some of the videos in the real time demo this morning with your r- r- leader of research, Katja, on this.
- SNSatya Nadella
Yeah.
- DPDwarkesh Patel
And only once I talked to her did it really hit me how incredible this is in the sense that we've used AI in the past to model agents. And just using that same technique to model the world around the agent and get this-
- SNSatya Nadella
That's right.
- DPDwarkesh Patel
... consistent real time. We'll superimpose videos of what this looks like atop this podcast so people can get a chance to see it for themselves. I guess it'll be out by then, so they can also watch it there. This in itself is incredible. You, through your span as CEO, have invested tens, hundreds of billions of dollars in building up Microsoft Gaming and acquiring IP. And, um, uh, in retrospect, if you can just merge every, all of this, uh, data into one big model that can give you this experience of visiting and going through multiple worlds at the same time, and if this is the direction gaming is headed, seems like a pretty good (laughs) investment to be have made. Uh, did you have any pr- uh, premonition about this or a good coincidence?
- SNSatya Nadella
N- no, uh, no. I mean, I, yeah, I- I wouldn't say that, um, (laughs) we invested in gaming to build models.
- DPDwarkesh Patel
(laughs)
- SNSatya Nadella
I mean, that, uh, we- we invest, quite frankly, I want to... You- here's an interesting thing about our history. We built, uh, our first game before we built Windows, right?
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
Flight Simulator was a Microsoft product long before, uh, we even built Windows. So- so gaming has got a long, uh, history at the company and we want to be in gaming for gaming's sake. And that's, I always start by, uh, I hate to be in businesses where they're means to some other end. They have to be ends onto themselves.
- DPDwarkesh Patel
Right.
- SNSatya Nadella
And then, yes, we are not a conglomerate. We are a company where we have to bring all these assets together and be better owners off by adding value, right? So for example, cloud gaming is a natural thing for us to invest in-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... because, uh, that'll just expand the TAM and expand the ability for people to play games everywhere. Same thing with AI and gaming. We definitely think that it can be helpful in maybe changing... It's kind of like the CGI moment even for gaming long term, and it's great as a, the biggest, world's largest publisher, this would be helpful. But at the same time, we gotta produce great quality games.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
I mean, you can't be a- a gaming publisher, uh, without sort of first and foremost being focused on that. Uh, but the fact that this data asset is going to be interesting, not just in gaming context, but it's going to be a general action model and a world model, uh, it's fantastic. I mean, like, um, you know, I- I think about gaming data as perhaps, you know, what YouTube is perhaps to Google, gaming data is to Microsoft. And so therefore, uh, I'm excited about that.
- DPDwarkesh Patel
Yeah. And th- sorry, that's what I meant. In just the sense of like you can have one unified experience across many different kinds of games. Um, how does this fit into the other, uh, separate from AI, the other things that Microsoft has worked on in the past, like mixed reality, um, maybe giving smaller game studios a chance to build these triple-A action games? And just like five, ten years from now, how, w- what kinds of ways could you
- NANarrator
Yeah, I think, you know-
- DPDwarkesh Patel
... interact with this?
- SNSatya Nadella
You know, I've- I've, I mean, I've thought about these three things as sort of the cornerstones, right, of, um, in an interesting way, even, like, around five, six, seven years ago is when I said, like, the three big bets that we want to place is AI, quantum, and mixed reality.
- 50:35 – 56:30
Legal barriers to AI
- DPDwarkesh Patel
Um, going back to AI for a second. So in your 2017 book ... Um, 2019, you invest in OpenAI very early. 2017 is even earlier. And you say in your book, "One might also say that we're birthing a new species, one whose intelligence may have no upper limits." Now, super early, of course, to be talking about this in 2017. We so far have been talking in sort of, like, a granular fashion about agents and, uh, Office, Copilot, and, um, uh, CapEx and so forth. But if you just zoom out and consider this statement y- you've made and you think about, like, you as somebody, uh, as a hyperscaler, as the person doing research in these models as well, um, providing training, inference research for building a new species. Like, in the grand scheme of things, how do you think about this? Uh, are, uh, do you think we're headed towards super humi- human intelligence in your time as CEO?
- SNSatya Nadella
(laughs) I think even Mustapha uses that term. Uh, in fact, he's used that term more recently around what this new species ... Are ... The way I come at it is, you definitely need trust. Like, the, I think the one thing that, um ... Before we kind of claim it is something, um, as big as a species, the fundamental thing that I think that we've got to get right is that there is real trust, whether it's personal or societal level. Uh, trust that's baked in. That's the hard problem. Because I think the one, biggest rate limiter, uh, to the power here will be how does our legal, call it infrastructure. We're talking about all the compute infrastructure. How does the legal infrastructure evolve to deal with this? Like, e- entire world is constructed with things like humans owning property, having rights, uh, and being, um, liable. Like, that's the fundamental thing that one has to sort of first say, "Okay, what does that mean for anything that now humans are using as tools?" And if humans are going to delegate more authority to these things, then how does that structure evolve? Like, until that really gets resolved, I think just talking about sort of the tech capability I don't think is gonna happen.
- DPDwarkesh Patel
Uh, as in, like, we won't be able to deploy these kinds of intelligences-
- SNSatya Nadella
Absolutely.
- DPDwarkesh Patel
... until we figure out how to-
- SNSatya Nadella
Because at the end of the day, there is no way. Like today, you cannot deploy these intelligences unless and until there's someone indemnifying it as a human. Uh, that's, I think the ... To your point, that's one of the reasons why I think about like even the most powerful AI is essentially working with some delegated authority-
- DPDwarkesh Patel
Right.
- SNSatya Nadella
... uh, from some human.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
Uh, you can sort of say, "Oh, that's all a- a- alignment," this, that, and the other. And that's why I think you have to sort of really get, uh, these alignments to actually work and be verifiable in some way. But I just don't think that you can deploy intelligences that are out ... So, for example, this AI takeoff problem may be a real problem, but before it is a real problem, the real problem will be in the courts. Uh, because the courts ... I mean, like, no society is going to allow for some human to say, "AI did that."
- DPDwarkesh Patel
Yes. Um, well, there's a lot of societies in the world, and I wonder if any one of them might not have a legal system that might be more amenable. And if there, you can't have a takeoff, then you might, uh, worry ... Like, it doesn't have to happen in America, right? Even if the, you, uh, legally-
- SNSatya Nadella
Yeah. It's a good point. But e- but even, like, it, it's sort of like even if in any ... One, one thing that we are, I think, we think that no society cares about it, right? There can be rogue actors.
- DPDwarkesh Patel
Right.
- SNSatya Nadella
I'm not saying there won't be rogue actors. I mean, there are c- cyber criminals and rogue states. They're going to there, be there. But to think that sort of the human society at large doesn't care about it is also not going to be true, right? So I think we all will care, uh, right? We deal, we know how to deal with rogue states and rogue actors today. The world doesn't sit around-
- DPDwarkesh Patel
Right.
- SNSatya Nadella
... uh, and say, "We'll tolerate that." So therefore, you know, that's why I'm glad, uh, that we have a world order in which, uh, even such, uh, you know, anyone who is a rogue actor in a rogue state has consequences.
- DPDwarkesh Patel
But, uh, if you have this picture where you could have 10% economic growth, it really, I think, like, depends on actually getting, like, something like AGI working, right? Because tri- tens of trillions of dollars of value, that sounds closer to, like, humans are, human wages are $60 trillion of-
- SNSatya Nadella
Yeah.
- DPDwarkesh Patel
... the economy. Um, getting that magnitude is just, like, you, you kind of have to automate labor or supplement labor in a very significant way. Um-... if that is possible and once we figure out the legal ramifications for it, it's, like, seems quite plausible, even within your tenure, that we figure that out. Um, uh, like, are, are you thinking about super intelligence, like the big- the biggest thing you do in your career is this or-
- SNSatya Nadella
Yeah. I mean, and by the way, you bring up another one. I mean, I know David Autor and others have talked a lot about this, which is that 60% labor. Like, what... I think the other question that needs to happen-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... is let's at least talk about our democratic societies. Um, I think that in order to have a stable social structure and democracies function, you just can't have return on capital and no return on labor. Uh, y- you know, we can-
- DPDwarkesh Patel
Right.
- SNSatya Nadella
... talk about it, but, you know, that 60%, um, has to be something that has to be revalued, right? So in my own simple sort of way, not may- call it naive is, hey, we'll start valuing different types of human labor. Uh, you know, what is today considered, um, high value human labor may be commodity. There may be new things that we will value, uh, including that sort of person who comes to me, you know, and, you know, and helps me with my physical therapy or whatever, right? I mean, we- it's like whatever is going to be the case that we value. But ultimately, if we don't have return on labor and there's meaning in work and dignity in work and all of that-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... uh, that's another rate limiter to any of these things being deployed.
- DPDwarkesh Patel
Yeah.
- 56:30 – 1:05:43
Getting AGI safety right
- DPDwarkesh Patel
On the alignment side, so two years ago you guys released Sydney Bing. And just to be clear, I think given the level of capabilities at the time, I think it was, like, sort of like a charming, endearing, um, uh, kind of funny example of misalignment. But that was because at the time it was, like, chatbots, they can g- go think for 30 seconds and give you some, um, uh, funny, uh, slash, uh, inappropriate, uh, response back. Um, but if you think about that kind of system that can, like, uh, I, I think to a New York Times reporter, try to get him to, like, leave his wife or something. Um, if you think about that going forward and you have these agents that are for hours, weeks, months going forward, just like autonomous swarms of AGIs who could be in similar ways misaligned and, um, just screwing stuff up, uh, maybe coordinating with each other, just, uh, h- what's your plan going forward to, like, when you get the big one, you, you get it right? (laughs)
- SNSatya Nadella
Oh, yeah. That- that- that- that is correct. And so that's sort of, that's one of the reasons why I think, um, we as- sort of, you know, when we even allocate compute, let's allocate compute for what is that alignment challenge.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
Um, and then more importantly, what is the runtime environment in which you are really going to be able to monitor these things? Uh, the observability around it. Like, that, by the way, you know, like, we do deal with a lot of these things today in the classical side of th- the things as well, like cyber, right? We just don't, like, we just don't write software and then just let it go, right?
- DPDwarkesh Patel
Right.
- SNSatya Nadella
You have software and then you monitor it, you monitor it for cyber attacks, you monitor it for, um, you know, you know, fault injections and what have you. And so therefore, I think we will have to build enough software engineering around the deployment side of things. Uh, and then inside the model itself, what's the alignment? Uh, and these are all, some of them are real science problems, some of them are real engineering problems, and then we will have to tackle it. And, and by the way, that also means that, like, take, uh, our own liability in all of this. So that's why I'm more interested in deploying these things in where, uh, you know, you can actually govern, uh, what the scope of these things is and the scale of these things is. And so you just can't unleash something, um, uh, out there in the world that creates harm, uh, because the social permission for that is not gonna be there.
- DPDwarkesh Patel
Yeah. What is, um... W- when you really get the agents that can, like, really just do weeks' worth of tasks for you, what is, like, the sort of, like, minimum assurance you want, uh, before, uh, you, you can
- NANarrator
It's different.
- DPDwarkesh Patel
... like, get a random fortune 500?
- SNSatya Nadella
I think, like, when I- when I use something like deep research even, right?
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
Um, the minimum assurance I think we want is before we especially have physical embodiment of anything.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
That I think is kind of one of those thresholds where-
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
... you cross. Um, that might be one place. Uh, then the other one is, for example, the permissions of the runtime environment, uh, in which this is operating. Uh, you may want guarantees, uh, that it's sandboxed, it is not going and, uh, out of that sandbox.
- DPDwarkesh Patel
Yeah. I mean, we already have, like, web search and all... You know, (laughs) we already have the out- out of the sandbox now.
- SNSatya Nadella
But even, but even the web s- what it does with web search.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
Um, and what it writes. So like, for example-
- DPDwarkesh Patel
Right.
- SNSatya Nadella
... like to your point about, like, hey, if it's just going to write a line of co- code, uh, in order to do some computation, where is that code deployed? Um, and is that code dep- eph- ephemeral, uh, for just creating that output, uh, versus just going and springing that code out into the world?
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
Uh, those are really other... Th- those are things that you could in the action space actually go control.
- DPDwarkesh Patel
Yeah. And s- separate from the safety issues, as you think about your own product suite, uh, and you think about, like, if you do have AIs this powerful, um, at some point it's not just like Copilot in the example you mentioned about wh- how you were prepping for this podcast. So more similar to, like, how you actually delegate work or, uh, work to your colleagues. Um, w- what does it look like given your current suite to add that in? And I mean, you know, there's one question about whether LLMs get commodified by other things. I wonder if, uh, these, like, databases or canvases or Excel sheets or whatever, if the LLM is your main gate point into accessing all these things, is it possible that the LLMs commodify Office?
- SNSatya Nadella
Yeah, I mean, it's possible to see... So it's an interesting one, right? I think, um, the way I think about the first phase, at least of it, would be can the LLM help me do my knowledge work using all of these tools or canvases more effectively? Like, one of the best demos that I've seen is, um, a doctor getting ready for a tumor board, um, workflow, right? So she's...... going in, um, to a tumor board meeting. And so she, one of the first things she uses Copilot for is to create an agenda for the meeting, because the LLM helps reason about all the cases, which are in some SharePoint site and says, "Hey, these cases obviously, you know, a tumor board meeting is a high-stakes meeting where you want to be s- mindful of the differences in cases."
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
So that you can then allocate the right time, right? So even that reasoning task of creating an agenda that knows even how to split time. Super. So I use the LLM to do that. Then I go into the meeting, I'm in a Teams call with all my colleagues. Guess what? I'm focused on the actual case versus taking notes because you now have this AI Copilot doing a full transcription of all of this. And just basically an intelligent, it's not just a transcript, but it's a, think of it as a database entry-
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
... of what is in the meeting that I, is recallable for all time, right? So that's, then she comes out of the meeting having sort of discussed the case and not been distracted by note-taking. And she need, she's a teaching doctor. She wants to go and, uh, prep for her class. And so she takes, and she goes into Copilot and says, "Hey, take my tumor board meeting and, uh, uh, and then take, create a PowerPoint slide deck out of it, uh, so that I can talk to my students about it." Like, so that's the type. So the UI and the scaffolding that I have are canvases that are now getting populated, uh, using AI, LLMs, and the workflow itself is being reshaped, knowledge work is getting done. Like, here's an interesting thing, right? If somebody is like ... One of the ways I think about it is, if someone came to me in the late '80s and said, "You're going to have a million documents on your desk." Uh, you know, we would have said, "What the heck is that?"
- 1:05:43 – 1:11:31
34 years at Microsoft
- DPDwarkesh Patel
Can I ask you about your, uh, some questions about your time at Microsoft?
- SNSatya Nadella
Yeah.
- DPDwarkesh Patel
Um, is being a company man underrated?
- SNSatya Nadella
(laughs)
- DPDwarkesh Patel
So you've spent most of your career at Microsoft. And look, I mean, you could say like, um, maybe one of the reasons you've been a- able to add so much value is you've seen the culture and the history and the technology and have all this context by rising up through the ranks. Should more companies be run by people who have this level of context?
- SNSatya Nadella
That's a great question. I mean, I've, I've not thought about it that way. Uh, but yeah, I mean, I, uh, I have sort of, you know, through my, whatever, 34 years now of Microsoft, it has basically been, uh, each year I felt more excited about being at Microsoft versus thinking that, "Oh, I'm a company person," or what have you.
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
Right? I mean, that is not like, I didn't go in there-
- DPDwarkesh Patel
Sure.
- SNSatya Nadella
... and saying it's ... It is about, uh, the, the, and I think that seriously even for anybody joining Microsoft, that means it's not like they're joining Microsoft as long as they feel that they can use this as a platform for their both economic return, uh, but also a sense of purpose and a sense of mission that they can accomplish, uh, by using us as a platform, right? So therefore that's the contract. Um, so I think yes, companies can, you know, have to create a culture that allows people to come in and become company people, like me.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
And Microsoft got it more right than wrong, at least in my case. Um, and I hope that remains the case.
- DPDwarkesh Patel
Well, how do you ... Like the, the sixth CEO that you're talking about that will get to use the, the research you're starting now, what are you doing to retain the future Satya Nadellas so that they're in a position to become-
- SNSatya Nadella
Yeah. Well, and-
- DPDwarkesh Patel
... the future leaders?
- SNSatya Nadella
Yeah, it's kind of f- f- fascinating. This is our 50th year and I think a lot about it, right? And the way to think about, you know, I think longevity is not, uh...... a, a goal, relevances.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
And so I think the thing that I have to do, and all 200,000 of us have to do every day, is are we doing things that are useful and relevant, uh, for the world as we see it evolving? Not just today, but tomorrow. Like, we have to basically... You know, and, and, and we live in an industry where there's no (laughs) franchise value, right? So that's the other hard part. Which is y- if you take the R&D budget that w- we will spend this year, is all about what is... It's all speculation on what's gonna happen five years from now. And so you gotta basically go in with that attitude that's saying, "Look, we are doing things that we think are gonna be relevant," and so that's what you gotta focus on, um, and then know that there's a batting average, and you're not gonna get... Uh, you have to have high tolerance for failure. That's the other thing which I think is, um, unlike, uh... You have to be able to sort of take enough shots on goal, um, to be able to say, "Okay, we will make it to the other side, uh, as a company," and that's what makes it tricky in this industry.
- DPDwarkesh Patel
I mean, speaking of, you just mentioned that you're, what, two months away from your 50th anniversary, um, uh, as o- of Microsoft's founding. Uh, if you look at the top 10 companies by market cap, or top five, m- um, uh, depending on how you count Saudi Aramco, basically everybody else but Microsoft is younger than Microsoft. And it's a really interesting observation about, like, why, why the most successful companies often are quite young. Um, the, you know, the, uh, the average Fortune 500 company will last 10, 15 years. What has Microsoft done to remain relevant for this many years? W- um, h- how, how do you keep refounding?
- SNSatya Nadella
That's a, that, uh, that, that is the... I love, I l- I love that even Reid Hoffman uses that term. I love that refounding thing.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
And I think that that's the mindset. Like, I mean, people talk about founder mode, and I sort of... I think for us mere mortal CEOs and others, it's more like, "Hey, the refounder mode." Um, and I think that it, it's, it's... To be able to see things again in a fresh way, uh, I think is the key, uh, to me. And so, you know, to your question, can we culturally create an environment where refounding becomes a habit thing, right? Which is, uh, like every day we come in and say, "Yeah, we feel we have that stake in this place to be able to change the core assumptions of what it is that we do and how we relate to the world around us, and do we give ourselves permission?" I think many times companies feel over constrained by the business model or what have you.
- DPDwarkesh Patel
Mm-hmm.
- SNSatya Nadella
Uh, and you just have to unconstrain yourself.
- DPDwarkesh Patel
Um, if you did leave Microsoft, what company would you start?
- SNSatya Nadella
Company I would start? Man, like that's where the company man and my, me services.
- DPDwarkesh Patel
(laughs)
- SNSatya Nadella
I'll never leave Microsoft. I think that if I were thinking of doing something, like I think picking a domain that has... Like when I look at the dream of tech, right? We've talked, we always have said technology is about de- the biggest, greatest demo- democratizing force. I feel like finally we have that ability, if you sort of say those tokens per dollar, per what is sort of what we can generate, I would love to find like some domain in which that can be applied, uh, where it is so underserved. Uh, that's where healthcare, education, public ser-... Well, like by the way, the other place is public service. Uh, well, uh, yeah, public sector, uh, would be another place where if you take those domains, which are the underserved places where my life as a citizen of this country or a member of this society or anywhere, what would I be better off, uh, if somehow all this abundance translated into better healthcare, better education, and better public sector, th- institutions serving me as citizens? That would be a place.
- DPDwarkesh Patel
Hmm. Um,
- 1:11:31 – 1:16:54
Does Satya Nadella believe in AGI?
- DPDwarkesh Patel
I... One thing I'm n- not sure about, hearing your answers on different questions, is whether you think AGI is a thing in the sense of like-
- SNSatya Nadella
(laughs)
- DPDwarkesh Patel
... will there be a thing which automates all cog- at least like starting with all cognitive labor, like anything anybody can do on a computer?
- SNSatya Nadella
See, this is where I... My problem with the definitions of how people talk about it is cognitive labor is not a static thing, right? Like, um, there is cognitive labor today. Um, if I have an inbox that is managing all my agents, is that new cognitive labor? And, and so today's cognitive labor may be automated. What about what is the new cognitive labor (laughs) that gets created? Both of those things have to be thought of, right? Which is the shifting... So that's why I think this distinction, at least in my head I make, is don't conflate knowledge worker with knowledge work. Uh, the knowledge work of today could probably be automated. Uh, who said my life's goal is to triage my email, right? Let, let an AI agent triage my email. Uh, but after having triaged my email, give me a higher level cognitive labor task of, "Hey, these are the three drafts I really want you to review." Like, that's a different abstraction.
- DPDwarkesh Patel
But will, will, will AI ever get to the second thing?
- SNSatya Nadella
May, but as soon as it gets to that second thing, there will be a third thing, right? So this is where I think, why are we sort of thinking somehow that we have dealt with tools that have changed what is cognitive labor in history? Why are we worried that, that all cognitive labor goes away?
- DPDwarkesh Patel
I mean, uh, I'm sure you've heard these examples before, but the idea that like horses can still be good for certain things, there are certain terrains you can't take a car on, but the idea that like hor- you're gonna see horses around the street, they're gonna employ millions of horses. They're just like, it's not happening, right? And then the idea is, could a similar thing happen with humans?
- SNSatya Nadella
But, but in one very narrow dimension, right? It's only 200 years-... of history of humans where we have valued some narrow sort of things called cognitive labor as we understand it.
- DPDwarkesh Patel
Yeah.
- SNSatya Nadella
Uh, let's just, uh, let's take something like chemistry, right? If this thing, like, quantum plus AI really helped us sort of do a lot of novel material science and so on, yeah, that's fantastic to have novel material science being done by it. Does that really somehow take away from sort of all the other things that humans can do? Um, right? So why can't we exist in a world where there are powerful cognitive, uh, machines knowing that our cognitive agency is not being taken away?
- DPDwarkesh Patel
Hmm. Um, I'll ask this question not about you, but, uh, uh, in a different scenario so maybe you can answer it, uh, uh, with- without- without embarrassment. The, um... Suppose on the Microsoft board, could you ever see adding an AI to the board? Um, could it ever have the sort of like judgment, and context, and holistic understanding to be a useful advisor?
- SNSatya Nadella
Thi- the first place, like, we added, like, I mean, just, uh, it's a, it's a great example. Like, uh, one of the things we added was this facilitator agent in Teams. The goal there, it's in the early stages of it, is, hey, can that facilitator agent with long-term memory, uh, not just on the con- context of the meeting but with context of projects that I'm working on, and the team, and what have you, be a great facilitator, right? I would love even in a board meeting, right, where it's easy to get distracted, after all, board members come once a quarter and they're trying to digest what the heck is happening with a complex company like Microsoft, I think a facilitator agent, uh, that actually helped human beings, uh, all stay on topic, focus on the issues that matter, uh, that's fantastic, right? That's kind of literally having, to your point about even going back to your previous question, uh, having something that has infinite memory, um, and then that can even help us, you know, after all, what is that Herbert Simon thing, right? Which is, we are all, like, bounded rationality, right? So if the bounded rationality of humans, um, can actually be sort of dealt with because there is a cognitive amplifier outside, that's great.
- DPDwarkesh Patel
Uh, speaking of the materials and chemistry stuff, I think you said recently that you want, in the next 250 years of progress in those fields to happen in the next 25 years. Now, um, when I think about what's gonna hap- be possible in the next 250 years, I'm thinking, like, space travel, and space elevators, and immortality, and cure all diseases. Twen- next 25 years, do you think?
- SNSatya Nadella
I think I, I ho- like, one of the reasons why I brought that up was I love that thing of, hey, look, you know, the Industrial Revolution, if you say was the 250 year, right?
- DPDwarkesh Patel
Right.
- SNSatya Nadella
I mean, if you sort of even take this entire change on from a carbon-based system to something different, um, then that means you have to fundamentally reinvent all the what has happened with chemistry over the 250 years.
- DPDwarkesh Patel
Right.
- SNSatya Nadella
And that's where I hope we have this quantum computer, uh, this quantum computer helps us get to new materials, and then we can fabricate those new materials that help us with all of the challenges we have on this planet, and then I'm all for, uh, interplanetary travel. (laughs)
- DPDwarkesh Patel
Amazing. Satya, thank you so much for your time.
- SNSatya Nadella
Thank you so much, Market.
- DPDwarkesh Patel
This was wonderful. It's wonderful.
- SNSatya Nadella
(laughs) Thanks.
- DPDwarkesh Patel
Great.
Episode duration: 1:16:54
Install uListen for AI-powered chat & search across the full episode — Get Full Transcript
Transcript of episode 4GLSzuYXh6w
Get more out of YouTube videos.
High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.
Add to Chrome