Pod Insights Decoder Recode Decode: Amazon Web Services CEO Andy Jassy (Live at Code 2019) About this episode Andy Jassy, the CEO of Amazon's cloud computing service AWS, talks with Recode's Kara Swisher at the 2019 Code Conference.
In this episode: How Jassy came to work at Amazon 22 years ago; how Jeff Bezos has changed in that time; how AWS differentiates itself from the competition; could it be disrupted by a small business?; facial recognition, sensor creep, and trust; alleged misuse of Amazon’s facial recognition tech by law enforcement; its work with the government, and employee objections to that; should tech companies work with ICE and border patrol?; potential antitrust regulation and whether AWS should be spun off; Donald Trump’s Amazon-bashing obsession; the HQ2 contest; the challenges of running AWS, including diversity; does Jassy want to be CEO of Amazon?; and how does AWS's culture compare to Amazon’s?
Learn more about your ad choices. Visit podcastchoices.com/adchoices
Transcript 9,553 words · 11 speakers
0:01 Speaker 00 What if managing company spend took less managing? Introducing AMEX Corporate, one intelligent platform connecting cards, expenses, and payments so you can keep work moving. Built to simplify, designed to scale. This is AMEX Corporate. Get started at go.amex/ammexcorporate. Terms apply.
0:29 Speaker 01 You could try to find the perfect coffee to water ratio. Or you could press a button. With Keurig, each K-cup pod is precision-measured for a perfectly delicious cup. No effort, no grind, no math required.
0:46 Speaker 01 Great coffee without the grind. Keurig.
0:52 Speaker 02 Hi, I'm Kara Swisher, editor-at-large of Recode. You may know me as a person trying to fool facial recognition sensors with increasingly silly sunglasses, but in my spare time, I talk tech and you're listening to Recode Decode from the Vox Media Podcast Network. Today, we're going to play an interview from this year's Code Conference. I talked to Andy Jassy, the CEO of Amazon Web Services, who has been at the company for 22 years. We talked about how Jeff Bezos has changed in that time, whether AWS should be spun off as an independent company, and why Jassy says the government should regulate but not ban facial recognition tech.
1:25 Speaker 03 What we have recommended to the government is that they put a constraint in place when law enforcement agencies are going to use facial recognition, that they only use results that have at least a 99% confidence parameter, and then only as one element of a human-driven decision, like one piece of evidence. And then at the end of the day, like anything else with, whether it's private sector companies or our police forces, you have to be accountable for your actions, you have to be held responsible if you misuse it.
1:52 Speaker 02 You can find full coverage of this interview and everything else from the Code Conference at vox.com/recode. But now, let's go to the Phoenician Resort in Scottsdale, Arizona, to hear my interview with AWS CEO Andy Jassy.
2:07 Speaker 02 We're going to talk now with Andy Jassy from Amazon. Come on out.
2:13 Speaker 02 We're not going to just talk about facial recognition, but since she just talked about this, actually, I want to talk about your business first, and then I do want to get into that topic, because AWS has recognition using K for your product. And I do want to talk about that a little bit. But you've been at Amazon since you were a teenager? You literally have been there 20 some years.
2:32 Speaker 03 22 years.
2:33 Speaker 02 22 years. You started as an assistant to Jeff?
2:37 Speaker 03 No, I took my last final exam at graduate school the first Friday of May of 1997. I started at Amazon the next Monday. And I didn't know what my job was going to be, what my title was going to be, what group I was going to work in. But it was very important to them that I show up that Monday. And I started off working in marketing. We just split up all the areas of marketing back then. They were all too big for each of us. And that's what I started off doing, is running customer retention and competitive intelligence.
3:01 Speaker 02 So it was very early. Why did you pick Amazon?
3:04 Speaker 03 I...
3:05 Speaker 02 People don't know you. You run one of the most important businesses, the most important businesses at Amazon.
3:09 Speaker 03 Yeah, all my experience leading up to Amazon was consumer-based. And I wanted a consumer company. And I wanted a place that was very entrepreneurial. And I wanted a place where if I was able to help start a business, there was no ceiling to what we could pursue. And my then fiance, now wife, is from Los Angeles. And we had lived on the East Coast for five years together prior to that point. And the agreement was to come out west for two to three years and then go back to New York. In fact, we wrote that agreement on a napkin in a bar when we made the decision to come out to Seattle. But that was 22 years ago. And my understanding is the statute of limitations on that napkin has expired.
3:44 Speaker 02 Okay, all right. So you decided to go there because what attracted you to Amazon itself?
3:49 Speaker 03 I loved the people. First and foremost, I was really taken by the people I met. They were really smart, really ambitious, incredibly mission-passionate. And everybody you met at the company was so focused on customers. Everything they did started with the customer and all the strategies and tactics moved backwards from there. And I believe that we had the chance to build a very unusual business, not just in books, but in other product categories as well. And I wanted to be part of it.
4:18 Speaker 02 Right. So talk a little bit about working for Jeff Bezos. He's changed a lot, but not really, in lots of ways. I remember meeting him back in the early, mid-'90s, I guess probably before you. What has evolved in him over those years?
4:35 Speaker 03 Well, Jeff is so unusual and so talented and there are so many things that are impressive about the guy. He's unbelievably inventive. He's a really big thinker. I mean, teams bring ideas to him and they seem big and then Jeff looks around another corner and makes it bigger. He has unbelievably high standards. You know, I worked for 18 months as what we then called his shadow. It's really like a chief of staff role. And I thought I had pretty high standards before I started that job and once I did the job, I realized my standards weren't high enough. But probably the thing that's most impressive about Jeff is he's an unbelievable learner. The way that he has evolved as a leader and as a manager and in the array and scope of what he works on over the last 15 to 20 years
5:20 Speaker 03 is very unusual.
5:21 Speaker 02 Okay, you've already got the job. So what has he evolved into? What has he improved at? And what don't you like about him?
5:29 Speaker 03 I think that...
5:30 Speaker 02 If you want.
5:32 Speaker 03 I think that...
5:33 Speaker 02 Jim Bankoff is too nice, but go ahead. That's my issue with him.
5:36 Speaker 03 You know, Jeff, in the earlier days, he was involved in virtually everything the company did, every decision the company made. It was as a books-only and then as a retail-only company, that was easier to do. But anybody who's had a chance to build a larger business with many more people and many more market segments realizes that you can't be involved in every meeting and every decision as the company grows. And there are some leaders and managers who can make that adjustment because you have to build different mechanisms to still be able to see what's happening in the business without having to be in every meeting. And some don't make that transition. And he was really capable of finding a great leadership team, being clear with them about what the objectives were, and then let them go around the business.
6:21 Speaker 02 Now, you all haven't left. You've all, the same group of people have been there forever. Why is that? Is it culty? What's going on there? Netflix has been like that too.
6:31 Speaker 03 I think that, I think there are a few reasons. I think, first of all... There are so many people who've been there 20, 21 years, 22 years at this point, that right, wrong, or indifferent, they almost feel like quasi-founders. There's that type of passion and ownership. And I think that we all believe that we have the chance to do something really unusual in business. And it's, you know, some companies are lucky enough to build an amazing customer experience in one market segment or business segment. Some do two. It's rare that you do three like Amazon has done. And so I think we just feel like we, you know, our work is not done. We have a great chance to build a lasting company that outlasts all of us. And it's fun and interesting. We get to be builders.
7:14 Speaker 02 You moved to AWS, which sort of was the saving grace of Amazon because it was under, you were there during, I was covering it for the Wall Street Journal, the Amazon dot bomb cleared, can this make money? AWS sort of changed the equation for you, for the company in that regard. It was a huge moneymaker. It created a lot of support for everything else that now is working really well. Can you talk about that transition for you?
7:36 Speaker 03 Sure. I was working for Jeff in this chief of staff role, and we had this idea to explore a business where we built infrastructure technology services that together comprised a platform that allowed other companies to build their technology applications on top of. And we, there were a number of things that made us think the idea might be a good idea, but probably the biggest one for us was that we felt like we were moving slower in our consumer business and delivering software than we wanted. And when I went and spoke with a lot of the different product leaders, they said, "Look, I know you guys think these projects should take two to three months end to end. We're spending two to three months just on the storage or the computer, the database, the analytics, nothing. We're building scales beyond our own initiative, and we're all reinventing the same wheel."
8:21 Speaker 03 And even though that's fairly obvious in retrospect, at the time, that was a big realization for us because Amazon is a very strong technology company. We figured if we had that problem, probably lots of other companies did too.
8:32 Speaker 02 Right. And so you did this in a pretty empty field. No one else seemed to follow you, now that everybody is there. But did you wonder why nobody followed you into it? Because when I talked to Sundar about why they didn't, he's like, "Oh, we forgot." We essentially, that's essentially what he said.
8:48 Speaker 03 Okay, that's interesting.
8:50 Speaker 02 Yeah.
8:51 Speaker 03 You know, when we were, we were working on defining the services, hiring the team, defining the services and building them for about two and a half years, in between when we had the idea and wrote the vision document and got approval, and when we launched our first service in March of 2006. And it was actually quite difficult to hire the team without telling anybody what they were going to be working on. And we were trying just to get to launch without our friends across the lake knowing.
9:15 Speaker 02 Microsoft.
9:16 Speaker 03 Yeah, and we felt, at that time, Amazon was not known as a technology provider to companies. We felt like it was really important for us to be first to market, to have a chance to be successful. So, we, I was, I was hoping we could just get to launch without anybody else knowing and beating us to the market. And in my wildest dreams, the many surprises we had, I never imagined we'd have a six-year head start.
9:40 Speaker 02 Yeah.
9:41 Speaker 03 I don't know exactly why others didn't follow. I think, I think for some of the older guard technology companies, our model was very disruptive to their existing businesses, and it's really hard to cannibalize yourself. And so I think they kind of wished it away.
9:55 Speaker 02 The Oracles.
9:55 Speaker 03 The Oracles, the IBMs, and the, you know, I think it was, it's hard when you have an existing business that's working, it's hard to cannibalize it with a model that is much lower margin. And I think that some of the other players, you know, probably were distracted with other things they were working on, and then their, their initial attempt at the business would turn out to be the wrong abstraction. It tended to be a higher abstraction when builders really wanted the individual building blocks to construct and stitch together however they saw fit.
10:22 Speaker 02 Well, that's changed. They did catch, they are starting to catch up. Now you have Microsoft, you have Google. How do you, how are you going to differentiate yourself? It becomes now a much different, what's the strategy in order to?
10:33 Speaker 03 We have a pretty significant market segment leadership position in this infrastructure cloud computing space. And there are a few reasons for it. And I think the first is, we just have much more functionality by a large amount than anybody else. And we're also iterating at a faster clip. So when you actually look at the details, that gap in functionality is widening. And that turns out to really matter if you're an enterprise or a government that's going to move all their applications to the Or if you want to be able to unleash your builders to build anything they could imagine. The second thing is we just have a much larger ecosystem of partners around our platform. It's not just the thousands of systems integrators who build practices on AWS, but most ISVs and SaaS providers will adapt their software to work on a technology infrastructure platform.
11:17 Speaker 03 Few will do two, hardly any will do three, and they all start with AWS just because of our leadership position. So you get to move to the cloud with a lot more of the software you want to use. And then the third thing is pretty different is that we're just at a different operating maturity than these other providers having been at it.
11:34 Speaker 02 Well, you said that. You said there's no compression algorithm for experience.
11:37 Speaker 03 Yeah, that's an expression we use.
11:39 Speaker 02 That sounds great, but it is Google and it is Microsoft. These are not small guys who don't think this is an important business.
11:45 Speaker 03 It turns out it's really different running large-scale infrastructure for yourself and your company, where you get to tell everybody the way it's going to be than it is for running it for millions of external customers with every imaginable use case all over the world where they get to use you without any warning. It just forces a different type of operating discipline and rigor. And you can see that borne out in the operational performance of the different clouds.
12:08 Speaker 02 We're going to take a quick break now. We'll be back to this interview with Amazon Web Services CEO Andy Jassy after this.
12:14 Speaker 04 It's me, the
12:20 Speaker 04 We don't need the jingle twice.
12:23 Speaker 04 Stop it. This is about vacation inflation and how Priceline negotiates amazing deals on hotels, flights, and rental cars. Yeah, but I didn't mention that you can save up to 60% off hotels in the Priceline app. Fine. No one deals more deals than
12:41 Speaker 04 Please stop. Touche. Priceline.
12:45 Speaker 05 How about a latte?
12:46 Speaker 01 Maybe dark roast?
12:47 Speaker 05 Vanilla?
12:48 Speaker 01 Caramel?
12:49 Speaker 05 Blueberry?
12:50 Speaker 01 You could get them to agree on one flavor or give everyone what they want. With Keurig, there's over 500 K-Cup pot varieties of fresh, flavorful coffee.
13:02 Speaker 01 No effort, no grind, no settling.
13:07 Speaker 01 Great coffee without the grind. Keurig.
13:12 Speaker 02 So what percentage market share do you have now? You have...
13:15 Speaker 03 I don't know the answer to that.
13:16 Speaker 02 Yes, you do.
13:17 Speaker 03 No, I really don't.
13:19 Speaker 02 Yeah, you do.
13:19 Speaker 03 We're the only ones that disclose our infrastructure cloud platform numbers publicly.
13:25 Speaker 02 Twenty-five billion, is that right?
13:27 Speaker 03 What's that?
13:27 Speaker 02 Twenty-five billion, is it?
13:28 Speaker 03 It's a $31 billion revenue run rate, growing 41% year-over-year.
13:32 Speaker 02 Right, so that's pretty good. Where do you see the competition? Which one of them do you worry about? Google or Microsoft? I would imagine Microsoft.
13:40 Speaker 03 You know, we don't spend a lot of time worrying about...
13:43 Speaker 02 Oh, that's the other one.
13:44 Speaker 03 You know that, Kara.
13:45 Speaker 02 Oh, God.
13:47 Speaker 03 But, you know, I think that...
13:48 Speaker 02 God, you can't do that.
13:49 Speaker 03 In my opinion, you know, we're pretty early in this space right now.
13:54 Speaker 02 Okay.
13:54 Speaker 03 And I don't think they're going to be... In the infrastructure technology space, I don't think there are going to be 25 winners because scale really matters.
14:01 Speaker 02 Right.
14:01 Speaker 03 But there's not going to be one. The market segments that we address in infrastructure, software, hardware, and data center services globally is trillions of dollars ultimately, so it's not going to be one. There are going to be several successful players. And, you know, I do believe Microsoft will have a business there. I think they're building a business and they're, you know, they're the clear number-two player at this point. And I think there'll be other players who are successful as well.
14:23 Speaker 02 How do you assess Google's efforts?
14:26 Speaker 03 Well, you know, I think they're...
14:28 Speaker 02 They've had some ups and downs. Diane left.
14:31 Speaker 03 I think they're working at it.
14:32 Speaker 02 Yeah, working at ups and downs? Or downs?
14:35 Speaker 03 I think they're working at the business.
14:36 Speaker 02 Yeah.
14:37 Speaker 03 And we'll have to see over time how big the business is.
14:40 Speaker 02 What about smaller players? Do you see it? Where would be the threat for Amazon Web Services?
14:45 Speaker 03 Well, I think that in all of our businesses, there are startups that none of us know about today that have the ability to disrupt. And there are... If you think that technology changes the last 10 years have been disruptive, and I think they've been unbelievably dynamic, I think the next 10 years are going to be faster than the last 10. And so there's all kinds of new technology that evolve that give people a chance to build businesses and pursue various segments. And so I don't know exactly who there'll be in our space, but I'm confident there will be.
15:13 Speaker 02 So what areas do you find the most promising of what you're offering in the features? And I will get to facial recognition.
15:17 Speaker 03 Yeah.
15:19 Speaker 02 When I wrote your PR person, he's like, "What's that? Like, recognition?"
15:21 Speaker 03 I don't think he said that.
15:22 Speaker 02 He was a joke.
15:26 Speaker 03 You know, there are so many that we're excited about. You know, I think that one of the obvious ones, machine learning and AI. I think most applications in the fullness of time will be infused in some way with machine learning and artificial intelligence. I think that when people think about, often people talk about hybrid infrastructure, which they mean today to mean some on-premises and some in the cloud. I think that in 10 years, when people talk about the on-premises part of hybrid, they're not going to be talking about on-premises, infrastructure and servers. All those are moving to the cloud. The on-premises part is going to really be devices. There are going to be billions of these devices sitting in our homes and our factories and oil fields and agricultural fields and cars and planes and ships, you know, everywhere.
16:08 Speaker 03 And those devices are going to really need to use the cloud to do large-scale analytics and then push the predictions to the edge on the devices themselves. So we're really excited about the edge.
16:18 Speaker 02 That's a worldwide sensor network.
16:20 Speaker 03 Yeah, and those sensors, by definition, have disproportionately small CPU and disk. So they really need the cloud to help them be effective and make decisions and do analytics. We're really excited about voice as well. And I think that when we first had apps in our phone, it was so neat to, in a few taps, and you had all this information. And when you use great voice apps, tapping on a phone feels so circa 2009. And so, you know, I think that voice apps have the potential to really revolutionize a lot of our customer experiences, too.
16:48 Speaker 02 Okay, so let's talk about that, the idea. When you think about the world, like sensors everywhere then would come into the cloud, right? This is...
16:56 Speaker 03 Well, they feed data to the cloud.
16:57 Speaker 02 They feed data to the cloud. But people are nervous about this. First of all, for the devices themselves. Now, you don't run that division, the Echo division, but it's an important part of Amazon's future. You have those there, and there's been all kinds of stories, whether by accident or not. And I'll use Nest as the example. They just happen to have a camera in there that they forgot to tell you about. And things like that. People get nervous about this idea of sensors to start with. And then moving it all up to the cloud for it to be used by various companies that you help create.
17:30 Speaker 02 Talk about that concept of how you sort of assuage people that this is not a freaky thing.
17:34 Speaker 03 At the end of the day, whatever company you're using a product or a service from, you have to trust them. And if you don't trust them, you shouldn't use them. And the effectiveness of sensors on devices is that they're collecting all this information. And today, all that information is, historically, has just been wasted. Nobody's been able to analyze it. Nobody's been able to see what customers are doing to provide a better customer experience. So the promise and the capabilities you can provide customers is really amazing. But you're going to want to make sure whoever you use that you understand their rules on privacy and how they manage their data. And you have to believe they're going to keep the data secure. If you don't trust that company, then you shouldn't use them.
18:18 Speaker 02 All right. Speaking of that, talk about recognition.
18:21 Speaker 03 Yeah.
18:22 Speaker 02 It's gotten a lot of -- people are worried about this. There's the idea -- there's some glitches today. This was not you. Customer and Border Patrol seem to have lost a lot of license plates and faces of people. Millions and millions and millions of data because one of their companies that they use to do this got hacked, had a malicious hack. And so now everybody has pictures of people crossing. And there's millions of border crossings happening at any time. And then recognition, you had some police issues. You told them they didn't get the settings weren't quite right. Can you talk about why you all did recognize -- explain what recognition is for those.
18:58 Speaker 03 Sure. So facial recognition technology allows you to kind of look at a set of faces and compare them to a data set of other faces you look at and see if there's some kind of match. And each result that you get has a confidence prediction level. And so some results, the system is very confident about and will have a 99-plus percent confidence prediction. And some, it'll have less confidence about it and it'll have a lower prediction number. And I think the issue around facial recognition technology is a real one. And I understand why people are concerned about it. And we're concerned about it, too. And if you knew what a lot of the senior leaders at Amazon spend their time -- ...on when they're not at work.
19:41 Speaker 03 They spend their time on issues of racial equality. And people are very passionate about it. I spend a lot of my time on that too, so I care a lot about it. But I really strongly believe that just because technology could be misused, doesn't mean that we should ban it and condemn it. And if you look at, just look at what you can do with computers and with servers. Look at what happened to Sony a few years ago where they had their email systems hacked into and all the corporate emails exposed. If you banned computers and servers, think about what a different world we would live in today.
20:12 Speaker 02 Yeah, but you don't have to go back, come on.
20:13 Speaker 03 No, but I'm just saying, new technology, there are ways, actually not even with technology, you could use a knife in a surreptitious way. There are things that you can do that you have to trust people to act responsibly with. And so when I look at facial recognition technology, our recognition service, in the two and a half years that we've had the service out there, there's been a lot of societal good done. You know, Thorn, the organization Thorn is using it to fight human trafficking. We have organizations that are reuniting missing kids with their parents through recognition or better security solutions or identity solutions or education solutions. And in the two and a half years we've had the service out there, we have not had one report of misuse by law enforcement.
20:54 Speaker 02 Wait, there was some problems they had, right?
20:56 Speaker 03 There wasn't a problem. There was a reporter that did a report on a police organization that used recognition. But they didn't use it appropriately and they didn't do anything. There was no misuse report or anything like that. They were interested in how they were using it and they were experimenting with it. But all that said, I actually understand why people are worried about it. We give very clear guidance to our, any of our customers who are law enforcement organizations who may use the service in such a way where you might impinge civil liberties, we strongly recommend that they don't use any results that aren't at least 99% confidence levels and that only is one piece of a human-driven decision like any other piece of evidence. And if people violate the terms of service, we suspend them from using not just that technology, all of AWS.
21:41 Speaker 03 But if you want more protection, which I think is totally reasonable. We should, the federal government should regulate it. And they should have ramifications if you misuse the technology.
21:51 Speaker 02 I think one of the things people are worried about, you get that, is that it's just a police force that is already not doing a great job at all, starts to get their hands on this stuff and misuses it, is not out of the realm of possibility, given the behavior of so many different law enforcement agencies.
22:08 Speaker 03 Nothing is out of the realm of possibility.
22:09 Speaker 02 It's not even just, but it's very in the realm of possibilities.
22:12 Speaker 03 Look, I mean, if you look at, there are lots of ways that you can get an investigation wrong, and that have been gotten wrong over time. But I think simply because you could get it wrong, doesn't mean you should ban the technology. And I really think, though, that people are looking for that, those extra set of protections around the federal government explaining how they want the technology to be used, and having real ramifications if you misuse it. And I wish they'd hurry up, because if they don't, what's happening is you're gonna have 50 different laws in 50 different states, and you're gonna have some that overreach.
22:47 Speaker 02 What would you like from the federal government in that?
22:48 Speaker 03 You know...
22:50 Speaker 02 First of all, you don't believe it's a violation of civil liberties to have your picture taken without consent?
22:55 Speaker 03 I think that, we believe this, both with pictures as well as video, that people should actually have to disclose they are somewhere. They have to disclose that they actually do have some kind of surveillance, and I think that's a good policy for both video and for photos. And what we have recommended to the government is that they put a constraint in place when law enforcement agencies are gonna use facial recognition, that they only use results that have at least a 99% confidence parameter, and then only as one element of a human-driven decision, like one piece of evidence. And then at the end of the day, like anything else with, whether it's private sector companies or our police forces, you have to be accountable for your actions, you have to be held responsible if you misuse it.
23:40 Speaker 02 And where do you play that role? You just kick them off of AWS? That's your role?
23:44 Speaker 03 If we find that, look, we can't put people in jail, if it's not part of the purview of what we do.
23:49 Speaker 02 Not yet, go ahead. Amazon jails coming soon to Prime.
23:56 Speaker 02 Just ignore me.
23:59 Speaker 03 What we can do is give very strong guidance. We also are offering police officers free training so that we can help them learn more about how to use the service in case they feel like they want to learn more and they can't understand it from the documentation in some fashion. But we really would like to help train more and more of the law enforcement agencies using it. And then if you use the service inappropriately, you're not going to be able to use our platform.
24:24 Speaker 02 All right. Getting into the same controversy, Amazon, you're moving heavily into government, working with government. There's been objections by different companies, like Google, for example, and Maven. What is the level of objection at your company towards the work you do? It's not as significant?
24:40 Speaker 03 No, I think if you have a company like we do with 625,000 people there, you're going to get every possible opinion you could imagine, which I think is great. I think that it makes us a better company and we get better ideas on the table. But by and large, the vast majority of the company is supportive of serving the government. And I think importantly, if our government doesn't have access to all of the most modern, sophisticated technology that the private sector has, we're in trouble, both in light of what the government has to get done and the role this government plays in the world. So we intend to continue.
25:18 Speaker 02 Continue to do that. And you've been getting some pushback from other companies like Oracle and others.
25:22 Speaker 03 Not about that.
25:24 Speaker 02 Not about that. But in terms of competing in this space, what changes when you're doing this?
25:30 Speaker 03 What changes?
25:31 Speaker 02 Changes for Amazon? Because people know you for other things and not this.
25:35 Speaker 03 Well, you know, I think if you want to have a public sector business with, you know, we have about 5,000 government agencies worldwide using AWS and about 10,000 academic institutions and about 25,000 nonprofits, you have to provide enough value that changes the equation for what these organizations can do for their constituents. And because the cloud allows them to meaningfully save money and to move much more quickly to provide better customer experiences for their constituents and then to spend their scarce engineering resource and what really differentiates the mission, instead of the undifferentiated heavy lifting of infrastructure, we're having a lot of success.
26:15 Speaker 02 Do you have any debates internally about, I know Marc Benioff and I have talked about this a lot, the pushback that he got working just for Customs and Border Patrol, not ICE itself? Do you think about that, providing services when things are happening at the border? Is that because tech companies now, look, everyone's in the business, but tech companies have put themselves out there as more.
26:34 Speaker 03 Yeah, I think at a high level, the way we think about it is that we want to provide the government with the ability to have the same technology that our private sector customers do.
26:44 Speaker 02 No matter what they're doing.
26:45 Speaker 03 Well, they have to act responsibly and they have to be held accountable if they don't act responsibly. And I think that's true across government in everything they do.
26:53 Speaker 02 Have you pushed back on anything? Do you do work for ICE right now?
26:59 Speaker 03 We don't, you know, as a matter of policy, we don't disclose customers who don't give us the right to reference them, which may mean they're not customers or they just haven't given us permission. But we, again, we will serve the federal government and they're going to have to use the technology responsibly.
27:14 Speaker 02 But there's no government agency you wouldn't work for? Unless it's the Defense Department or some, there's something they're doing, you would just...
27:22 Speaker 03 Any government department that's following the law, we will serve.
27:25 Speaker 02 Okay, okay, well, that's a lot of them. We're going to take another break now, but we'll be back after this with Amazon Web Services CEO Andy Jassy, live on stage at the Code Conference.
27:36 Speaker 04 It's me, the We don't need the jingle twice. Stop it. This is about vacation inflation and how Priceline negotiates amazing deals on hotels, flights, and rental cars. Yeah, but I didn't mention that you can save up to 60% off hotels in the Priceline app. Fine, no one deals more deals than Please stop. Touche. Priceline.
28:34 Speaker 02 Let's talk about regulation. You have this business that's growing astonishingly. And you have clients like MLB, NASCAR, Netflix, and obviously you compete with Snapchat at Google.
28:45 Speaker 03 Yeah, we have Snapchat as a customer.
28:46 Speaker 02 You have Snapchat as a customer, right. One of the things, obviously, government is -- you're interested in government, but government is interested in Amazon right now, probably a way that's a little different. There's a lot of questions of the FTC looking into Amazon. I think Facebook and Google are sucking up most of the auction, but Amazon is the third one that's often mentioned around antitrust and things like that. Are you surprised by this?
29:08 Speaker 03 I don't know if we're surprised by it. I think that when you're able to build multiple customer experiences in different business segments that people really respond to, your business is going to get larger. And I think as your business gets larger, there's going to be more scrutiny, and we expect it. And you have to run the business to be comfortable with that, which is the way we've tried to do it.
29:29 Speaker 02 So do you expect -- what regulation are you yourself expecting? One of the questions is whether AWS should be spun off. You and I have talked about this, and when we last talked, you said, "I don't see any reason for that."
29:40 Speaker 03 Yeah.
29:41 Speaker 02 How do you feel about that now?
29:42 Speaker 03 I feel the same way. I mean, I just don't see it. I mean, I would never say--
29:46 Speaker 02 You're one of the biggest businesses around.
29:47 Speaker 03 Yeah, I would never say never about anything. I just don't see it. Typically, when companies spin off subsidiaries, it's either because they want to get that company off the financial statements, or they just can't afford to fund the new business the way it needs to be funded. Neither of those are the case, so I just don't see it.
30:04 Speaker 02 Do you see any benefits to spinning it off?
30:07 Speaker 03 Not really. I mean, I think that AWS customers shouldn't want us to spin it off, because what they really want us focused on is providing highly secure, highly operationally performant services, where we're iterating at the rate we are right now, over 2,000 services and features a year we deliver. They want more capabilities. They don't want us distracted with building new financial systems, or an ERP, or an HR system, or doing analyst calls or earnings calls. There's a distraction in spinning off an organization.
30:39 Speaker 02 You don't want to do analyst calls?
30:40 Speaker 03 Well, I mean, I sometimes end up doing some analyst conversations, but no, I think that we'd like to stay focused on providing capabilities.
30:49 Speaker 02 So are there any benefits to doing that if you're not forced, say, by the federal government?
30:54 Speaker 03 Not that rise to the, you know-- To outweigh the benefits of staying together.
30:58 Speaker 02 Why does everyone think you're going to do it?
31:00 Speaker 03 I don't know.
31:01 Speaker 02 Right, because you keep saying you're not.
31:02 Speaker 03 I know, I'm obviously not being compelling enough.
31:05 Speaker 02 Yes, compelling. What if the government forces you to?
31:11 Speaker 03 Well, I mean...
31:12 Speaker 02 You're the obvious candidate to spin off, possibly.
31:15 Speaker 03 Again, I can't speak to what the government's thinking or will do, but at the end of the day, we operate in the United States and we will follow United States law.
31:24 Speaker 02 Right.
31:25 Speaker 03 So if we were forced to do it, I guess we would have to do it.
31:28 Speaker 02 Is that a big worry within the top... You're one of the top three executives at Amazon, right? It's you, Jeff Wookiee. There's just a couple of you that are running the most important businesses. Is it something you all talk about a lot?
31:39 Speaker 03 We don't spend a lot of time talking about it. We're very aware of it. And what we really talk about as it relates to this topic is that over the last five years as the scale of the company has gotten so much larger, it's kind of amazing the way the business has grown, not just the revenue part of the business, but just the number of people at the company that...
31:59 Speaker 02 How many do you have at AWS?
32:01 Speaker 03 We don't publish that number, but it's tens of thousands. It's a big team and it's growing really fast. And we just talk about the fact that while we don't necessarily know of anything imminent, just as you get bigger, you're going to get more scrutiny. And so we just have to make sure as a group and as a leadership team, we run the company such that if people want to inspect more closely, we'll be proud of it.
32:22 Speaker 02 So how do you deal when you're getting, say, Trump starts tweeting? Trump is, seems obsessed with Jeff Bezos in some fashion and Amazon itself and keeps calling Amazon Washington Post. What did you think when he bought the Washington Post? You thought, "Oh, shit."
32:37 Speaker 03 I'm not good for Jeff, you know? And I think good for the media business because I think, as you know, you know this better than me, there's a lot of change happening. And I'm impressed with what they've done. But, you know, I think that you can only control what you can control. And anyone who's worked on a business knows that. And we have been a business through several administrations and the thing that we have tried to stay focused on is customers and doing what's right for customers over a long period of time. And in any short period of time, there are lots of things that can happen. But if you do the right things for customers over a long period of time, usually things work out and so we think about it.
33:15 Speaker 02 Does the company get bothered by those attacks? Because they're frequent, they happen...
33:20 Speaker 03 I'm sure we don't prefer it, but at the same time, it's really, it's not a big distraction internally. You know, we were really, early on at Amazon, you know, if you look, for instance, at our stock, it has always been, you know, very up and down. And Jeff used to say constantly, you just can't be distracted by what's happening in the stock market at any one time. If you think you're 30% smarter when the stock is up 30%, you're gonna have to think you're 30% dopier when the stock's down, and that's not such a pleasant place to be. And so, we have, we have always had a lot of attention on the analyst side and the press side, and the company's by and large been pretty good about staying focused on what matters most long-term.
34:03 Speaker 02 Did you, you were not involved in the headquarter, were you gonna put part of the headquarters of AWS there, the second headquarters or the third, whatever you people were calling it, headquartered to, I just, I thought it was a bit of a circus. But, what did you, were you surprised by the backlash at Amazon? And this wasn't something that was your responsibility necessarily, but you're at the top echelons of this company.
34:23 Speaker 03 The backlash?
34:24 Speaker 02 The backlash against the, New York, yeah.
34:27 Speaker 03 I think we were surprised, you know, I think that there was a lot that changed in between when we made the decision to do it and when we started to take action, starting it, and there were, there were different politicians in place, and there were, you know, different people whose voices were really being most listened to. And, you know, I think it's a shame for everybody involved. I think it would've been good for New York, and we were excited about it. But when you start a second headquarters, and you know, to your point, it was two second headquarters, one in Virginia and one in New York. When you start one of those, it's a lot of work and energy to put 25,000 people in a new location, and it's just hard to start one of them when there's that much vitriol. It's not the right place to put your employees.
35:11 Speaker 03 So we were surprised by it and disappointed by it, but you know, we still have 5,000 employees in New York City, and we'll continue to grow, and Virginia is really excited and supportive about the, they own the 25,000-person.
35:22 Speaker 02 So is there gonna be another headquarters too?
35:24 Speaker 03 We don't have a plan for that. We're gonna have some amount of people in Nashville, which we've announced, but
35:29 Speaker 02 Right, but that's not headquarters three.
35:30 Speaker 03 We don't plan to replace New York City.
35:32 Speaker 02 Right. Why do you need another headquarters, may I ask? I found it.
35:36 Speaker 03 Yeah, it's, when you're
35:37 Speaker 02 Besides entertaining us with mayors acting like horses' asses, but go ahead.
35:43 Speaker 02 The mayor thing.
35:44 Speaker 03 When.
35:46 Speaker 02 You ignore me.
35:47 Speaker 03 I'm not ignoring you.
35:47 Speaker 02 Yeah, no, it's OK. I like it.
35:49 Speaker 03 I'm enjoying your comments.
35:50 Speaker 02 All right, OK, good.
35:53 Speaker 03 You know, when you're growing at the rate that we are across the number of businesses we are, you can't convince everybody to move to Seattle. You have to have talent all over the country and all over the world. And so we were trying to find, and we have teams in probably about 20, 25 different cities around the world, and we were trying to have a few places where we had our largest number of employees, because it makes it easier to do the right career development and move people around. And so we were trying to pick places that we thought had great talent and where the cities would want us to be there, and we're excited about Virginia.
36:26 Speaker 02 And in terms of the greatest challenge you face, being a, you're a CEO of Amazon, right? You get the CEO title within it, and you don't want to spin off. What is the greatest challenge you think you face? Is just getting more of the business or capturing more of the market share or keeping up with innovation?
36:43 Speaker 03 There are so many.
36:44 Speaker 02 You are essentially C. You are, you have, are running your own company within a company.
36:48 Speaker 03 Yeah. There are lots of challenges, and if you look at AWS, even though it's a $31 billion revenue run rate business, growing 41% year over year, cloud computing, public cloud computing represents just about 3% of the total workloads in IT. So the overwhelming majority of these workloads live on premises. So we are so early in our space. So we have, we have a lot of countries that we haven't expanded infrastructure into that we will need to. We, we're building a pretty big field, you know, of sales, account managers, and professional services and support, but we're still not at the coverage levels that we need to help customers all over the world the way they'd like and we'd like.
37:32 Speaker 03 We have loads of new services that we're going to build, lots of new technologies that we're pursuing. And we're, we're so early in our business. You know, and I think that probably, you know, one of the big things that. We all think about as well as leaders is just the culture. You know, I think one of the real secret sauces of Amazon, the first 24 years has been the culture that we've had and the leadership principles that we feel really passionately about. And I think that one of the things that's hard as you grow, both in size as well as geographic diversity, is making sure you keep that culture intact and consistent and strong. And so we've worked really hard.
38:08 Speaker 02 Well, since you mentioned it, diversity, diversity. You don't have the most diverse company, side diversity.
38:13 Speaker 03 I think diversity is incredibly important. You know, if you're building products that have broad appeal like we do, if you don't have diversity in builders, you're not going to build the right products. And so we're working really hard at it. You know, we've made a lot of progress the last few years, but we're not where we want to be. I think the tech industry in general is not where it needs to be.
38:32 Speaker 02 Why is that? Why, Amazon is, you guys, I think it's because you've all been there and you're, nobody's leaving for some reason. Like, there's not openings.
38:38 Speaker 03 I mean, some of it, when you look at something like some of the senior-most leaders, you have a lot of longevity there, which has really helped the company. But you're also right that it makes it harder to move into those. But we have some unbelievably strong women leaders at the company. I mean, if you look at Teresa Carlson, who runs our public sector business at AWS, complete superstar, and Stephanie Landry runs Fresh and Prime, and Maria Rens, who runs our delivery experience, and Tony Reed, who runs the Alexa experience. We have really strong leaders, but I, you know, I think that in general, tech companies, most tech companies have been focused on hiring more women, understandably, and I think we're all making progress even though it's not where it needs to be. I think that women look like a pretty dominant majority compared to African-American people
39:23 Speaker 03 inside tech companies. I mean, we, we have to do better and we have to be more inclusive. And if you look at where the economy is headed and where the opportunities are in our country, there are significant segments.
39:35 Speaker 02 So why aren't you?
39:36 Speaker 03 Well, we're working at it, it just takes time. We have to, you know, you have to develop more leaders and you have to actually be recruiting in the right places and, you know, we're working hard at it, but, you know, it's a work in progress for us.
39:48 Speaker 02 All right, last question, then we'll get questions from the audience. Do you want to be CEO of all of Amazon?
39:52 Speaker 03 Oh, well.
39:53 Speaker 02 Say Jeff decides he wants to go to Mars, which he does.
39:58 Speaker 03 Yeah, I don't think Jeff's going anywhere anytime soon, and we're real fortunate about that.
40:03 Speaker 02 So you're not like Tim Cook or something like that?
40:05 Speaker 03 You know, again, I haven't thought anything about it. You know, if and when it becomes something that's a real possibility, then I'll think about it, but I have enough things to think about without imagining scenarios that aren't real.
40:18 Speaker 02 All right, okay, good answer, good answer. All right, questions from the audience? Right here. Go ahead.
40:24 Speaker 06 Hey, Andy. Jason Del Rey from Recode. I cover...
40:27 Speaker 02 Amazon.
40:28 Speaker 06 You all. First, I'm based in New York City, so when you decided not to put your HQ two or three in New York City, I lost out on 25,000 potential sources, so thank you for that. I'm sure you're happy. Mike, I wanted to ask about some of the employee pushback to recognition, facial recognition software. And there was an all-hands meeting in the fall where there were questions asked of leadership about marketing the technology to law enforcement in general. And according to a transcript of your response that was published in BuzzFeed, I'm going to paraphrase, you said something like, "There are a lot of opinions, you know, that's great,
41:13 Speaker 06 but we feel really good about this and really good about the way law enforcement's using that." And I heard from several employees myself who felt like that was dismissive of their concerns. So my question is, what would you say to those employees who still feel like their concerns about the technology have not been taken seriously?
41:31 Speaker 03 Yeah, well, I would say a few things. First of all, what I said at that all-hands meeting is almost identical to what I said when Kara asked me the question earlier. And, you know, I've, look, I think that...
41:44 Speaker 03 One of the leadership principles that we hold most dear at Amazon is have backbone, disagree and commit. And what that means is that you're not only empowered to disagree and speak up if you disagree with something at the company or a customer experience or whatever it is, but we expect you to. And we celebrate that. But then at a certain point, the leaders have to make decisions on where we're going to head as a company. And when we make those decisions, we expect people to disagree and commit. And if that's, that can be about a customer experience issue, it can be about pricing, or it can be about the decision that we're going to continue to offer facial recognition technology to law enforcement or agencies. So it wasn't meant to be dismissive in any way, it was just trying to answer the question.
42:28 Speaker 07 Okay. Andy Simson from Swoop. Last time you and I talked about culture on a plane to Aspen, you got just over $5 billion in revenue. Now you have six times more. You said that it's really important to keep the core principles of Amazon intact, but has culture changed in some other way in this six-fold growth? And is the culture at AWS in any way different than other parts of Amazon?
42:52 Speaker 03 I think the culture in AWS is quite similar to the rest of Amazon. And, you know, and I think the culture has, in the meaningful ways, is largely the same. You know, we really think about how we're running the business in AWS as running a large startup. And we have a little bit more process than we did when we first started AWS, but we really operate in a way where we start with the customer experience, that is always paramount. But then, we all believe that speed disproportionately matters to our customers. And I think that's true, by the way, in every stage of your company, for every size company. Sometimes I think larger companies get this wrong and think that speed doesn't apply
43:36 Speaker 03 to them, especially if they have a pretty significant market segment leadership position. But just look at the Fortune 500 since 2006, more than half of those companies aren't in the Fortune 500 anymore. Business changes fast. So we, the culture is very much the same, very customer-focused, very Bias for action. We have very high standards because our customers have high standards and expect that from us. I think the biggest differences today from when we spoke when it was a $5 billion business is that back at that time, we didn't have much of a sales function. It was pretty small at that stage and we had lots of customers all over the world who wanted our help and we just didn't have enough people to do it. And that has changed significantly from that time. Quick?
44:20 Speaker 08 Andy, Mark Mahaney at RBC, I think you actually would really enjoy analyst calls.
44:25 Speaker 03 Yeah.
44:26 Speaker 08 You made that question, you made that statement about 3% maybe of workloads or workloads that could be in the cloud or in the cloud today. That's the lowest number I've heard in years. Is there something that you've seen as you've built the business over the last couple of years that makes you think that TAM has grown materially? That's a very, that's great penetration if it's true.
44:46 Speaker 03 Yeah, what I was saying was that if you look at all the IT workloads or applications that exist in the world, only 3% of them are in the public cloud right now. All the rest of them are on premises. And so I think that that is changing at a very rapid rate. I mean, you can just look at the evidence of our business growing the way it is and some of the other companies who are also growing in the space to see that. But I think some of the things that are changing that fast are, number one, if you don't have to lay out capital expense for servers and for data centers and instead get to spend that as you consume it on infrastructure, it's hugely useful. So you get to turn capital expense to variable expense. That variable expense is lower than what you can do on your own because we have such a large scale.
45:29 Speaker 03 We've lowered our prices on 72 different occasions in the last 10 years, largely in the absence of competitive pressure to do so, just because it's just part of the DNA of how we operate. And you get true elasticity. So you don't have to provision ahead of time and then live with the peak if you guessed wrong or too high. You give back what you're not using. So the cost equation is very compelling. But the number one reason that companies are moving so fast with the cloud is because of the agility and speed in changing their customer experience. On premises, if you want to try something new, it takes you 10 to 18 weeks to get a server. And then you have to build all the surrounding infrastructure software around it. In the cloud, you can spin up thousands of servers in minutes.
46:11 Speaker 03 And then we have 165-plus services you can mix and match and use in whatever way you want. And so that changes the way you can get from an idea to implementation in several orders of magnitude. So just the cost and the speed are compelling companies and enterprises and governments at a faster rate than they ever have considered before to move to the cloud.
46:30 Speaker 02 One more question, very quick, sorry.
46:33 Speaker 09 Sure. Andy, congratulations for building such an amazing business. I remember when Reinvent was about 1,000 people. I wanted to ask you about cybersecurity. My name is Yoav Leiterstorf. I had a firm, a VC firm, focused on cybersecurity. And actually, half the companies we fund are doing cloud security. And I guess part of that is because the cloud providers aren't taking care of all the issues that there are in security. And you're doing Reinforce this month. I'm just wondering, what are your plans? And are you going to continue to support an ecosystem of vendors? Or are you going to try to do it all yourself in security?
47:09 Speaker 03 Well, Reinforce, just for people that know, is a new security conference that AWS is doing in a few weeks in Boston. And you know, the way I would explain it is that AWS is responsible for making sure that the infrastructure that we provide, so our servers, our data centers, our network, and then all of our services like compute and storage and database and analytics and machine learning and messaging, that all of those services are secure. And then our customers, when they're using that infrastructure, they build applications on top, but they have to do some things themselves to make sure that their applications are secure as well. If you exposed your credentials on GitHub,
47:52 Speaker 03 it doesn't matter what we do with the infrastructure, people can get into your application. So it's always been a shared responsibility that way. And there has been, for many years, and continues to be, this very vibrant, expansive ecosystem of third-party software providers in the security space who are helping customers in every imaginable way with security. And some of them are doing it at the application layer, and some of them are helping them be more secure on the infrastructure itself and setting up permissions and accounts and guardrails the right way. But that is a very big business where you have a lot of third-party companies having a lot of success in partnership with us.
48:29 Speaker 02 All right, Andy, thank you. You're such a good sport. Thank you so much.
48:33 Speaker 03 Good to see you.
48:35 Speaker 02 Thanks again to Andy Jassy for joining me onstage at Code, and thanks to you all for listening. You can follow me on Twitter at Kara Swisher. My executive producer Eric Anderson is at Erica America. My producer Eric Johnson is at HeyHeyESJ. If you liked this episode, we'd really appreciate it if you shared it with a friend. And make sure to check out our other podcasts, Recode Media, Pivot, and Land of the Giants with Jason Del Rey. Just search for them in your podcasting app of choice. Thanks also to our editor Joel Robbie. Thanks for listening to this episode of Recode Decode. I'll be back here on Wednesday. Tune in then.
49:17 Speaker 10 20th Century Studios presents Whale Fall, the most incredible survival thriller of the year.
49:25 Speaker 10 In theaters Friday. To survive the unthinkable, do the impossible. You're gonna become so dangerous that the predators can't help but let you go. Austin Abrams. I'm going home. Josh Brogram. Survivor, survivor. Whale Fall, rated PG-13, maybe inappropriate for children under 13. In theaters Friday. Get tickets now.
Previous Recode Decode: Twitter's Vijaya Gadde and Kayvon Beykpour (Live at Code 2019) Next Recode Decode: RAICES immigration activists Jonathan Ryan and Erika Andiola (Live at Code 2019) Page updated October 13, 2026 · transcript and summary are machine-generated