BIG INTV: CTO Elizabeth Stone On The Future of Netflix: “Personalized, Interactive, Immersive”

11:00WIRED:Yeah, I'm sort of a poster child for learning on the job and learning from people around me. Sometimes people reporting to me, which has been the interesting learning. So even starting at NUNA, it was the first time working with engineers, first time working deeply with HR and legal and recruiting. So I wasn't a hands-on engineer from the earliest days of my career. I had to learn from someone who was already a VP of engineering, who all of a sudden was reporting to someone who didn't have that background. And so the thing I learned is the substance of tell me more about your job. Tell me how you think about decisions.

11:38WIRED:Tell me about challenges on the team. Like curiosity is the name of the game there to help build understanding and trust, but also understanding what I could add.

11:47Elizabeth Stone:Right.

11:47WIRED:So often what I can add is I might not be the best engineer sitting in the room, but I can use my background to say, well, I understand the business strategy. Thanks to you, I understand our technical strategy. How do we make sure those things come together? And so it's a translation role. And I think that that's probably a role I've had since before I ever even knew what economics was, which is, can I listen carefully to people? Can I connect with what you're saying and thinking with what you're saying and thinking, even if you speak different expertise or language? Yeah.

12:21Elizabeth Stone:I'm curious. I have to ask about the gender of it all and how that has shown up for you. I mean, you hold an incredibly big job today. You've had a very distinguished career. Working in tech or adjacent to tech, very sort of masculine industry. I mean, running wired, I will tell you the number of men who look shocked when they meet me and find out that I run wired. And it's a woman and it's shocking to them. And it's infuriating to me. I'm curious how you have seen that show up, if at all, in your career and how you navigate that piece of it.

13:04WIRED:in my very first job I was the only female trader on the trading desk at Merrill Lynch and to do that as a 21 year old it's kind of eye-opening you know in a big trading floor in New York City of like where am I how did I get here and am I supposed to be here It was true in grad school, very rare for women in PhDs in economics. Each job I could name, though, like gender was notable. I don't experience it day to day. I don't usually like the word lucky, but I have been attracted to places and companies where it didn't feel like it was a limitation or that it was really shocking people. I am sure I've subconsciously absorbed adaptations, you know, given those environments.

13:58WIRED:When I had less confidence in myself earlier in my career, it was whatever you do, don't be noticed for being different. Right.

14:05Elizabeth Stone:You know, find your way to fit in.

14:07WIRED:Fit in, fit in, fit in. Now the advice I usually give people with some more confidence or experience in different environments is don't try so hard to fit in that you lose what you personally can add. And I know now that I have a style where people would say, why aren't you sharper elbowed? And people sometimes will treat that as a gendered thing. And I'll say, don't confuse that for a whole bunch of other things that are necessary for my job. But I'm more comfortable having a different style than people around me now than I was 20 years ago. And I'm very comfortable saying, yes, there's definitely a path.

14:45WIRED:And if you want to use me or all these other great women in these roles as examples, I'm very happy to help support and mentor and encourage in that direction. Otherwise, I just feel like I'm doing my job, like other people are doing their job. And I've built more confidence about that over time.

15:00Elizabeth Stone:Right. You're not doing your job as a woman. You're just doing your job as anyone would.

15:04WIRED:You know what I mean? Yeah. And I do. I think it's really important to frame it that way. I want to be on a leadership panel, not only women in leadership panels, as one example. Do both, but I don't like to box myself in just based on the identity. That makes sense.

15:28Elizabeth Stone:We have to start talking about Netflix. Okay. And I'm curious, I think first and foremost, I mean, as you said, you've occupied a few different roles at the company. Now you're the CTO. What drew you to Netflix to begin with? What was the draw? What was the challenge that was presented to you that just felt like something you couldn't turn down? This may be a surprising response, but it was the people even more than the challenge.

15:55WIRED:It was the people and the culture that drew me to Netflix. So I met the person who would be my first manager when I came to Netflix, Caitlin Smallwood, who built the data science team at Netflix over more than a decade. And I thought, I want to be on this person's team. And as she described how leaders engaged with one another, the candor of the culture at Netflix, it was magnetic at the time where I thought, like, I feel like I could be at a company like this for a really long time. And so I was kind of running towards what Netflix represented in terms of the people and culture and that particular manager.

16:39WIRED:And I'm a strong believer that people join and leave companies very often because of their manager versus the whole broader backdrop of the company. that the reason why i stay at netflix is absolutely about the challenges one of the things that's changed between the time i got to netflix that was 2020 a picture tiger king picture everyone is stuck at home yeah watching netflix yes it was almost like a netflix can't lose yeah like unbelievable product market fit first mover advantage all the popular titles it felt like I was going towards a company that had it in the bag and made. Now, what changed by the time we landed in 2022 was, wow, we have tons of competition.

17:31WIRED:And it's one of the better things that I think happened to Netflix and to my own career to have to rethink some of these challenges, to say, well, it's not going to work just to keep doing the thing we were doing and expect that we're still going to continue to win with consumers or be a consumer's first choice, we have to earn that and continue to earn it. It was really a push towards like we can't get comfortable.

17:56Elizabeth Stone:Right.

17:56WIRED:So whether it's thinking about account sharing or it's thinking about having an ad supported plan. So that's hard technical problems. It's hard data science problems. It's hard business strategy problems. It's growing the company with new skill sets that we didn't have when I got here. and that's a fascinating challenge that continues to attract me and a lot of people on the team many of whom would say it feels like I'm working on a startup within Netflix if you're on the live team or the ads team or the games team you're basically a startup within a very mature company and I like that it requires both types of thinking we've got mature spaces where we want to continue to be really strong and we have to grow new ones from the ground up all at the same time.

18:42Elizabeth Stone:And I want to ask you about a lot of those challenges you just talked through. Before I ask you anything else about Netflix, though, I have to ask you because before I came in to do this interview, I asked the Wired team, you know, I'm interviewing the C2 of Netflix. What should I ask her? And they all said, please ask her, why can't we cast video from our phones to our televisions anymore? Why is this? Why is this happening? And I thought to myself, oh, I haven't thought about doing that literally ever, like since maybe I was like in a hotel room five years ago and wanted to probably watch like the finale of Love is Blind.

19:22Elizabeth Stone:But is there anything you can tell us about that sort of recent development in, you know, in the business and in sort of what you offer to the audience?

19:32WIRED:We are still allowing it for select devices. So it's not completely off the table. One of the things that has changed is what TVs are capable of and devices that are connected to smart TVs or connected TVs. So if you have a Chromecast device, if you have a Roku streaming stick, if you have a Fire Stick, those tend to be the much more popular ways to get access to streaming apps like Netflix. And since those devices tend to have remotes that make it much easier to control playback, pause, add your subtitles, that's a better member experience than thinking that people are going to purely cast from their phone and the app in their phone.

20:17WIRED:So it's kind of the evolution of TV devices and that that's a better member experience. And that's honestly where we see most people engaging. And then there's some cases, like I think Chromecast is a good example. For the more recent Chromecast devices, nothing has changed. It maybe just is changed in terms of what are the controls that are actually on your phone versus you use the remote on the TV.

20:40Elizabeth Stone:Well, I'll let them know. We seem to have a disproportionate number of casters on the Wired staff, and I'm not sure why.

20:48WIRED:We might need to look into this. I am available for IT support. Call Elizabeth. Bob afterwards. Yeah, I'm going to need some specifics on the devices they're using, what they're trying to stream. Hopefully it's only Netflix. I can't speak for other streaming services.

21:03Elizabeth Stone:Of course. We're all Netflix all the time over here, obviously. All we watch is Love is Blind at Wired. I got that. I want to ask you a little bit about the challenge that you have in live programming, which is it feels like has become a much bigger piece of the Netflix strategy in the last few years. You had this this 2023 Chris Rock comedy special. You had the Jake Paul Mike Tyson fight. Sixty five million concurrent streams, I think, on that one. But it hasn't always been smooth sailing. Like you guys have taken some L's on live. I think, if I'm remembering correctly, there was actually a Love is Blind special that was supposed to be live that ended up just getting canceled, right?

21:47Elizabeth Stone:So take us, like, behind the scenes with that a little bit.

21:51WIRED:Yeah. So maybe I can talk through that arc that you summarized with the first live event being Chris Rock. Right. In some ways, that's a great out-of-the-gate. We were able to use technology and infrastructure that we had in place for video on demand. But the technology you would build for streaming video on demand is going to be different than something that actually requires me to be able to take a signal from a camera at a stand-up comedy or on an NFL field to the cloud, to our Open Connect delivery network, to a device, wherever you're watching, essentially instantaneously. The thing we learned to do really well with video on demand is anticipate when you're going to press play and have those files be located very nearby when you do so.

22:46WIRED:So that there's very little transit of video and audio files when you click play on that title. It just magically appears and it was on a server not too far in many cases from your home. So we have 6 ,000 locations around the world where we've got servers that have our content sitting nearby. We can't do that for life. You know, we're worried about how many seconds latency is it from something that happens on the field to your TV screen. And so we had to make adjustments to our infrastructure. If you're talking about one million people watching Chris Rock, relatively doable. Many multiples of that signed in to watch the Love is Blind reunion.

23:24WIRED:We had made some changes to our infrastructure to do a better job with that transit of files and the quality of that experience. And there was a bug in our OpenConnect backbone. And in 2023, we weren't that sophisticated in the testing we had around live. So we didn't catch it until everyone is logging in and trying to watch the reunion. And incredibly painful. I mean, there's people at Netflix who they're going to remember it for the rest of their lives.

23:57Elizabeth Stone:Do you still talk about that, like the Love is Blind livestream incident? We do. We do.

24:02WIRED:The learning culture here is pretty profound in not hiding from failures or lessons learned. Internally, people would probably say we almost got lucky with Chris Rock because live is challenging for some of those reasons that I mentioned. And we continued to talk about Love is Blind because we don't want that to happen again. What's the technology we need to build? What's the testing practice, the operational sophistication we need to have so that we don't have that type of incident again? I can promise you in any technology field at this scale, all around the world, especially for live events, it's very hard to promise.

24:48WIRED:there's never going to be an issue, but it's how quickly you detect it and you recover from it that matters. When we got to the Paul Tyson fight, you're right, 65 million concurrence, it was 108 million viewers. Wow, watching that one event. I cannot believe that the service was still standing. Was it a great member experience? Not for a lot of people who experienced the rebuffering. But at that type of scale, my reaction standing in the launch room, which is in the same building I'm sitting in now in Los Gatos, California, was watching a group of engineers who had learned from Love is Blind and stayed totally calm, cool, and collected under pressure and did everything they could with truly record-breaking numbers that we watched the numbers ticking up, and I just felt my stomach turning in that moment.

25:49WIRED:We had more people in the waiting room than we expected for the entire event before the undercard fight started. So that's something that we only could have felt prepared for by going through the prior experiences. And then coming out of Paul Tyson, we had roughly four to five weeks to get ready for the two NFL games on Christmas Day last year. And one of the most memorable pieces was not just standing in the launch room watching the team navigate this unprecedented experience with Paul Tyson, but that I woke up, woke up is probably a little bit generous. I don't think I slept the night of Paul Tyson, but I woke up, I got out of bed the next morning And I already had a memo in my email inbox saying, here's all the stuff we need to improve before the NFL games.

26:38WIRED:So the team had not only gotten through the event, but were immediately learning from it, thinking about what we needed to really hustle on. And now we just crossed 200 live events this year. Wow. Multiple WWE events each week. We've had David Chang dinner time where we do some exploration of new experiences like live voting is on the David Chang show. So I won't say it's easy at this point, like we just close our eyes and it happens, but it's come a really long way since that first Chris Rock. And it's kind of the relentless, we want to have huge, exciting live events and we set a high bar on video on demand.

27:18WIRED:So we need to do the same for live.

27:27WIRED:How is your credit score calculated?

27:29Elizabeth Stone:Why is health care more expensive in the U.S. than any other country?

27:33WIRED:We answer those questions and more on Control F, a new podcast that uses data to reveal the hidden stories in our daily lives. Every episode, we tell you about the data that companies and the government track about us and how that data shapes our reality. Join the search. Listen to Control F from KUOW in Seattle, wherever you get your podcasts.

28:04Elizabeth Stone:So setting this high bar, this approach of sort of take big swings, fail or sort of come up short, iterate, iterate, iterate. How does that strategy apply in the context of, let's say, games, which I know is another area where Netflix is really pushing? Tell us a little bit about that sort of piece of the business, how it's working now. maybe what audiences should expect from Netflix in that context moving forward? Sort of where are you on that journey?

28:33WIRED:We've had to really evolve our strategy since we first entered games. If Netflix doesn't offer something differentiated or like, why would I go to Netflix to play a game relative to all the other options? And I don't think we knocked that out of the park for the first couple of years. There wasn't a reason. I don't think members even knew. like why is there a games row on Netflix?

28:59Elizabeth Stone:I admittedly had no idea.

29:01WIRED:Yeah, had no idea. And I don't, upon reflection, everything is clear in hindsight. And so we've made changes to that strategy. So that includes some of the things that we're now launching, which is thinking about games on TV, games that build on Netflix's IP, games that, you know, what we call the Party Pack. There's a new game launching each week this month, which is really about community and playing games as part of groups on the TV, using your phone as a controller. So there's technology innovation there. There's game innovation there. And in some ways, it builds on a strength we have, which is understanding streaming, especially on the TV.

29:46WIRED:We still have mobile games, but we're starting to evolve our game strategy towards the what's the special sauce that Netflix could bring. So that's required some pretty big pivots. It's required up-leveling and a lot of innovation rather than imitating what the market is doing. So some of that innovation is using the phone as a controller, thinking about how to use some of the infrastructure we have, like Open Connect that I mentioned, to support streaming of games in addition to film and TV. So we're still early on in that journey too, but after some fits and starts.

30:21Elizabeth Stone:And it's 2025, so obviously we have to talk about AI. We have to. The reality is, though, and you've said this before, and it's certainly true. I mean, Netflix has been using machine learning technology for a long time, right, for a long, long, long time in very well-known and sort of ways that I think have become synonymous with the company and with sort of your approach to programming in many ways before sort of generative AI became the only thing anyone was talking about. So I'm curious if you could talk a bit about that history with the tech and sort of how audiences who are listening to us talk today and all they're thinking about is like LLMs and video generation and sort of those conversations, how they should think about AI and machine learning in the context of Netflix and the company's history.

31:11I read somewhere yesterday someone who I assume was smart, but it's the Internet, so you never know.

31:17Elizabeth Stone:They basically said something to the effect of, like, Netflix is an AI company. I don't know if you would agree or disagree with that, but certainly Netflix as a technology company, I think there's a fair argument to be made there. So I'm curious for you to sort of elaborate on that a little bit.

31:34WIRED:There's very few companies that have the combination of strengths I like to think Netflix has. And it's on the content side, it's on the tech side, it's on the member experience or product side. So I don't like making it one or the other. Because I think the thing that's been made Netflix successful is we're this great combination. And that is an important kind of principled and thoughtful lens to put on. So what do we make of AI or Gen AI? Tech for the sake of the tech is not that interesting, especially in Netflix's business. So that's where that longer term arc of, yeah, we're very familiar with machine learning and AI recommendations is a great example of how do we make sure that we're personalizing?

32:20WIRED:We think this is the right title for you in this moment. We'll give you lots of choices, but here's a place to start. And now there's an opportunity for LLMs. LLMs take in lots more information. They could understand content more deeply. They understand members' preferences more deeply. Maybe we can do a better job with those recommendations using the advances in Gen AI. But it's with that objective in mind that it's interesting for us. Another good example would be we're piloting conversational search. So a member could use more natural language saying what mood they're in, what theme they're looking for, what actor they want to watch.

32:58WIRED:Those are all things that serve the member very well. And then there's examples of machine learning and AI that are creator-facing so that creators are able to use those tools. That inspires a lot of great innovation in technology, but it's a means to a bigger end that I think is more inspiring. And that's been kind of the focus point for Netflix, and that hasn't changed even as the tech has advanced.

33:21Elizabeth Stone:I'm curious, too. I read – this is a long time ago. I mean, this is 2018. Netflix's chief content officer said it was 70 % gut and 30 % data in terms of commissioning strategies and audience projections. I mean, obviously, that's a very old number. And then the quote continues, most of it is informed hunches and intuition. Is that still the case? Or is data now an even bigger part of that story for Netflix? If it was all data, all we would have are hits.

33:55WIRED:yeah we just knock it out of the park like it's that's the perfect film for that audience that's a perfect tv show i i wish it was that formulaic it's not so our content team has to have a lot of judgment a lot of intuition a lot of deep audience understanding a lot of cultural understanding of what's relevant and interesting or innovative right now there's not a formula that spits out the perfect set of things or the storyline or the script or that like data can inform some of that judgment and intuition but a creative exec's decision making has a lot of gut and judgment to it so it's another place where I would say it's a combination and definitely parts of my team try to be helpful for those decisions or try to be helpful as we're thinking about how can we make a title that we greenlit the best version of itself?

34:48WIRED:But that's a human endeavor. That's not an AI endeavor.

34:52Elizabeth Stone:You know, I know that the company has embraced, with caveats and with guardrails, but has embraced AI as a tool for creativity. So, for example, you know, it was used to cut cost on some really ambitious graphics work in the Eternot, which is a sci-fi series that I, full disclosure, have not watched. It's been used in other productions that have run on Netflix. is there an acceptable world to you as the CTO of Netflix where I turn it on and I am met with a script that was generated using AI, characters that were generated, you know, AI-generated video? Is that an acceptable world for you? Or do you see Netflix continuing down this road of sort of AI as tool for human creativity, but do not expect to see sort of wholesale AI generated video on this platform?

35:47WIRED:I think the human creativity is at the center of all the storytelling we've got on Netflix. So will there be fully AI generated content out there? I'm sure we see it already. Is it compelling? Does it move you? Does it have the human emotions that we crave seeing in film and TV shows. I haven't seen that. And even in cases, let's take a world before Gen AI, the applications for machine learning or AI, like visual effects, have long been around. But that was to help a creator tell a story. The way I've seen Gen AI used is very similar to that. It can make things possible that weren't previously possible.

36:31WIRED:It can help a creator bring their vision to life earlier in the creative process to make sure there's alignment around it. It can lower the barrier to entry for people to create great titles. But the human is still kind of the center kernel of that. And I believe that very deeply. And we've been able to leverage the tech in great ways over time with that type of framing. So I'd be foolish to say there won't be AI-generated content out there, but I don't think that's going to replace the humans that are involved in the storytelling that Netflix is bringing to life.

37:09Elizabeth Stone:When you sort of future cast out, let's say five years, let's get really crazy, and you think in sort of your seat as CTO, the Netflix that you want me, the end user, to experience in five years, what's different about that? What stands out to you when you're sort of looking at the horizon and thinking about the version of this company and this service, this product that you want to be creating, what's different in that timeline? What does that look like for someone?

37:40WIRED:More personalized, more interactive, more immersive. What does that mean in practice? So we'll have a greater variety of content, some of which we've talked about. So expanding into live or games or podcasts. We'll have more mobile first experiences. So things that we're trying now like conversational discovery and exploration. We're testing new features that allow members to clip and share content. So it's more social and fandom oriented, more immersive in the ability to influence the outcome of a show like live voting with the upcoming Star Search, which yes, we are bring back. I've heard. There's a role, you know, it's a less passive entertainment experience.

38:29WIRED:It's a more active one. And I think technology together with the content, if we do this right from a product experience, we'd make it feel like in five years when you log into Netflix, it's an experience just for you. And we understand what you're looking for across a wide variety of things. and you can be a part of that story as you build it. So it is an evolution. It's not, it doesn't change overnight. It's something that all of these elements of new content types or new types of products experiences build towards, but personalized, interactive, immersive would be my best summary of what it's going to feel like in a few years time.

39:07Elizabeth Stone:Well, I will call you in five years and give you my feedback on how you did. I look forward to that. Excellent. Now, before we wrap up, we're going to play a little game that we came up with here. You should have been warned. If you weren't, it's too late. It's called Control-Alt-Delete.

39:26Elizabeth Stone:So I want to know what piece of tech would you want to control? What would you want to alt, so alter or change? And what would you delete? What would you vanquish from the earth if given the opportunity? And you can feel free to sort of take a liberal interpretation here.

39:41WIRED:um so the tech i would like to control i spend a lot of time with my museum of music devices i have every music device that's ever been made i have iowa sound systems if you remember those i've got boom boxes i've got ipods i've got walkmans what are you doing with all of this stuff I they're nostalgic for me do you still listen to CDs? yes wow I do I still have cassettes wow am I a hundred years old? maybe she might be yeah so I I would like to simplify the web of things that is music speakers where music lives to be a better archive. Of course I have my iPhone and I have Spotify, Sonos sound systems and so on, but there's something there that I feel like it doesn't capture the scrapbook of music like I would like it to.

40:50Elizabeth Stone:Wow, okay, so you want to control a variety of musical devices and how they manifest in your ears, basically.

41:00WIRED:Yeah, and then just be able to move across them more easily. Got it. I like that.

41:05Elizabeth Stone:What about Alter?

41:11WIRED:Alter, Peloton kind of saved my life during COVID. Mine too. I did a lot of cycling outside and in other communities and that went away during the pandemic. I'm, how would I describe myself? a little bit of an obsessive endurance athlete. And I would like to be able to alter Peloton for making your own training program. So not having it be like, oh, I just have to do the stuff that's on demand that you're offering to I could construct my own training.

41:44Elizabeth Stone:This actually, I do need to say, I'm not saying you stole my idea because I've never said it out loud. But I really wish there was a way on the treadmill in particular where I could like program a workout with like hills, intervals, whatever I needed to do that day, and then just press play and have it run through that workout so that I'm not manually futzing with the knobs.

42:06WIRED:Yeah. Or like with a great coach layover, great music going back to like, I want to be able to pull from a whole archive of music. Oh, I see how it all comes together here. Yeah. I now want to construct the perfect training program that I could do on my Peloton or my treadmill. but it's a little less like on demand and a little more choose your own adventure. Personalized, interactive, immersive. Wonderful. With great music.

42:30Elizabeth Stone:I will say Netflix being on the Peloton is like one of the best things that's ever happened to me. So thank you for that. Delete.

42:43WIRED:The apps that control the apps that turn on the lights, close the blinds, turn on the speakers, change the thermostat that never seem to work and I feel like are contributing to my learned incompetence in functioning in my own home.

42:57Elizabeth Stone:Yeah, we're all dumb enough at this point. Why am I sitting on the couch trying to get this app to work

43:02WIRED:to turn off the lights instead of just getting up to turn off the lights? So I could probably do without that. Love it.

43:10Elizabeth Stone:Elizabeth Stone, CTO of Netflix, thank you so much for being here. This was great.

43:14WIRED:Thanks for having me. Great conversation.

43:23Elizabeth Stone:This show is produced by Jessica Alpert with help from Adriana Tapia and Sam Egan. Sound design, mix, and original music by Pran Bandy. Kate Osborne is our executive producer. And I am, of course, your host, Katie Drummond, Wired's Global Editorial Director.

44:17WIRED:Transcription by CastingWords all the noise. Click here wherever you get your podcasts.

44:26Elizabeth Stone:From PRX.