Transcript

Duolingo’s battle for learning in an AI world, with Luis von Ahn

0:01 The very best founders I know are brilliant at building systems. They connect teams, they remove bottlenecks, and they eliminate single points of failure. And yet When it comes to their own wealth. Most are running a disconnected stack. A tax accountant here and a state attorney there, a wealth manager who doesn't talk to either one of them.

0:22 Creative planning was built to fix exactly that. One integrated team of tax professionals, state planners, investment specialists, all coordinated by a dedicated wealth manager who sees your full financial picture and keeps every piece working together. Proactive tax efficiency, state strategy, investments all under one roof. Creative planning where wealth works together. Learn more at creative planning dot com slash masters of scale.

0:53 Hey folks, Jeff Berman here. I am thrilled to share some of the new names who will be joining us at this year's Masters of Scale summit. This may be our biggest stage yet. Reed Hastings, Meredith Whitaker, Van Jones, Amjad Masad, and more. Will be there with us October 20th through 22nd in San Francisco. If you're building something great, or you want to build something great, We want you there with us too.

1:20 Join us at masters of scale dot com slash apply twenty six. That's mastersofscom slash apply. Twenty six. What kind of world are we gonna have? The type of world that I would like to have is a world where People really are bettering themselves. That is one way in which they find meaning in life. Will it be the case that in 10 years we can get

1:46 A significant fraction of humanity's population, billions of people. Actually learning something. As opposed to either doom scrolling or just giving up on life because AI can do everything. We got to more than a hundred million active users, but that's just Too small of a fraction of humanity.

2:10 That's Luis von Ahn, CEO and co-founder of Duolingo, the language learning app that cost a billion dollars in annual revenue last year. Duolingo is known on social media for its slightly unhinged owl mascot, but it was Luis himself who went viral last year over an internal AI memo. That sparked intense public controversy. Today we talk through his tough lessons from that experience and what he's learned about the limitations of AI since that memo. We also talk about why he's made a strategic shift in 2026, emphasizing user growth over revenue growth. even though 90% of Duolingo's users are currently non-paying. Plus, in the spirit of Duo the Owl, we talk about fun.

2:56 As a learning strategy, ведеs, chess. or new business idea so Let's get to it. I'm Bob Safian, and this is Rapid Response. I'm Bob Safian. I'm here with Luis Von, CEO and co-founder of Duolingo. Luis, great to have you with us. I'm very happy to be here. Thank you for having me.

3:23 You first hit my radar when I was editor of Fast Company. You helped invent capture and then recapture the the security word puzzles that became kind of ubiquitous across the web. And then you launched Duolingo, which is this sort of playful but pragmatic language learning system that really took off and now it's crossed a billion dollars in annual revenue last year. Duolingo has always been Unexpected? Certainly s your social media campaigns have gotten a lot of attention. H how important is the unexpected and playful to what Duolingo does. I mean ultimately's pretty

4:02 Pretty important I mean what we wanna do is we wanna teach people and we wanna literally teach billions of people. Most people will tell you they want to learn something. But what people say and what people do happens to be very different. Certainly trying to be a a fun brand, an unhinged brand. uh that gets people to learn and

4:21 So, you know, our belief about teaching is that It's really important that we teach you something Well. But it's equally or more important that we keep you entertained while you're learning. Because otherwise.

4:34 You know, most people just Won't learn all that much. And I I mean a lot of your users are adult. So it's it's not just kids that you're trying to get to be playful, and and it's just important for uh for those of us who are trying to be uh lifelong learners. Yeah, a lot of our users are adults. Basically our users.

4:52 Um the whole range from age they need to know how to read. So I you know, age call it six seven to like a hundred. When we started Duolingo, there were people that would say, No, no, no, no, serious learners are not gonna want this. Бикас пікос і хас тобі серіос, і цера. That making something less fun.

5:12 Helps. Whenever we make it more fun, even the people who call themselves serious learners end up doing more. Mm and i it the decisions you make about Duolingo like Mm. Is there ever a distinction between engagement and learning?

5:27 There are You know, some parts of the Tech world where there's sort of been blowback about. Emphasizing engagement too much. They mostly go hand in hand because the the the most important thing is that people spend the time learning.

5:41 If you wanna learn Spanish and you're an English speaker. You kinda have to put in about five hundred hours. Like that's That's basically it. Uh it's kinda similar with exercising. I mean, whether you're doing the elliptical or the treadmill, it probably matters a little bit.

5:54 But what matters the most is that you're actually doing it a certain amount of time every day. When people are learning on dualingual, the The setting is a pretty is it's a little different than a classroom. In a classroom I think you can frustrate quote unquote users, in this case the the learner a little more.

6:12 Because they're you know, they're they kinda have to be in the classroom. It's not so easy for them to get up and walk away. Yeah. Right. We are we're always one click away from Instagram or TikTok or whatever whatever you want. So The level of frustration we can gift to users has to be lower than what a classroom does. And that probably makes it so that

6:33 We probably are teaching a little less. But We just wanna have a lot more time and so in the end it is uh actually ends up being good for learning. Mm. I mean r right now a lot of the um the learning that business people are being forced to do is is technological is is about AI. I I know that that's not

6:52 been the focus of of what the learning is on Duolingo, but are there are there things about the way we should uh be approaching learning about this new technology that y y you would take away from uh from what you do with Duolingo. It it certainly isn't often framed to us as being fun. I think the most important thing I would say for learning anything. It doesn't have to be fun.

7:16 It just has to be that it keeps people motivated. There are multiple ways to keep people motivated. We've dueling we've chosen mainly fun. That's the main thing we've chosen, but it you don't have to do that. I mean for example Seeing results keeps people motivated. In the case of learning AI, I I would say

7:34 That's probably a better motivator. Um of the form like I'm gonna learn AI, but the first thing that I'm gonna do is Make myself a dashboard. Or like a mini dashboard or something. But I I think it just if if you find the right motivation, um, you know, that helps a lot. Mm.

7:50 I mean I I I I have to ask you, because last year at this time you sent out this all hands email about AI that that rattled things a bit, right? Uh No new hires unless teams showed that AI couldn't do the job and existing plays is assessed on their AI use. And it it really sparked this this blowback on social and the stock price. You're not unfamiliar with taking risks. W was this a a bigger risk? Then you realized at the time?

8:18 Uh absolutely. I did not think this was going to be controversial because internally, inside Duolingo, this was not controversial. We started as a technology company and you know, I used to be a computer science professor that actually taught the AI class at Carnegie Mill University. So we've we've always used AI in as much as we can. So internally this was not. Controversial at all. Externally. I think um

8:43 I was not very clear. And given how I wrote it. And without giving it more context, I opened it up to people thinking that what I was trying to do was to fire a lot of our employees. Uh, but that was not never the intention. I mean, we've never done a layoff. We still have never done a layoff. In fact, last year when I sent that memo, we increased our number of employees, not decreased our number of employees. There was that misunderstanding because I think there's there's a lot of

9:08 fear that AI is gonna substitute jobs, et cetera, et cetera. The way I see it here is Our employees are just way more productive if they use AI. Um and so I actually want to hire more people because they can do more. It also seemed a little bit like um like you were sort of forcing people. You weren't making it playful to learn how to UI A AI. You were sort of like it was almost like a bludgeon, right? Which uh which I guess wasn't the intent necessarily either, but

9:40 But I do think it's something a lot of folks are struggling with. Like how do you get the people on your team who are more resistant to this new technology to to get on board? Yeah, I mean the good news here is at Dualing we just we hire a lot of people who are Pretty young. I mean, we have a lot of people who are straight out of college. And we have just not found a lot of resistance here. Internally we have this thing we call the golden rule of AI usage, which is

10:10 You have to use AI for the benefit of our learners. Like everything we do with AI should be for the benefit of our learners. So for example. Uh if it helps you be more uh Efficient uh

10:21 Putting out more features, that's for the benefit of our learners. If we put out a feature that helps our learners learn better because they're now, for example, interacting with an AI to practice conversation, that's for the benefit of our learners. Sometimes that when we use AI, we're able to save costs. But That is not the goal of our usage of AI.

10:38 That is an okay thing, but it is not the primary goal. It has to be about helping our learners, versus we're gonna use AI to save ten milyen bucks. Uh that's just not all that motivating. And I and I just wanna make make sure everyone because there have been some reports about you kind of walking back some of the things you said in in that memo, you're clearly still a believer in AI, right? There's no doubt from you that sort of this is the direction the business has to has to continue to go. I'm a big believer in a a in AI, but that comes with some

11:10 You know, it definitely comes with asterisk and learnings. Uh I mean for example. My initial memo said that we were going to uh Um Evaluate every employee on their usage of AI.

11:24 I don't think that was right. Many people came to me and said Look. For the job that I'm doing. I'm finding that I'm just using AI for AI's sake because you're gonna evaluate me on that, but it's not because I actually think that for this particular thing we can do it better. And I think ultimately, you know, in the case of performance reviews, what you should evaluate is

11:46 you know, how much that particular employee is contributing to the company. It turns out for most of our employees, using AI helps them contribute more to the company. That is true. However There may be cases, projects or particular roles where it may just not help all that much. And so saying blanket statement, we're gonna evaluate you on the usage of AI.

12:08 With Uh not needed. And so we've removed that. The other important thing that I think is important to mention, uh, when it comes to AI is that We're trying to use AI as much as possible.

12:19 But We really don't want to decrease quality. For some things AI's quite ready to do high quality stuff. For some things it's just not. And so we're not going to decrease quality.

12:32 Just for the sake of using AI. So where are you seeing AI not able to del deliver that kind of quality? In a lot of places Um For example, we hire a lot of artists and designers.

12:46 Uh and our app is very, you know, uh high craft on the sign, etc. We're just not seeing AI get to the level of creativity. Or the level of polish that you know, our top people have by any means. The other place where I think it's just not

13:03 The highest quality The one of the biggest problems I think with AI is that it demos really well. So what do I mean by that? It's just like look, it can write a story. And if you see one story, probably wrote a really good story, like the one story they showed you and I'm like, My God, it wrote a story. But when you go and you know, in our case, you we may need to write One thousand different stories.

13:23 uh for you know for people to to learn a language, then you'll find that like I don't know, twenty percent of the things were Just pure slot. Whenever we scale a lot. uh with things with AI, we have to really be careful that slot doesn't get through. And if the quality is just not high enough, even though AI is really nice in that it can do it pretty fast,

13:42 We just don't don't go for it. Yeah, and I mean I've heard some of that with with coding too. Like it can be terrific in in writing code and then it can produce stuff that like, well, that just doesn't work and y you need to Dig into that to test all those uses. Yeah. For the longest of times. If you would

14:00 Read Twitter. The vibe there was you probably should just fire all your engineers. Біз АІ конкодвані хума. І что на тру. Internally you would see that your engineers were all using AI, but somehow

14:17 You hadn't seen us speed up an engineering progress. Like the number of features we put out every month or whatever was just not increasing. And when you went to see why that was, you would see that oh, it turns out In some cases it does something amazing, and in the other cases, just like you said. It kinda doesn't work, but even worse.

14:36 It's so hard to figure out why it didn't work. That you spend about as much time as you saved on the cases that it worked. Figuring out why the other thing didn't work. It's getting a little better because coding agents are getting better. You can find a project with one engineer that maybe was ten times faster, but Th overall, this is just not yet true. Yeah.

14:57 In twenty twenty six, you've sort of signaled a uh a strategic shift for Duolingo prioritizing user growth over profitability, which seems like another risk, especially for for investors. So I uh why make this move and why why why now? I decided that This year. we would really spend it trying to work much more on user growth rather than revenue growth. And the reason for that is

15:26 There's a number of reasons, but probably the biggest one. Yes. I really think that over the next few years. The way most people learn is going to change. I mean, AI is just going to allow us to teach significantly better than we have in the past. And we're already seeing it, our product today, in terms of teaching. Is way better than it was

15:45 Four years ago. We do studies to see how well our product is teaching. It's just getting better every year. So I think there's a huge opportunity in front of us. And because of that, whenever you see that there's a huge opportunity in front of you, I think what you want to do is just capture as much of the market as possible.

16:02 The other thing that happened was that Are user growth? Particularly Towards the end of twenty twenty five, did slow. It doesn't mean that we're not growing. We're still growing. We're still growing exponentially, but it did slow. And I you know, I sat here and I thought Wow.

16:16 At a time when we should be Actually accelerating user growth. It seems to be slightly decelerating. I really don't like that, so we're gonna make that change. Now of course it's Whenever you go out to the stock market and you say this year we are going to

16:32 Lower our estimates for revenue. Uh the stock market No likey. No We knew this was going to happen. We knew but I I really do believe That this is the the right

16:46 Thing to do. For the long term of the business. And and just so I'm I'm sure that I and our and our listeners understand your your business model, most of your users are free users already supported by ads that come after a lesson. Most of your revenue is from premium users or premium fees. And so user growth means

17:07 More emphasis on the free level. Yeah, that's the main thing that helps us grow users. About ninety percent of our monthly active users are not paying us. And about ten percent pay us. But

17:20 The amount of money we make is flipped. So about ninety percent of the amount of money we make comes from the ten percent that pay us. So most of the money we make comes from from subscriptions. Increasing the fraction of People who pay. It's actually not that hard for us.

17:36 And we can do it at any point in time. If at the end of a lesson I give you one ad Versus if I give you two ads. That makes a huge difference in the number of people that pay. So it's not just that you get that you get extra revenue from having two ads, it's that for me as a user, I'm like, I don't wanna listen to the ads or wh watch the ads. And you and subscribing. Yes, but the thing is again, we're trying to be a a long term business and it does come at a cost.

18:01 Because maybe two ads i is not that annoying. But at some point, you know, if you're playing Seventeen adds at the end of a lesson. Uh people are probably just gonna leave. Particularly for your users who can't afford the premium, right? You're kind of

18:15 You're kind of driving them away. Correct. So that's why we don't pull that lever, but it is an easy lever to pull. And again, it's not just ads. You can do a lot of things. Almost every freemium business eventually piles on ads, Netflix did it, YouTube did it. The question isn't whether Duolingo will pull that lever in 2026 as much as how big it can grow before it has to, and what that says about the future of education. We'll talk about that more after the break. Stay with us.

18:50 Hey listeners, Bob here. If you listen to Rapid Response on Masters of Scale, you may be missing half the show because every Friday we release a second Rapid Response exclusively in the Rapid Response feed. The guests and topics are just as compelling and timely. From Ford CEO to NASA's administrator to the lessons from The Devil Wears Prada. It takes about 10 seconds to find. Just search Rapid Response wherever you listen to podcasts and hit follow to make sure you never miss an episode. I hope to see you there. Humans will never be more intelligent than AI. Those were great at AI and those that went out of business because they weren't. How do we build a future?

19:38 That is human centered. I'm Rana El Calyubi. And on my podcast Pioneers of AI, we answer that question and so many more. As an AI scientist, entrepreneur, and investor, I know what it takes to build AI that works for everyone. Every week, I sit down with the pioneers shaping our future. And we take you behind the scenes of the AI that's transforming our lives.

20:05 Find pioneers of AI wherever you tune in. Before the break, Duolingo's Luis Van Aan shared his missteps and learnings about AI and why Duolingo is deemphasizing revenue growth in 2026. Now we dig into the limits of unhinged marketing, what happens when stakeholders don't speak the same language, as Luis puts it. And what not to do if you want to get hired at Duolingo. Let's jump back in. As you're gonna try to like open up this top of the funnel, really, bring bring more people in. Does that mean like

20:45 Like more of the social media stunts that Duolingo is is kind of known for. You know, I I mean I'm remembering sort of staging the death of your mascot, the Green Owl Duo last year. But I also I saw something your chief marketing officer said that's social output that was Eighty percent unhinged and twenty percent wholesome that now is gonna become more balanced. Хісториклі ар со мідіа газ би Um

21:11 An incredible growth level. Our mascot. Yes. has become you know Really popular.

21:19 Um and to the point where we don't have much control over him. Like you said, he staged his own death. But We really do think that it is important to also. Convey. some things in our marketing that I think historically we just have not because it was actually quite effective. If you go look, it's it's about as effective as a classroom. That's just not something we have really said a lot in our marketing because we spent a lot of our marketing doing stunts, unhinged things, having the outworking, which has been

21:48 Successful in its own right. We need to thread the needle here. To try to continue being fun. But also move a little more towards having a message that is like, hey, you know what, Duolingo actually really works.

22:01 In this user expansion, like how much of the plan is around expanding what you teach. I mean, I've always thought of Duolingo as for languages, but you know, courses on math and music and chess. Like why did you add those areas and sort of is your vision of Duolingo's future to be more expansive? Yeah, we may not teach everything, but we wanna teach the subjects that are meaningful to you that also take a long time to learn. Math is an example, languages is an example, music is an example. Chess.

22:35 Even though it's our newest subject has gotten the most traction outside of languages. Interesting. Languages by far are still the biggest. And languages the other thing that may sometimes may not understand is Just how big language learning is in the world. There's about two billion people in the world learning English. I don't know of a subject. Including math.

22:54 That has as many people learning it. Math. There's about a billion people learning math. So language is the largest thing. Um Chess is the second largest for now for us, but I believe that over time math is gonna overtake chess. That's that's my hope.

23:08 At the moment no numbers support my hope. But that's my hope. As you're talking, I'm I'm thinking about another digital learning platform, Khan Academy, that rose up alongside Duolingo. I've had Sal Khan, the founder and CEO on the show, also an early AI believer, um recently teamed up with Ted to offer a degree in AI skills. Will Duolingo become more like Con Academy in subject range? I mean, it's almost like the two businesses are converging. Did do you and Sal ever talk about that?

23:40 We do. I I I know Sal. Uh we do. I don't know if we're gonna exactly converge who we're trying to reach in many cases is pretty different. Sal has this incredible videos that explain subjects incredibly well. Our approach is to be more like a mobile game. Um

24:00 So you know, the i in some sense they're quite complimentary. Khan Academy has been mostly geared towards kind of schools and school usage, et cetera. Whereas where You know, so many schools do use us, but we're you know, w really for the whole population I mean, Khan Academy is also a nonprofit.

24:17 Duolingo is not your publicly traded company worth more than four and a half billion dollars. You went public in twenty twenty one. How has your role as CEO changed since taking Duolingo public? I would say the first f couple of years it didn't change all that much, but I think it has changed more recently, especially When I I had to s say that we were gonna have a different strategy.

24:43 Now as a publicly traded company, I have to convince the board, the employees, and Public market investors. Uh who I don't have a direct Conversation line two.

24:55 People on the internet. And there's our users. Probably the most important constitu constituency But the source employees. Also our investors and they're not

25:05 really aligned. Maybe in the long term they all are. But certainly in the short term. I mean, users what they want is to get as you know, the the entirety of the learning as good as possible and as free as possible. Right. And again, um it is a caricature. But investors you just want us to charge. Every user as much as we can.

25:23 And employees are somewhere in the middle. Employees probably want us to give uh Duolingo entirely for free, but they also want to make a lot of money. They say that you get the investors you deserve, right? I mean, if you if you talk too much about delivering uh results today, you get a lot of short term investors. And there's that issue of, you know, are you in business to make as much money as possible? versus having uh a a sort of more social mission. Is teaching people a conduit to making money, or is making money a conduit to teaching people? It's a really good question, and I may be

25:58 An incredible optimist, and it may be that I'm wrong. But it is my belief that if you look in the long term These two are the same. But not in the short term. I agree. In the short term You know, the it really it really matters.

26:14 which one you're you're prioritizing, but I think in the long term. What's gonna matter the most and what's gonna also make the most money Is uh A company That

26:26 really is is putting The education of the users first. I think that will be the larger company in the long term. Uh now Can I prove that to you? I can't, but that is you know, that is my optimistic thinking that in the long term these two are actually aligned.

26:42 Hm. How much of of the decision to sort of broaden beyond languages is prompted by by new AI tools that enable instant live translation across languages. Like do you worry that people won't be as motivated to learn a language as time goes by? No, I I d I don't worry about that. We have two big buckets of users.

27:08 The first is People who are just hobbyists. And You know, if they're learning language as a hobby Whether a computer can do it or not doesn't seem to matter to them. It's similar to by the way, we also teach chess.

27:20 You know, a computer beat the world just champion in something like nineteen ninety seven. It turns out humans are just not particularly good at playing chess when it comes to computers. We're just not. But This is still the case. There's millions of people learning chess. Because it's a hobby.

27:33 Similarly, languages you know, a a big group of our users is a hobby. And I don't think they're gonna change. And then the other huge group of our users are people who are learning English. either for a a job that they want to get Or to Come study here in the US or in the UK, et cetera.

27:50 And in that case, they actually need to learn it themselves. That's an even qu bigger question, which is some people have said, Well with AI, why do you need to learn anything? I mean we would be uh It would be A chaotic society if nobody learned anything. I mean, people still need to learn stuff.

28:07 So I I you know I I I I do think that what people are gonna learn is probably going to change some. And I don't know exactly how, but uh again, people still do need to learn stuff, including math. They need to learn math just so that they can learn, you know, logical thinking. I mean, I You've been a professor teaching in the classroom. You've now run Duolingo for for more than a decade.

28:26 Uh what do you think people misunderstand about learning? About teaching, like what makes it hard. What turns people off, what turns people on. In my experience The hardest thing

28:40 It's keeping people motivated. There's been a technology. That teaches any subject. really well that has existed for a very long time, which is called a book.

28:50 A book you can actually learn almost any subject you want. I mean you can you can learn quantum physics by just reading books. You can do that. But it just turns out the vast majority of the human population does not have the stamina to do that. That is just not a thing that they would do. So I think the hard thing is getting people's attention, and I do think it's getting harder and harder. Because people's attention spans are getting shorter and shorter.

29:12 I think that is the the hardest thing about about teaching somebody, keeping keeping them engaged. And different people can also learn better in different ways, right? I mean, is that is that part of the challenge It is definitely the case that different people do learn in different ways. And we do we do personalize whenever you start a lesson in Dualing, it is personalized to you. Um

29:32 But I will say I think that is Exaggerated. It is rare that we find changes that you know only apply for this group of people versus that group of people.

29:43 You hear these things that were like Oh no no, but people in China are very different. They use apps very differently, or people in whatever Uh friends are very different, etc. It's very rare that you find a feature that works in one country but not in another one. It's very rare. It it typically most people are pretty similar.

30:02 For example, if you have a progress bar that you're filling up and it's three quarters filled. Pretty much every human will want to fill it to a hundred percent. That is that is just the universal truth. And there's just th I think there are more universal truths than there are kind of uh very specific differences between people.

30:19 You are not shy about unconventional ways of of approaching things. I was I was looking at this um viral clip of you telling a story that you paid a taxi driver to report on how job candidates treat them during the ride to interviews, which is a a particularly tricky way to learn. Uh A little clarification, by the way, I don't know how this got it wasn't a taxi driver. We would send them like a Limo driver. That works for us. And by the way, it's not like I sat there paying this person uh what w all that happened is that at the end of an interview we would've go asked just this person, you know, did they treat you nicely? Um and uh it turned out there was signaling that in most cases they were treated nicely, but there were some cases, particularly one case, where we were about to make an offer.

31:01 And the person said, No, they really were not very nice. Um and we didn't make an offer. When you're trying to hire somebody's usually because there's a hole in the organization that you're trying to fill. We have this internal rule that We'd rather have a hole than an a hole.

31:15 That is what gave rise to that. So What's at stake for for Duolingo right now? What we're trying to achieve. Especially because of AI.

31:26 Is you know, what kind of world are we gonna have? The type of world that I would like to have is a world where People really are bettering themselves. That is one way in which they find meaning in life, et cetera. Will it be the case that in ten years? We can get. A significant fraction of humanity's population, billions of people.

31:44 Actually learning something. As opposed to either doom scrolling or just giving up on life because AI can do everything. What I'm hoping we can do is is just add meaning to Yeah. that many people at the moment we have s you know, north of a hundred million active users, which It's nice. That's nice. Good job Luis. We

32:04 We got you know, we got to more than a hundred million active users, but that's just Too small of a fraction of humanity. For me to claim victory. I'm hoping that We'll have billions of people learning meaningful things in ten years.

32:17 You're you're not easy on yourself. I mean a a hundred million people that that that does deserve a a pat on the back. Not enough. Uh yeah, I mean look it's good. If you look at education, it's unheard of to have something that is teaching Uh.

32:31 you know, more than a hundred million people willingly. That's kind of unheard of, but still I just think that there should be a world. Where Really People are spending

32:41 some amount of their time on on the screen. Uh doing more meaningful things than, you know, scrolling forever. Uh I it's not like I'm against social media companies. I'm not necessarily against them or anything, but like it seems kinda not all that meaningful to me. Well, Louis, this was great. Thanks for coming on and talking. Thank you excellent questions.

33:02 Luis is an optimist, as he says. His bet is that in the long run, AI will deepen our love of learning, not replace it. I hope he's right. So far the story is different. Layoffs, hiring freezes, universities pleading with students to write one sentence on their own. But as Luise argues about company values too, the long term is what matters. We have the agency to impact how the story unfolds on AI from here. If we put in the work, the 500 hours or whatever it takes, AI may be fluent in almost any language. But it can't replace the learning we get from doing something the hard way. I'm Bob Safian.

33:44 Thanks for listening. Rapid response is a wait what original. I'm Bob Safian, our executive producer is Eve Tro. Ar senior producer is Alex Maris. And our associate producer is Masha Makutonina.

34:08 Mixing and Mastering by Aaron Bastanelli and Brian Pugh. Our theme music is by Ryan Holiday. Our head of podcasts is Lital Malad. For more, visit rapidresponsshow.com.