Back-to-school with Big Tech

This is Fresh Air. I'm Terry Gross. It's back-to-school season, and in many classrooms, digital devices have replaced or supplemented books. Tech giants like Google and Microsoft have tried to get schools to buy their products and have offered tech training to teachers and students in subjects like coding, the use of AI, online safety, and basic computer tools. So what are the tech giants getting in return for this investment, and how much are they influencing what and how children and teens are taught in schools? My guest, Natasha Singer, is a New York Times reporter covering the intersection of technology, business, and education. She's written a new book titled Coding Kids, Big Tech's Battle to Remake Public Schools. She writes that in interviewing more than 100 education and tech industry figures for her book and reviewing hundreds of school, government, and corporate documents, quote, it became clear that a handful of giants have used their money, their might, and their mammoth platforms to influence nearly every step of the education supply chain, unquote. Singer was a member of a Times reporting team whose coverage of tech industry privacy violations won a George Polk Award in 2019 and was a finalist for a Pulitzer Prize in national reporting. Natasha Singer, welcome to Fresh Air. You describe how tech companies have provided, things for schools. Are they donating them, or are the schools buying them? So, public schools in the United States spend billions of dollars a year buying technology for the classroom, including software, math apps, but also computers for every kid. And so those are purchases, and companies like Google and Microsoft have education divisions that sell technology to schools. Then, these companies provide software services, like if you're going to buy a Chromebook for every kid, then you need a system to track where the Chromebooks are, and if you're going to give tests, you have to lock down the computer, so that might cost like $12 per kid per year. So there are services that schools pay for. And then on top of that, the companies provide a lot of services that they don't charge schools for, including if you're a client that uses the Google Apps for your schools, then, you know, you get Gemini, Google's chatbot, and, you know, the teacher training is often funded by tech companies or even led by tech companies. And there are other things, like there are digital lessons for kids on how to be proficient with spreadsheets or how to, you know, make PowerPoints or, you know, how to analyze data or how to code that the companies have funded or have made themselves, and those are often free. How do the companies compete with each other to convince the teachers? How do they convince schools or, you know, school boards to buy their products as opposed to their competitors' products? So we already have an entrenched system where school districts refer to themselves as, we're a Google district, or we're a Microsoft district, or we're an Apple district. And so imagine that if you are a wealthy suburban district that can afford to buy MacBooks for all the middle schoolers and high schoolers and iPads for all the elementary schoolers, and Apple comes with a new thing, you know, it's going to fit with the system, you're using, and you're going to want to use maybe the Apple coding lessons. And if you are a huge district like Chicago Public Schools or Miami-Dade County Public Schools, and you are a Google district and your students all have Chromebooks or you're a Microsoft district, and Google comes and says, like, you can just start using Gemini with your students because you already have all the software and apps that Gemini will be embedded in. And so part of it is the, like, incumbent advantage, is that you're going to be able to use all of those things that are in school districts. And then when you see newcomers trying to break into this market, so OpenAI with its ChatGPT chatbot is one of the newcomers. And education is a crucial market for tech companies because if you want your product to scale and have permanence and have lifelong customers, the best thing to do is either target elementary schoolers and they're going to use your app for life, or you target teachers who are going to introduce them to kids. And so if you're going to use your app for life, So we see that OpenAI made deals with some large school districts and some small districts, and they agreed to provide ChatGPT-EDU, which is a form of ChatGPT that doesn't use teacher and student data to train OpenAI's model and has some increased safeguards compared to the consumer product. OpenAI did deals with schools to train teachers on these tools and to allow teachers to use ChatGPT-EDU for free for, you know, initial period of a year or so. And then, you know, the school will see how it works. Are teachers more efficient if they use a chatbot to help make their lesson plans or their quizzes? You know, are the teachers doing interesting things with a chatbot with kids? And so now, in the last few weeks, Anthropic, which makes the Claude chatbot, announced it had made a program for teachers with online lessons that teachers can sign up for to become proficient at Claude. And so, you can see that the newcomers are trying to break into the space where companies like Google and Microsoft and Apple already have heavy footprints. Do the companies make suggestions of what to teach with their tools? Or what kind of, like, computer literacy, whether it's AI or coding or just digital knowledge, do they make suggestions of what to teach and how to teach it? They do. And I think that there's a very interesting, very interesting tension here between the desire of schools and teachers and parents to want students and kids to learn the latest technology. We don't want kids to be left behind. We want them to be able to operate in this machine-laden world we're living in. But at the same time, when you have tech companies shaping the curriculum and training the teachers, they're going to teach what is good for them, but also what their worldview is. And so, you know, I've talked to these companies, and they're like, yes, AI literacy. Every kid must have AI literacy. And sometimes when I say, what does AI literacy mean? It means learning how to use their particular chatbot. And that's a very, very narrow way of thinking about AI. And so, you know, there are also moves to teach much bigger themes about AI to students, including, like, where do these generative AI tools come from? Who makes them? Who made them? You know, who profits from them? Because we have a handful of companies that have a massive amount of data and have scraped the internet and books and newspaper archives to build their language models that are then able to make these powerful tools, right? And so, and then the question is, how do you use these tools? And what about the biases of these tools? And what about the fact that, like, if the tech industries are pushing it, they're pushing their worldview, which is machine-centered. And if you read the, the marketing to teachers of these companies, which I do, they're all saying something really similar. Like, Microsoft says something like, teachers, prepare your kids for an AI-driven future, right? And the Google one says something like, teachers, use our tools and help prepare students for an AI-first future. And the tech companies want kids to be prepared for the world the tech companies are building. But we're not saying to kids, do you want an AI-driven future? Do you want an AI-first future? Like, we're skipping over all the hard questions and the tools are just helping kids and teachers into being the AI-driven world that the tech companies want and that benefit the tech companies. It's also a world that may leave students without jobs because a lot of the jobs will have been taken over by AI. And one example of that that you write about is that the tech companies were really pushing coding. Like, you have to teach students how to code. They'll have great jobs. And we need more tech engineers who know how to do this. We need more coders. And, of course, coding is becoming really obsolete because AI is taking over the job of coders. The story of how big tech companies like Google and Microsoft and the nonprofit groups they funded catalyzed computer science in schools is extraordinary. And I don't think we've seen anything like it in the history of public education in the United States. And that's part of the backbone narrative of my book. Because you see around 2011, 2012, tech executives start to say publicly, we have this really big challenge. We're growing really fast. We're innovating. We can't find enough talented software engineers to hire. And the blame lies with the public school system that's not teaching enough kids to code. And it becomes a kind of national emergency, right? There's this urgency that's created to get computer science in schools. And the argument, to parents and schools and kids and teachers, is really winning. It's that if we just teach every kid to code, every kid computer science, they will be able to get these high-paying, high-powered jobs. And they'll be able to change the world. And also it will help the tech companies and the economy. And so schools really are enthusiastic about this. And you see this prominent nonprofit group called Code.org that has been funded by Google and Microsoft and others really put coding on the map in families. And schools begin to teach this. And more and more kids begin to take it, and you see an increase in the number of kids majoring in computer science in universities. And then a decade later, generative AI comes along. ChatGPT and Claude are being used by software engineers and other tools like Replit to generate computer code. And the tech industry's interest in funding computer science in schools wanes. And the tech industry pivots to AI, spending hundreds of billions to build these data centers. And the tech industry is laying off tens of thousands of people. And now you get students who are graduating from college with computer science degrees who are struggling to get jobs. And so I do think what is our responsibility as a country who told this generation that if they just coded, they'd be magicked into money and power, and especially the tech companies who drove that narrative? What's our responsibility to those kids? So how did – Getting coding taught in schools pay off or not pay off for the big tech companies? I think the main payoff for companies was goodwill, right? Here are these companies coming into schools and fostering a subject that is really popular, that is advertised to help kids become not consumers of technology but creators of technology, right? The argument was code or be coded. If you don't know how to code, these apps are going to enact themselves on you and steer your life. And instead, you could be the one making the products. I think one of the things that tech companies got was a lot of goodwill for the work they were doing to get computer science into schools. I think we see that large tech companies in the past and even now tend to hire from elite universities, elite state schools, Berkeley, UT Austin, when they're looking for kids in computer science. If you are in a small state college, it's much harder even if you've been coding for a decade to get hired at these companies. So I don't know the extent to which the companies directly benefited in terms of hiring, but it certainly widened the hiring pool for other industries like banking and pharmaceutical companies who also needed technologists. And, you know, Walmart, you know, huge companies, everybody needed technologists. Was COVID a big turning point for tech giants like Apple and Google and Microsoft in terms of getting their products into schools? Because all the students were learning remotely. Yes. The big wave of getting computers in schools had already started. You know, like in 2013, companies were shipping, you know, nearly 7 million devices to schools. But it really grew. You know, in the pandemic when everybody was doing remote learning. And it also put everything, as you know, online. And so parents were sitting at home and they were watching their kids, including their younger kids on iPads and their older kids on Chromebooks. And the teacher is using Zoom to teach them. And then they need to type things into the computer and do all these online lessons. And it, you know, became a tech-centered kind of education by necessity. So what are some of the ways that the different companies who make different chatbots are trying to get their chatbots into the schools now? Different companies are trying different avenues to get their AI chatbots into schools. And different school districts are doing different approaches to AI in schools. And I visited Miami-Dade County Public Schools, which is the third largest U.S. school district. Because they. They had decided to give Gemini, the Google chatbot, to every high school student. That's more than 100,000 students. And it was super interesting because they were very careful in vetting the chatbots and figuring out whose brand of chatbot they want to use. And they had very rigorous training of teachers on privacy and safety and other things. They tried to control all the things they could control. And when I went to visit, which was early on in this process, they were experimenting with different ways to use chatbots, including this English class I went to, where they were studying the idea and writing about the idea of free will versus destiny, like in Oedipus Rex. And the teacher asked students to write just a short essay in which they had to argue whether humans had free will or whether they, you know, were ruled by destiny. And the teacher put a rubric into Gemini to grade students on, did they have a thesis sentence? Was it strong? Did they have strong evidence to back up their thesis? And did they have a concluding sentence? So the students, you know, it was short, wrote their essays. Gemini gave them instant feedback. Some of it was, yes, you've got all the stuff. You've got all the points and your thing is logical and great. And sometimes Gemini said, well, actually, your evidence does not back up your thesis sentence or actually only have one piece of evidence and you need three. And so Gemini was not giving kids feedback on their writing, whether they were literate. Whether their literary, whether their word choice was wrong, it was just evaluating the content based on what the teacher had asked it to do. And, you know, teachers, if you've got five classes of 30 kids and you're going to have to grade 150 pieces, even if they're only a couple paragraphs long, it's a whole bunch of work. And so when I spoke to the teacher, she said, I'm going to read every single one the same way I always read every single one. What this does is in class really quickly gives kids feedback on. What they're missing and what they might strengthen. And they can change it and then they're going to submit it to me. And I'm going to see what the chatbot told them and whether the chatbot was helpful. So I think they're really interesting ways that we're going to see AI being used in schools. Now, you might not want that. You might be like, I don't want my kid talking to a chatbot. I don't want my teacher giving kids feedback through a chatbot. People have all kinds of feelings about this. But I think we're at the beginning of this. And we don't know how it's going to pan out or what uses are going to be good. Or what uses might be extremely problematic. We don't know yet. We're just at the start. You've spoken to a lot of parents and a lot of teachers. How would you describe their biggest concerns about the use of AI in the classrooms? Terry, I think there's a large thing going on in society. And then specific things that parents are worried about. So we're in this extraordinary moment where big tech companies are facing a reckoning. And AI is the catalyst. So we see. We see communities pushing back against AI data centers, for example. And we're seeing in New York and Los Angeles and elsewhere, parents' groups push for a moratorium on AI in schools. And I think part of that is just this feeling that tech companies have runaway power in schools. And that parents don't want their kids to be guinea pigs for AI. And they saw what happened with social media. And they want there to be safeguards and proof that AI is beneficial for learning. Before schools use AI tools with kids. Because it hasn't really been tested yet. It doesn't have a history. We just don't know. Microsoft has been funding studies. And there was a study of adults using chatbots. You know, white-collar workers. And it was a study with Microsoft and Carnegie Mellon. And it found that kind of offloading tasks like research and writing to bots hindered the critical thinking of the human. Because you're basically turning the human into a handmaiden of a chatbot. And the handmaiden is reduced to, like, checking the accuracy of what the chatbot produced. Or, you know, is this writing good? And then smoothing out the writing. And then there was another Microsoft-funded study with Cambridge University Press in the U.K. Which looked at students that were randomized to either use a chatbot or not when reading text passages. And the kids that were randomized to use a chatbot to summarize the reading. To help them did less well in reading comprehension and retaining the information than kids that took notes by hand on the text passages. Do you think you can recognize the difference between an AI-generated essay and one that a student wrote? And do teachers think that they can tell the difference? So, I think that there are some telltale signs of AI writing that teachers can spot. But students are clever. And the students that want to do this. First, get an AI to generate an essay. Then they run it through another AI and ask it to take out the signals that it's an AI. Whoa, really? Yeah, no. That's so clever. There are these tools called humanizers. Right? Oh, I didn't know about that. Really? Yeah, no. AI can insert mistakes and slang words and things to make it look kiddified. You know, so I think that teachers know they have to get ahead of this. By not giving kids writing to do at home. And, you know, you see some schools are going back to blue books and saying you're going to have to write this by hand in the classroom. But I also think that, like, teachers are trying to reframe the way they teach to get kids to think critically and to say, like, okay, you've written this. Like, tell me the thinking that went into this. Right? Like, how did you arrive here? And so, I think we're in a, I mean, I think we're also in this race where, like, some teachers are using AI to generate their lesson plans. Right? and assignments and their quizzes and then kids are using AI to do them, and then teachers are using AI to grade them. So we just have bots talking to bots, and is that what we want education to look like? Well, we need to take another break here, so let me reintroduce you. If you're just joining us, my guest is Natasha Singer. She's a reporter for the New York Times covering tech in schools, and she's the author of the new book Coding Kids, Big Tech's Battle to Remake Public Schools. We'll be right back after a short break. I'm Terry Gross, and this is Fresh Air. This is Fresh Air. I'm Terry Gross. Let's get back to the interview I recorded yesterday with Natasha Singer. She's a New York Times reporter covering tech in schools, and she's the author of the new book Coding Kids, Big Tech's Battle to Remake Public Schools. She writes that in interviewing more than 100 education and tech industry figures for her book, and reviewing hundreds of school, government, and corporate documents, quote, it became clear that a handful of giants have used their money, their might, and their mammoth platforms to influence nearly every step of the education supply chain. It seems like there was a turning point fairly recently in this thinking of a lot of school administrators, and in the American Federation of Teachers, the Teachers Union, about the use of AI, and you offer the AFT, the union, as an example. Can you talk about how the AFT and the American Federation of Teachers kind of shifted course on AI? Yes, and I've been talking with Randy Weingarten, who's the head of the American Federation of Teachers, for more than a year about this. So last year, in July 2025, the AFT announced that it had done a deal with Microsoft and OpenAI and Anthropic to help train teachers on AI. The AFT was going to start a National Academy of AI Instruction. And at the time, Randy Weingarten was the head of the American Federation of Teachers for more than a year about this. Randy Weingarten said that this was really important because teachers needed to learn how to use these tools, partly because kids were using them. So teachers needed to learn how to use them just to catch up, but also because it could be efficient. You could generate emails to parents. You could generate lesson plans. It could free teachers up to spend more time with kids. And second of all, because Weingarten said it would get teachers a seat at the table with these big tech companies and help the companies shape their products to serve schools and teachers and students. So that was last summer. And she got some flack from union members who felt she had cozied up to big tech. And she said she knew that, but she felt it was the right thing to do. And then a few months ago, she gave this talk in Washington, D.C. at the National Press Club in which she said she was wrong, that she had mistaken engaging technology that keeps kids hooked for something that was definitely going to affect them. And she said, well, we're going to have to do something about AI chatbots, particularly these ones called companion chatbots, which befriend you. And it's, you know, ersatz emotional support. And she didn't want kids befriending chatbots. But more than that, she was concerned about the effect that AI could have on young kids. And she said she thought elementary schools should not be using AI with kids. And that was a major change from, you know, just a year ago. What was the turning point? I think that Randy Warngarten is a very savvy politician. And the wind around AI changed in the United States. We see communities pushing back against these AI data centers that big tech companies want to build. We see this concern about these flock cameras that are going up that surveil people. We see pushback against Meta's glasses that have cameras in them that people are calling spy glasses, surveillance glasses. And then also there have been a few lawsuits, right? We see that parents are suing some of the AI companies saying that their child developed a relationship with a chatbot. And the child was feeling bad about themselves. Some of them, you know, are adolescents. And the chatbot essentially helped the child die by suicide. And they're blaming the companies, you know, for making defective products that don't have guardrails for kids. And so in the midst of our concern about social media, the fact that these similar companies concerns are beginning to bubble up for AI, I think, bothered Randy Warngarten. But also the notion that we don't know how these tools are going to work in schools, and whether they're going to hinder the basic critical thinking that kids need. So during the second Trump administration, he created a White House task force on AI education. And there was a meeting of that task force hosted by Melania Trump. And I want to play a clip. digitally-forward country, and they decided that they did want high school students to use AI because they did a poll and they found the majority of high schoolers in Estonia were already using like off-the-shelf commercial chat GPT and other chatbots. So they went to OpenAI and they said, we want you to make a product specifically for Estonian schools in Estonian and that is Socratic and that asks questions and doesn't answer them. So if we're reading Catcher in the Rye in 11th grade in Estonia and we are giving students chat GPT and the student comes to chat GPT and says, tell me the plot of Catcher in the Rye, the chatbot might say, well, what do you think Holden Caulfield's motivations were instead of writing the essay for them? And so Estonia has created a specialized school-based chatbot. They are trying it out right now. They are doing research with a university in Estonia and also with Stanford to look at usability, an uptake and eventually to look at how it changed learning. There's been some pushback. Some kids don't want to use AI. Some kids find it ridiculous. If you say to it, how can I build a birdhouse? It might say to you, you know, how do you think you can build a birdhouse, right? Like a chatbot that only asks you questions and is Socratic, you know, can't fit in all situations, but it's like, I've spoken to the Estonian minister of education and the president of Estonia and other folks there asking questions about it. And the minister of education in Estonia said to me, like, we are trying this in high schools. We're going to see if it works under no circumstances. Are we going to try this with kids under 13 and school should not be using off the shelf chatbots with kids period, right? They should be using specialized tools that are designed for schools. Let's go back to the beginning of this story. You trace the efforts of big tech to get their equipment in schools and to convince teachers like how to, to teach computer science. You trace it back to Apple. So tell us what you know about the very beginning. It's such an interesting story and it sets the paradigm for a lot of things we still see big tech companies do today. And in the early days of personal computers in the early 1980s, tech companies wanted to get their desktop computers into schools. And we have a tradition in the United States of powerful, charismatic inventors thinking that their technology is going to transform schools, right? It goes back to Thomas Edison. And Steve Jobs is one of those people. And Apple was still a fledgling company. And he wanted to get Apple's desktop computers into every school in the United States. And he figured, you know, Apple should give a computer to every school because many schools didn't really even know, like, what would you do with a student computer in those early days? And it was going to cost Apple millions of dollars to do that. And, you know, back then Apple was still a small company and they couldn't afford to do it. So Steve Jobs decided he was going to personally go to Washington, D.C. and lobby Congress for a tax break. And he calls the campaign Kids Can't Wait, right? It was really urgent. He felt this personally. He had fallen for computers at a young age and it was transformative. So he felt there was a creative imperative, but there was also a business imperative. Because companies also wanted to sell home computers. And if your kid comes home from school and says, Mom, you know, like, we're using this Apple computer in class. Can we buy one? It's also going to help not just turn kids into lifelong Apple customers, but also get parents to buy these computers for home use. So Congress does not pass the Apple bill to give a national tax break to computer makers, but California passes the Apple bill. And so computer manufacturers can now get a tax break for donating computers to their parents. And so they're going to have to give computers to school. And in 1982, Apple gives Apple to a computer to nearly every school in California, thousands of schools. And it also ushers in this new thing, which is called computer literacy. So how is that shaken out? Like, who has the biggest school market now? Do you know? Oh, well, in the United States, you know, Google came along. And in the early 2000s, universities and schools were paying a lot of money for legacy software systems like email. And Google offered this free package of Gmail and Docs and other things. And suddenly you saw entire universities overnight switching tens of thousands of students to Google. And then after major universities switched to the Google apps, you saw huge school districts like Chicago Public Schools switch to the Google apps. And then once Google really began to dominate the school market with software, they launched a new device called a Chromebook. And it was this sturdy laptop that ran on Google's operating system and also was internet connected. And that meant, like, all students' documents and everything would be up in the cloud. And schools began buying these things in bulk. And Google's argument at the time was, this is going to democratize access to education. These things are cheap. And they're sturdy. And every kid could use one. And you can buy one for everybody. And also, they're going to teach kids important career skills. And it was this altruistic message to schools that equated laptops, technology products, with equity and democracy. So it seems, from what you're saying, that it's part altruistic and part all about future profits. So the altruistic part is kind of clear. I mean, if you're offering conferences and offering, you know, instructional classes, that's, you know, going to be helpful. Can you talk more about where the profits come in? Well, first of all, United States public schools are spending upwards of $30 billion a year on hardware and software. So there is a huge market for technology in schools by itself. But the long-term profit comes from getting a student on your operating system, whether it's the Apple operating system or the Google Play system. And so, you know, there's a lot of money going into getting a student on your operating system, whether it's the Chrome operating system or the Microsoft operating system. And then getting them in your ecosystem and creating loyal customers and keeping them for life. Because if you're a kid who's going to be using a MacBook in school, you may be also likely to ask your parents to make your first phone an iPhone. And you might be an iPhone customer for life. And then you're going to buy Apple Music and Apple TV. And then you're going to buy, you know, Apple MacBooks for your whole family, right? You're like, or you're going to buy an iPhone. And then you're going to buy an iPad. You're going to be a Google person for life. And you're going to buy Nest cameras. And you're going to keep all your email on Gmail. It's not about the immediate capture of a student this week with one device. It's about the long-term providing training wheels in schools for your brand of technology. It's about getting kids in your ecosystem and trying to keep them there for life. Well, let me reintroduce you because we have to take a short break. My guest is Natasha Singer. She's a New York Times reporter. She's a New York Times reporter covering tech in schools. And she's the author of the new book, Coding Kids, Big Tech's Battle to Remake Public Schools. We'll be right back. This is Fresh Air. This is Fresh Air. Let's go back to my interview with Natasha Singer, a New York Times reporter covering tech in schools and the author of the new book, Coding Kids, Big Tech's Battle to Remake Public Schools. Meta, the company founded by Mark Zuckerberg, which also obviously it owns Facebook, but also Instagram. They had to pay, was it $17 billion in settlement after they were found to have knowingly designed a system on Instagram that would be addictive while saying to the public that it was not. Do you think that that lawsuit has added to the skepticism around AI and around the tech industry in schools? And I know that, you know, Instagram is not AI, but Meta is a really big company that we now know, like lied to the public about knowing that they had designed something that was intended to be addictive. So I think the settlement is this extraordinary societal and cultural moment because you have Meta, which is the biggest social media platform in terms of how the public thinks about social networks, right? And Meta agreed to settle this case with the attorneys general of 47 states. And I think it does play into this groundswell of parent concern over tech and AI in schools in several ways. The first is this notion that big tech is not AI. And the second is that big tech has runaway power and is making decisions for kids that can negatively affect them. And social media is the case study. And, you know, I did a lot of research in my job at the New York Times on these lawsuits by the attorneys general. And they had all these documents showing over the years employees at Meta writing to Mark Zuckerberg worried about their underage kids on our platform and parents are complaining that we're not getting them. And they're like, well, we're not getting getting them off. There are these beauty filters that we have that girls are using to transform their face, and it's actually causing anxiety about their looks. We have kids who are reporting that they're being bullied or that they are getting unwanted sexual advances, and the scale of that is really problematic. And every time, you know, people are making suggestions to Zuckerberg of hiring more safety engineers or changing features to be off by default. Either he or somebody who works for him, you see in these emails, says, no, we're not going to do that. And so some of the things that the companies are agreeing to do now, like turn off notifications in the middle of the night that could wake kids up, turn off notifications during the school day, turn off the beauty filters, put in age verification in a way that's actually rigorous to prevent kids under 13 from being on the platform, are things that inside Meta, people were asking for. And so I think that the news of the settlement is big. The proof is in the pudding, because Meta has also promised to make changes for years that parents feel did not go far enough. I think the way it relates to AI is also, though, in the way the system works, because the underlying mechanism of social media, which is that they work like slot machines, right? There's an endless speed. You just pull the lever, you get more videos. You pull the lever, you get more videos. You get more photos. You pull the lever, you see more posts, and you get more emoji hearts and things, right? It keeps you hooked with variable rewards. And when you use a chatbot and you're a kid, it's very similar. You're asking the chatbot something, you're getting an answer, it's human-like, it's interesting, it's charming, and then it says at the bottom, like, can I also help you do X, Y, and Z? Right? There's never an end. And so I think that parents have very similar worries about AI that they have had with social media. They do not trust tech companies to put in safeguards to protect kids from AI, and they do not want their kids to be guinea pigs for untested AI in schools. Do tech leaders talk to you, or do they see you as being too critical? Both, I think. For my book, I spoke to Brad Smith, who's the president of Microsoft, and he has been one of the leaders in the tech industry pushing for computer science in schools. And I also spoke to Maggie Johnson. She's the head of Google.org, who is his counterpart at Google in terms of, you know, supporting computer science in schools. And I'm grateful that tech executives talk to me because I think it's really important that we understand how tech companies are thinking about what they're doing in schools. And Brad Smith said, philanthropy in the private sector is doing this work because public schools have failed to do what they're supposed to do. And his view is that public schools should be teaching modern subjects like computer science. And they were not equipped to do it fast enough. And the tech industry stepped in to help. And is it better than if schools were able to do this themselves? He said, no, but it's certainly better than nothing. And he said to me, you know, think about where we would be on computer science if the tech industry hadn't stepped up. What's your take on that? Well, I've been having this conversation with him for 10 years. And when we started having it 10 years ago, I said to him, I get that. But when big tech companies put their finger on the scale, it tips education in certain ways that not only favor the tech industry, but can obscure equally or more important things. So for example, schools went hog wild for coding. But what about data science? Data science is a really important subject that teaches kids to analyze data. Data science is a major subject that kids would really benefit from using. And also, you know, you learn math, you can learn some coding. Yet, we are not pushing for data science in school. And one of the reasons is that there aren't trillion dollar companies pushing for data science. And there aren't tech billionaires going around saying, if you just learn data science, you'll get this amazing career, right? And yet, data science is crucial. And when I talked about it with Brad Smith, he said, yeah, we need to push for all the, you know, new subjects, not just computer science, except that the tech industry pushed for the thing that benefited them. Natasha Singer, thank you so much. Appreciate you coming on our show. Terry, thank you so much. Thank you so much. Thank you. Thank you. Thank you. Our digital media producer is Molly C.V. Nesper. Roberta Shorrock directs the show. Our co-host is Tanya Mosley. I'm Terry Gross.