aiEDU Studios

Teens show us how they use AI

• aiEDU: The AI Education Project • Season 1 • Episode 28

Share episode

Teens don’t treat AI like magic. They treat it like a wrench — something that's useful, but only as good as the person using it. 

On this episode, we sat down with students who’ve put AI tools to work in surprising ways. Between all their experiences, we saw a grounded view of AI as a study aid, a creative partner, and sometimes a risky shortcut that demands stronger digital literacy. 

Jeremy’s experience highlighted accessibility and caution in the same breath. He told us how AI helped him while recovering from a concussion by simplifying textbook passages, but also mentioned being fed inaccurate weightlifting advice by a chatbot. His fact-checking routine for AI (ask for links, check dates, consider the consensus, and talk to a human expert) could be a model for any family to use. 

Julia pushed the limits of AI tutoring by attempting seven AP exams. She passed the exams where she already had background knowledge, but stumbled in subjects where she didn't. Her story showed how AI can accelerate learning but still can't replace it. 

If you care about raising capable, critical-thinking teens in an AI-saturated world, this conversation will offer helpful insight and practical advice. 


aiEDU: The AI Education Project

SPEAKER_04:

What I tell my sister, uh, who's 12 about AI is that it's not conscious, right? It's it's a machine and it's not it's not a human being. So she has to know when talking to AI, her her prompts are just being run through an algorithm. And there's no actual thinking behind that, uh behind the screen.

SPEAKER_01:

Alex, you must talk to a lot of students in schools about AI. What are you hearing from them? And from your experience, how much do teens already know about AI?

SPEAKER_00:

Yeah, I mean, there's like one story that I think really illustrates this. I was actually going back to my alma mater, uh, Copley High School, and the principal has sort of organized two sort of concurrent presentations. So one, I was presenting to like all the teachers in the school, and it was like 200 of them, like a third of them actually had me in class, which was really embarrassing and nerve-wracking. Um, and you know, gave our spiel, and you know, everybody like, you know, their eyes were dinner plates, and everybody was like, oh my gosh, like this AI thing is a really big deal. And then after that, they actually brought me into a student assembly. And so I kind of gave a modified version of the presentation for the students. I sort of pulled out all the stops like humanoid robots and deep fakes and you know, just like the cool stuff like AI generated music. And like as my presentation was going, the kids were, you know, getting louder and louder, which is to say they were not paying attention, they were talking to each other. So after the presentation, a couple kids like approached me and they're like, Oh, you know, Mr. Katron, uh, you know, can we give you some advice? And it's like, yeah, of course. Like, what, you know, what's your advice? And they said, Well, you kind of showed us all this stuff as if like it was new and exciting, but like we know about all that. Like, you know, it's just I feel like our teachers think that this is also new, but like for us, this is kind of just like the norm for like how we what we experience when we're online. Um so I don't necessarily think that kids know everything that they need to know. But if we're not acknowledging that they are also coming into this conversation with frankly a lot of knowledge, I think it's gonna be really hard to connect.

SPEAKER_01:

You are so right. I think it's an opportunity to ask them, like, teach me what you know. Like, we do know less than they do for the most part.

SPEAKER_00:

Yeah, I mean, so to that end, I mean, let's find out just how much teens know about AI and how often they're using it on this episode of Raising Kids in the Age of AI, a podcast from AI EDU Studios in collaboration with Google. I'm Alex Katron, founder and CEO of AIEDU, a nonprofit helping students thrive in a world where AI is everywhere.

SPEAKER_01:

And I'm Dr. Lisa Pressman, developmental psychologist and host of the podcast Raising Good Humans. On this episode, we're hearing from teens across the country who are on the leading edge of AI. They'll tell us how they're using AI and how they think about when to use it and what they know AI can and can't do for them.

SPEAKER_00:

First, we're gonna hear from Jeremy. I'm Jeremy, I'm a rising senior from California and I'm 17. Jeremy uses large language models to help him study for courses. And then last year, he suffered a concussion while participating in a swim meet. And that inspired him to put AI to a new use to help him keep up with schoolwork while he recovered.

SPEAKER_04:

I did get a um concussion last year, and because of that, it was very, very hard to comprehend like long textbook passages. Like I remember for AP AP biology, there were like a bunch of new words, a bunch of complex process uh processes that I did not understand whatsoever. And when I was reading it, it was like the words were just swimming like in front of me. I could I literally could not comprehend what I was looking at. So I gave it to AI and I asked it to summarize it to me, to explain it to me like I was a fifth grader. And it was able to do that in these very simple terms, uh, in a very clear tone. And I think by doing so, it really helped me understand uh AP biology, even though I was in a mindset that wouldn't really allow me to.

SPEAKER_00:

Jeremy turned to AI for another challenge that stemmed from his athletic pursuits. So chlorine is rough on the skin and the hair, and no one in Jeremy's social circles was giving him any tips. And he just needed some help figuring out what to do about that. He needed sources he could trust.

SPEAKER_04:

I mean, skincare isn't really a big thing that a lot of guys talk about, at least in my community. And because this isn't commonly like discussed, I had no idea what I should be doing. I had no idea I was supposed to put conditioner on. Um, I had no idea I was supposed to put sunscreen on. So when I realized things were like wrong with my skin, like it was peeling, I was getting like a lot of freckles, I was looking a lot older than I actually was. I was like, okay, something's wrong. Like my scalp's itchy. I need to ask AI uh what to do. So I told it my problems and it diagnosed me, and it was like, obviously, you're missing conditioner in your routine. Obviously, you're missing this in your routine. And I asked it to give me a daily morning skincare routine and a nighttime routine, and it's helped me out ever since. And I've been able to have very, very clear skin. So I'm super grateful grateful for that.

SPEAKER_01:

So while Jeremy found it helpful for the skin routine, he also knew from experience, gratefully, that it wasn't always accurate or trustworthy. But when he and friends were working out together and trying to lock in on their nutrition and fitness goals, he realized that not everyone knew that you have to fact-check answers from AI tools, especially when it comes to your health and safety.

SPEAKER_04:

I do have a friend who was like, he's super into weightlifting and he wanted to know if it's okay for him to take a performance-enhancing substance. And he is younger than me, so I think obviously it isn't safe. But the problem with a lot of AI, AI bots is that if you phrase the question and you and you lack context, they can give you the answer that you want. So he did ask AI whether or not it'd be safe for him to take, and it said that it would, and he and he did end up doing it because AI told him it was okay. Uh, so I think that kind of brings us to the whole point on digital literacy and like media literacy. You have to be able to like spot this, spot this stuff out and go back to your cognitive reasoning and know that even though AI said it's okay, that doesn't mean that it actually is okay. And I do have processes or steps to take to kind of confirm that what AI is telling me is correct. I remember I was asking it in answer to one of my uh to one of my MCQ, like the multiple choice questions, and it gave me a blatantly incorrect answer. And I was pretty confused by this because I mean, we went over in class, my teacher told me what what the right answer was. And I think that was kind of like the first time I understood that AI could be wrong. And then also to secondly, when I was asking it some uh questions about like lifting, it gave me some advice that I knew was definitely incorrect. And I think because there's so much misinformation out online, you're bound to go against something that's wrong. So as whenever whenever you're you're asking a question, I would always recommend to uh ask it for links for sites uh to source, so then you can read those yourself and verify like when was this published? How long ago was this published? What's the general consensus of this study? Or you could take the advice and go to someone who you know is proficient in that material and ask them what they think about that. I think a good general thing to do would always be to try to educate yourself on how AI actually works, knowing why you're being fed a certain answer is really enlightening. So you always take AI with a grain of salt.

SPEAKER_00:

This is what we want for every single kid is to just know that they need to look and that they need to ask and check. And you know, his strategies are pretty much exactly what I would say. It's like, you know, go and find like a trustworthy source, go and find an expert, get the links, get the links.

SPEAKER_01:

One thing that I noticed is the conversation between Jeremy and his friend, where Jeremy is kind of like, wait a second, hold on. And he's checking him. And I think we also can do that whenever anybody at this point tells you like an interesting fact. I think it's an opportunity to ask, like, oh, where did you learn that? And like, did you check it? Especially if you're picking up on things that are kind of ridiculous. Okay, so the next teen we heard from on the other side of the country took an experimental approach to using AI as a study aid.

SPEAKER_03:

My name is Julia. I'm 18 years old, and I'm going into my freshman year of college, and I'm studying biology.

SPEAKER_00:

During her senior year of high school, Julia decided to get as many AP credits as she possibly could. It's pretty ambitious. So she was enrolled in three AP courses in school and she signed up for those tests. But Julia also decided to sign up for four additional AP tests, ones that she wasn't actually taking the courses for at her high school. So you're probably already guessing where this is going, but let's see how it actually worked out without a teacher or classroom experience to guide her.

SPEAKER_03:

I took three in school that I did take the exam for, and I took four with learning with AI and learning the textbooks on my own. Two of them were in history, was AP European history, AP world history, and the other were economics, so microeconomics and macroeconomics. I ordered the books online and I studied from that. I decided to use AI in the course because I realized that I couldn't do it all on my own. You need some help with it. And I realized that, you know, when you get several like 500, 800 page textbooks, you can't just like sit down at night with your pencil and write notes and read. You actually need to get that recall and get that memory. So I worked, I kind of worked with the different features that this AI offered. And it kind of broke the course material down into like major unit study guides. Seven tests is a lot in two weeks, you know. And there was one day where I had back-to-back tests, and I was sitting in the testing room for like eight, nine hours going through everything, you know, taking the breaks. You can go to the bathroom, get your little granola bar in the corner, and that's about it. I took seven tests, three, of course, that I took the classes in those, and I passed those ones, thankfully, for college. And then the other out of the four that I took by myself on AI, I passed the two history ones that I had a background in, whereas the other economics ones that I didn't have a greater background in, I didn't pass. So I learned kind of from this whole experience was okay, I'm going to work with the AI with things I already know a little bit about. And that's going to take my knowledge a lot farther than, you know, initially. I do think there's still a value in learning in a classroom, of course. You know, otherwise, you know, why would I go to college if I didn't value that?

SPEAKER_01:

Julia also told us about other ways that she's finding AI helpful, including a surprising use in her job as a summer camp counselor.

SPEAKER_03:

Like as a camp counselor now that I'm working in the summertime, it's like, okay, how do I lesson plan like an activity that like your average six-year-old like will understand, or how do I explain this, you know, in terms that they will understand? I think one of my favorite examples was we were going to make an escape room for the kids. And it was, okay, I have this list of materials and I have, you know, these supplies. And how would a, you know, a six, seven, eight, 10-year-old enjoy this and also understand it and it worked at the way their brain's working right now. So I saw a lot of success with the kids enjoying that and myself enjoying the process of it.

SPEAKER_01:

I would be remiss if I did not mention that's a lot of AP exams to take on. And she kind of found out for herself that she didn't really learn enough unless she had something that the AI tutoring could build on. So there really wasn't a shortcut. So I think that was that was good. I can see how it could be really helpful to figure out camp counselor activities and kind of where a six-year-old might be at and what you could do with different materials. Yeah.

SPEAKER_00:

I mean, the escape room sounded fun.

SPEAKER_01:

So totally. And there's no way I could do that without AI.

SPEAKER_00:

Well, I'm really bad at escape rooms in general. So I would, I would not want to. I mean, my escape room would be really easy. So I guess there's that. We have another interview, and this is from a teen who found a really creative way to use AI to help her earn a role in her high school musical.

SPEAKER_02:

Hello, my name is Yuki. I am a senior. I'm 17 years old. And I'm currently a member of my school's choir. So I've been um, I am pretty knowledgeable about music and such, but when it comes to musicals and more of the drama theater, I'm not as experienced. And because my choir director is one of the judges for the auditioning process, I can't go to him for to ask for feedback. And a lot of my friends may be auditioning themselves. So I found it pretty hard to be able to ask someone in person for feedback or suggestions for the audition. So what I did was I asked an AI chatbot to provide me a sample of songs that I could audition with based on the range that I provided it and the musical that I'm auditioning for and the character that I'm going for. And I found it super, super cool how the AI chatbot not only provided me this song, but also um asked me if I wanted like a specific 30-second sample from that song that I could use for my actual audition. So it gave me a song from Chicago. I believe it was called um Mama Does It Best or something like that. I'm sorry. I'm not good at memorizing song titles, but um it was so good. It took my range and like the um the voices that I'm comfortable with, and it really also took into account the role that I'm going for and the mood of the characters in general. And I think this song did me perfectly, and I was luckily able to get that role. So it was just such an amazing experience.

SPEAKER_01:

Yuki doesn't only use AI as a musical advisor, although I thought that was very cool. She uses it to help her study for school too. But it was another use, one that is of a particular concern in general, which is um not uncommon among teens. She asked it for some help to resolve a tricky friend situation.

SPEAKER_02:

My best friend and I, we went to a trip to Disney World together for school. And there we got into a pretty big fight. I didn't really have anyone else to talk to about that because I didn't want to bring um any of my other closest friends into this argument between us two. And I didn't, I don't really talk about conflicts I have with my friends to my family. So I did like ask it to provide feedback on like a text message I was gonna send. And so I inserted the text message to the AI and I asked it if it sounds passive aggressive or how I can fix my tone and make it seem more like I wanted to resolve the conflict instead of like fueling it further. And the AI did tell me I should definitely like um speak to her at least like on call or in person instead of through text, because um, like there's always this question of like our tone and like being misinterpreted when it's like through a text message instead of hearing our voices or seeing our face, facial expressions. So that was definitely good advice. And I did call my friend afterwards. And I also had this AI help me with coming up with a few like phrases or ways to go about starting that conversation. So I think I had a like one of the most mature conversations I've ever had. And I think the having getting that advice from AI has also made me like reconsider how I navigate like personal conflicts and how I go about communicating with others since then. I will say that I've found that while AI can't help with like the personal conversations and personal conflicts, there is this kind of like limitation of the AI not actually having emotions. And it's like programmed to provide like the response that users are looking for. So I think in that way, there's there's not really a way for AI to be like truly objective either.

SPEAKER_01:

Yuki is incredibly mature and insightful. She asked good questions. She got a response based on asking a particular type of question about tone. On the other hand, the best way to figure out how to manage conflict is to experience it and um trial and error and falling on your face a couple of times. And as she pointed out, which is heartening, this is not in context. It says what it thinks most people want to hear and it's not a person. So I think if you could just know that every single teenager in conversation with a large language model asking relationship advice knew all of those things, I would feel more comfortable. And so that means we need to be talking to our young people. And, you know, it it means this is a thing to be paying attention to.

SPEAKER_00:

Yeah, it's we're still kind of making sense of like where the lines need to be drawn. And I think like the the devil is in the details. The use of AI to help you find a song, you know, for an audition. I I think this is maybe going into the realm of AI helping kids that don't necessarily have access to all of the same resources as other kids. Because I think there are going to be, you know, some families that can afford to hire a voice coach that would do all of that work. I mean, I think it's really unfair to say that, you know, someone should have to choose, you know, no support at all because UK was like, you know, ultimately the AI was sort of like getting her in the right direction, but she was ultimately the one going and learning it and singing it.

SPEAKER_01:

Oh, yeah. I mean, I actually thought that was a great example of thoughtfully using AI in a way that does democratize so many things. I really thought that was cool. It was more like, oh, and then it's also being used for this. And it's this bright young woman. So it's just, it's a real wake-up call.

SPEAKER_00:

I mean, so Eliza, so I feel like we learned a lot. I certainly learned a lot. I heard you say you learned a lot, so I'm gonna go ahead and go on a limb. Where's your head landing? I mean, uh, are are the kids actually gonna be all right? You know, it is all the other stuff we were talking about maybe not as important as we thought.

SPEAKER_01:

I always think the kids are going to be all right. Um, I think one of the go-to places for parents and culturally about kids is that they're not gonna be all right and they tend to be wired for resilience. So I I feel pretty good about that. Though these are challenges that are very tricky. We have to be comfortable setting boundaries and understanding that we can be open with our kids about, you know, I don't know enough to give you as much freedom as you would like right now. We're gonna learn together. Maybe you'll know a little bit more than I do. Maybe I'll know a little bit more about the big picture. And I also say that, like, with the understanding that I really don't think there's any reason to panic when you're paying attention, you're curious, and you have a relationship with your child.

SPEAKER_00:

I think the the takeaway for parents is like kind of you're all in good company, like we're all figuring this out. But being overwhelmed or feeling like you don't know where to start is like not an excuse to bury your head in the sand. And that's sort of like the one thing is like you can't, this is not gonna go away even if you ignore it. You know, it's it's it's coming.

SPEAKER_01:

And I think getting to do this with you, Alex, and learning about the landscape has helped me understand that like everything, you know, nobody's going home with the 20 takeaways that are guaranteed to have best outcomes. But I think that's the reality of being a human being. And now we can together kind of remind ourselves that that's what's cool is that we're still human beings raising kids, figuring this out. And I have appreciated this very much, both as a mother and as a developmental psychologist. I think, I think we've got this. Even though there's so much more to learn, we come to a close on our first season of raising kids in the age of AI. Alex and I want to thank you all for listening and for taking this journey with us.

SPEAKER_00:

There indeed is so much more to learn and explore in this topic. Um, one way to continue your learning journey is to check out aiedu.org. You'll find more resources and updates, and uh it's one of uh many places that you can go to find out where AI will take us in future generations next.

SPEAKER_01:

Don't forget to follow the podcast on Spotify, Apple Podcasts, or wherever you listen.

SPEAKER_00:

And we want to hear from you. Take a minute to leave us a rating or a review on your podcast player of choice. Your feedback is important to us.

SPEAKER_01:

Raising Kids in the Age of AI is a podcast from AI EDU Studios in collaboration with Google. It's produced by Kaleidoscope. For Kaleidoscope, the executive producers are Kate Osborne and Lizzie Jacobs. Our lead producer is Molly Sosha, with production assistance from Irene Bantiguay, with additional production from Louisa Tucker. Our video editor is Ilya Magazanen, and our theme song and music were composed by Kyle Murdoch, who also mixed the episode for us. See you next time.