The AI Revolution: What Parents Must Know to Guide Their Kids

Welcome to Where Parents Talk. My name is Leanne Castellino.

Our guest today is the head of research at Common Sense Media. Amanda Lenhart is a qualitative and quantitative researcher whose particular focus is studying how technology affects families and children. Common Sense Media is a nonprofit organization founded in 2003. It's dedicated to providing trustworthy information, education, and an independent voice to parents, educators, kids and families.

Amanda is also a mother of four, and she joins us today from just outside Washington, D.C. thank you so much for making the time. Oh, thank you, Leanne. It's great to be here. Really important topic, of course. Anything to do with technology and parenting today sort of rises up the priority list. For sure. ChatGPT has only been around since 2022, which is hard to believe for many of us, but the ongoing evolution of AI and open communication platforms continues to evolve rapidly.

Common Sense Media recently undertook a study. Can you take us through the impetus for this study and what you were looking to find out? Sure. So, you know, I think Common Sense Media has lots of different arms. We have an education team, we have a research team. We have a ratings and review youth team. We have an outreach team that reaches out and talks to parents all the time.

And I think in 2020, late 2022 and early 2023, what we were just hearing from parents and educators was like, whoa, what is this? What is ChatGPT? What are chatbots? What is this generative AI thing? And what do I need to know about it? So we, as an organization, started to kind of pull together and began to undertake research as a part of a variety of things that Common Sense Media has been doing about and to help and parents and administrators think about generative AI.

And so this research is the fruit of about a year's worth of work to pull that all together. It's interesting because it's all evolving in real time. So as the research on your end is, you know, being undertaken, things are changing. So can you tell us how the study was conducted? Sure. And that was absolutely. One of the critical challenges of this study is, you know, how to stay on top of the work, to stay on top of all the changes with the work.

So this piece of research was done as a. It was done as a survey of parents and kids. And these are paired. So it's a parent and. And one of their teenage children. And so the idea was to try to be able to see and compare together how, how families are thinking about generative AI. And I will say this is one part of a larger project. So we will have more, more that we'll be releasing from the same data set.

So again, it was an online survey conducted on what's known as a probability panel, which is a panel that is created to match the US population. We did also over samples of youth and families of color to make sure that we were able to have enough data to be able to have significant, statistically significant findings about those groups as well. So yeah, and then we pulled that all together, did a bunch of analysis and released it as this piece of research.

So let's dive into some of those findings. Can you take us through some of the key findings of this study? Sure. Well, you know, one of the things first off, I think everybody wants to know, like how many kids are using it, how many parents are using it. And we do see that it's about 7 in 10 adolescents. And so I should say this work is of kids 13 to 18. But if you're 18, you're still in high school. So it's high school seniors is the, is the oldest part of this study.

And so we see about 1 in about, about 7 in 10. So about 70% of kids say they've used it, not everybody who's used it that frequently. So use of it still isn't like integrated into a daily practice for most teenagers. And we see that of that group that say they have ever used it, about 40% of them say they've used it for school. And so we did dive into that. Parents, fewer parents have ever used it. A lot of parents say they're aware of it, but only half have ever talked about it with their kids.

So I think there is a little bit of a disconnect where parents don't have the time. The kids have a little bit more time, a little bit more play, time to play. I think we see that in how much people are using these tools. We also see different uses of the tool. We wanted to know why are you using it? What purpose in your life does this serve? We see that the top reason for most teenagers who use it is to help with homework.

But then the second most important reason was to help me keep from being bored. Right. So there's this real element of using it for sort of fun and pleasure and amusement as well as for sort of more utilitarian kind of academic uses. And I can pause there. There's a lot more we can talk about, we can walk through in greater detail about some of the school based findings. But those are sort of the highlights around use of generative AI.

So let's Unpack a little bit there of what you said, seven out of ten. So 70% of those surveyed. Just jumps off the page, I would think. Did that surprise you or strike you as a researcher? You know, we had done a lighter weight data collection the year prior and the numbers were lower. It was closer to a little bit. Less than half of teenagers at that point would use it in early 2023, which I was surprised by because the technology was really quite new. I think some of it too is, is how.

What are people necessarily thinking about? So some of our earlier work focused on ChatGPT as the one thing we thought people would know about because it was the most, the earliest mover in the space. But this other work is a little. Taking a little bit more of a broader lens. But I, I would say at the end of the day, I'm not that surprised.

I think it is something that people have heard of, they've tried, they've played with, but I think the majority of kids, it's not something they're using all the time. And I think that's an important distinction when we're talking about how integrated this is into teens lives. I guess the other question that just comes to mind as you talk about those statistics is just how far behind are parents in this equation? Yeah, I mean, parents are. They're a little behind.

Again, I want to throw parents a bone here. As a parent myself, like, keeping up with the rapid change in these tools is extremely difficult, especially if you're not somebody lucky enough to have it integrated into your work life. Where you might have to do it, you might have to understand it. It can seem, it can seem opaque, it can seem hard to understand.

Actually, my colleagues in our parent team created what they call Parents Ultimate Guide to AI to try to help parents with some of these questions. Like, I don't even know what it is. I need, I need like a handbook. I need a quick. A cheat sheet for what AI is. So I think a lot of parents feel that way. And if you don't see a good way to integrate it into your life or how you might use it, it may not be something that you've had a chance to experiment with.

So I do think that's a real challenge for parents. I think parents know it's something they need to talk about with their kids. I think that sometimes they don't know where to start. And they also tell us they're not getting information from their child's school. So 80%, more than four in five parents said that their school had Never talked to them about generative AI at all. And I think parents are hungry for more guidance and more help just in understanding schools approaches to this.

I think educators also don't always know what to do either. So I think a lot of us are in this, like, phase where everything's so new and it feels overwhelming. And so sometimes some educators, I think, really embrace it and love it, and others, I think, just sort of feel a little paralyzed and don't quite know where to begin. Certainly new, ever evolving.

And so the onus then, like many things with technology, you know, you look at social media and some of the other pieces over the last little bit, the onus then really becomes that of the person, the individual, the family, the parent, to say, okay, you know what, I'm going to have to sit down and sort this out. Is that the correct approach? And how should a parent go about informing themselves about generative AI?

Yeah, I mean, I think, I think this is going to be a team effort, really, because I think for a lot of. I think there's sort of multiple dimensions of use. And this is something that came out of our work, which is when we sort of did a sorting of all the different ways that young people told us they use generative AI kind of fell into two groups, which is one that's sort of very utilitarian for sort of work, kind of.

I'm translating something, I'm summarizing something, I'm using it to write something, and then these more personal sort of uses where I'm asking it to help me plan something, I'm asking it for health information, I'm asking it to help me. I'm asking it to help me make a joke or to create new and fun content. So I think there's sort of two different types of use, and that's something that's, I think, also important for parents to understand that this isn't just for school.

But there's a lot of like, creative and sort of pleasurable uses of this, you know, as where, where to begin. For parents who want to know, I think it's, you've got to start by trying it yourself and going to one of the chatbots might be the easiest way to use generative AI. And we should say that there's sort of two different types we're talking about here.

Generative AI is the kind of AI where it's creating new content versus traditional AI, which is following more of a. Of a, of a kind of a coded pattern and sort of its outcomes are expected as opposed to Generative AI, which has, like, kind of expected outcomes. And if we want to go a little more deeply into, like, what's actually happening with these models, we can. But I think the important thing for parents to know is that you don't always know what you're going to get with generative AI.

And so it's actually really important for parents to start playing and to try it themselves so they have a sense of what it's like. So along those lines, then, Amanda, what would you say is the key takeaway for parents from the Common Sense Media study? Yeah, that's a really good question. I think the key takeaway is it. Is it is time to try to play with this with your children. And I know it's harder with older teens. They don't really want to.

They don't want to sit down next to you necessarily and play with ChatGPT like a younger child might be more interested in doing. But I think it's time to ask questions, to ask your school about what they're planning to do, to ask them to share information with you as a parent so you have an understanding of what their expectations are in terms of your child's use for school.

And then I think, you know, I think it's time, you know, the takeaway for parents is that your children are using it and they're reasonably savvy, but not perfectly savvy. And part of what we as parents need to help young people understand is the ways in which ChatGPT isn't always perfect. Right? So one of the big challenges with ChatGPT is it.

It lies, right? It doesn't. It's not actually. It's trained on the universe of data. And that can be accurate data or inaccurate data, that can be biased data, it can be racist data, it can be perfectly wonderful data, and it's all in there. And so what it spits out can reflect all the different things that's been trained on. So it can spit out things that are racist or sexist or might attack an identity of your child.

And depending on sort of where they are and who they are, they may feel uncomfortable by some of the things that they see. That. That these chatbots or generative AI spits out images that they might spit out. If you're trying to do it for school, helping your young people understand that you need to check another source or two or three. Right? The chatbots make up things. Right? The whole point is that they're just.

They're deciding what's the next most likely thing that should Be there, the next most likely character, the next most likely word, the next most likely pixel. But does not necessarily mean they're going to check if that thing actually exists. So we know that these platforms, the term is hallucinate or make up information. So they've made up books, academic articles that don't exist. They get dates wrong, they get math wrong. They're getting better, but they're not perfect.

And so helping our young people realize all the ways in which these tools can't always be trusted, especially for things where being accurate is important. It's probably not important if that funny picture of a bunny in your backyard that you asked to make for you is accurate, but it's probably a lot more important if it's you're asking it to help you summarize something or help you brainstorm something. It's a really important point.

And I wonder if the Common Sense Media study was able to discern just what level of understanding kids in these age groups had about what's going on in terms of the accuracy of the information, the items that you just outlined you trustworthy, is it as we continue to see the guidelines and guardrails for AI evolve in real time. Yeah, that's a, that's a tricky one. I think what we found is that a percentage of young people know that the content's not correct.

About half of teenagers who've used generative AI say that they've checked, double check the output against another source. The other half haven't. We don't know if they do it all the time, so there's not necessarily a sense of how frequently this happens. But we know that they have done it so they understand on a certain level. But again, that leaves about half of young people who don't know that.

What we also saw in the study that was actually really critical is one of the things that came out is that teachers don't always talk about this either. As I said, teachers are sometimes paralyzed. They don't really know what to do. We have a whole swath of teachers who'd never mention it. It's like it doesn't exist and they don't talk about it at all. Another group of teachers who just ban it and say that you can't use it at all. And then we have some teachers who really embrace it.

But teenagers don't always know what the rules are. They don't know where the guardrails are. They don't know. They don't know what's appropriate and what's not appropriate. And they really want that guidance. What we do find though is that once teachers do start to talk about this, and this is about a third of teachers have done this, a little bit more than a third.

Once teachers do talk about it in the classroom and have lessons about it, young people's understanding of generative AI really, really blossoms. It's like teaching, it works, right? Like, I mean it's not, it's not an earth shattering finding, but I think it's really encourages, I think teachers to take time to have these conversations and for parents to encourage their child's teacher to, you know, take the bull by the horns and start talking about this.

Because once you start having these conversations, it really opens up, it just really opens up a much more nuanced understanding of the content. Young people who've had these lessons are much more likely to check the accuracy of their work than young people who've never had them. So you can even see it in the behaviors that young people exhibit. So that's a thing where I would also encourage parents to start getting involved in their child's school and in their use of these tools.

As a researcher, Amanda, I'm curious as to what struck you, if anything, about the findings in terms of where we are and where we need to go next. Yeah, I mean, I think there's a lot more work to be done, I think in terms of providing young people with a set of clear guidance about when it's okay to use these tools and when it's not. I think some of that's going to come down to explaining like why, why we think learning is important, right?

Hey, the one conversation we might want to have is like, hey, we think it's not okay to use these tools to write a five paragraph essay for school. Because one of the things you really need to learn, even if in your regular life, later in your life, ChatGPT is going to be, or all of these platforms are going to be, these chatbots are going to be writing your papers or writing your work for you. Right now you need to know what's good, what's not good.

You need to learn the basic skills so that you can manage those tools later and so that you can evaluate it and determine whether it's, whether the, what the, what the tool is providing you is worthwhile. And so that's why we're going to ask you not to use these tools right now. Later in your academic career it might be okay. But for right now, part of what we're doing is learning how to do this.

And these tools are Disrupting that, and I think that's a really important conversation to have is to talk about why this matters and why you want, what you're hoping young people will learn and why ChatGPT or Gemini or Claude or any of these chatbots or other image generators or video generators or other, other instantiations of these tools might be getting in the way of that. Right.

I think that's really, I think helping and bringing young people along for many young people is really helpful in helping them to understand how to integrate these tools into their, into their academic and personal lives. One of the challenges in many households will be generations, right. The generational divides in terms of understanding of these different technologies and the appetite to want to learn about them.

What would you say to parents who may be like some of the educators that you mentioned earlier, might be actively resisting learning about these different technologies because, you know, they don't have to know about it, but now by function of the fact that their child is exposed to it. This study, other studies, you know, now it's become incumbent to at least be informed at a, at a very basic level.

You know, what would you say to that parent about making them less resistant and having them try to embrace learning about AI? Yeah, I mean, I would emphasize the things about it that are positive. I think there are a lot of positive things about it that can bring to our children's lives and our lives. And I would encourage folks who are skeptical to try one of the chatbot platforms and ask it to do something for you that maybe you needed to do, like ask it.

You know, one possibility would be ask it to make you a packing list for a trip. And this is one thing that we did over the summer with my 12 year old, 12 year old was complaining about what do I pack? We're going to this, the mountains and we need. What do I pack for the mountain? Going to the mountains. Well, why don't we see. Let me see what one of these chatbots says. And the chatbot went through and gave us a whole list of things.

And then we checked the list because it included some things were like, well, maybe you don't need to bring flip flops to the mountains. And so that was helpful for us. It saved me the time of writing out a list for my child. My child got to interact with it themselves. So it was something that we could both sort of saved us some labor, but also helped us to think about like how these work well and how they don't always work well.

And so it was a, actually a helpful way for all of us to learn on what's good about these tools and what's not. So I would encourage people to try things like that. Ask it to plan something for you. Ask it to plan meals for the week. We've done that. And you know, sometimes it's offering new things. We were like, I don't, I don't want to eat that. That doesn't sound good. And sometimes it offers delightful things. You're like, oh, yeah, we used to make that. Let's have that for dinner this week.

So there are lots of ways that as parents, there are things these platforms can do to help us that might make our lives a little easier, might be an entry point for those of us who are a little skeptical to think about, to learn about it and also to have it be a little bit useful.

Amanda, in terms of context, how different or similar would you say that AI is when you compare it to social media and what parents need to know about that whole world in terms of being able to then support their children, navigate it again as things are constantly changing and evolving? Is there a comparison to be made there? Yeah, that's a great question. I mean, I think we see that most often in some of the questions around the regulation of these spaces. Right.

You know, generative AI is really new. Social media has been with us for some time and I think there's a sense that, you know, with social media we kind of let things go a little too long. We should have been having more conversations sooner about ways we could make those platforms a little bit more positive for our young people. I think with ChatGPT, we're trying to get ahead of that a little bit more.

I don't think they're especially analogous, but I think they do each come with both positives and negatives. And that's, I think, the thing, the thing to take away. Right. There's concerns with these platforms about what kind of information they're collecting about you. Right. As a part of using them, we often share personal information about ourselves and our children share personal information about themselves. And in many of these chatbots and other platforms, they retains that information.

And sometimes you can find out what it is. Actually just did this with my husband this morning. He's like, you won't believe what my, my chatbot has been keeping about me. And he asked it to show and it walked through and it turns out it knows that he likes basketball and it knows that he's also interested in work related topics. And so I think that's something to think about. On how much talking to your young people about how much personal information they're sharing with these.

And this comes up in also in even educational platforms. So some of the tutoring chatbots, which I think have a lot of promise for helping young people to personalize their learning and help you address specific challenges that you're having on a particular academic topic. But part of what they do to create a rapport with you and your child is it retains information about things that knows about you and returns those back to you to create more of a relationship with you.

I think we could also say there might be concerns about having a chatbot know you and create a relationship with you. And I think that's another risk that's starting to come up in some of the data where we see this, where young people are asking chatbots for advice.

They tell us that they practice difficult conversations with it, they talk to it when they're bored, and in some ways are creating, or could be creating what's known as parasocial relationships, sort of a non human, but an effective relationship with the chatbot. And I think we have a lot of research to do to figure out what that really means for our young people. So going back to the Common Sense Media study, Amanda, which incidentally is called the dawn of the AI era, what would you say?

How can the findings of that study be applied in households, in homes, whether it's a parent who does have an understanding of AI or not?

Yeah, I would say the thing that comes up most for me is sort of one of the big learnings is that we need to talk about this. We need to open up the lines of communication around generative AI, whether that's between child and parent, between student and teacher, between parent and school. There's a whole triangle of information where we really don't have enough flowing back and forth.

And so I think, I think all of the entities, the student, the parent, the educator, the administrator, all have responsibilities in those different, in those different quadrants. But I do think that a parent's takeaway here is like, it's time to talk about it. If you don't feel comfortable, it's time to learn about it. There are things you can do with your child.

Other things that I think are really helpful is, you know, as parents, we can be the guides to helping our young people have a more nuanced understanding of these. Hopefully, educators will join in that conversation. But for right now, parents may be the ones who are carrying that, carrying that bucket right now. And I would suggest, you know, having conversations with your young people in which you both, you both play with the tool together.

And in particular, one device I really like is asking young people to ask a chatbot about something they know a lot about, something where your child has a real depth of knowledge. Maybe it's a sports team, maybe it's musical artists, something that they really care about, and then see what the chatbot returns. And it's such a great way of sort of highlighting the chatbot hits and where it misses. Right? It can bring in and allow your child to see very viscerally.

They're like, wait a second, that's not what you call that dinosaur. Or like that's not the right thing. Or that book doesn't exist. So it can be a really powerful way of showing your child some of the challenges that that AI presents. So I like that. But also encourage play, right? Like play with it together if you can, if you have that relationship with your adolescent.

Because that's another way of having these conversations about bias, about accuracy, so that your young person can really, as they use this, understand what they're getting themselves into. In that same vein, are there any potential risks of, let's say, parking this discussion, you know, not having it in the short or medium term as all this technology continues to evolve and become, you know, more part of everybody everyday life? Really?

Are there any risks to parents who don't want to take those steps in the short term? I mean, the risks are that your, your child will be misled by these platforms and will in some ways be harmed by kind of biased material that it can, that these different tools can offer to your child, that they will have, that if they use it without using it thoughtfully in relation to academic work, that they may lose out on skill development. Right.

That there might be some skills that they could be developing that they don't develop because they're using this as a substitute.

So engaging with your child about why, why we go to school and why that's important and what are appropriate ways to use this, and hopefully hearing from your school about the school's parameters around what are appropriate ways to use this, I think will really go a long way to ensuring that your child leaves their educational experience with a lot of great tools about how to write prompts and how to understand generative AI, but also with some real cautions and knowing kind of where it, where

the use is really positive and helpful and efficient and where the use is problematic. What could the next step or the next phase of the Common Sense media study look like and include in your view?

Yeah. I mean, so we, we already have a next round that we're working on that will come out in the early part of next year, and that's really going to focus on trust and how young people think about trust and the relationship between the things that they see that are produced by AI and how they trust information that they see online. And that's, I think, a critical issue. Right.

It goes back to these ideas around trust in how you trust the material that it's sharing with you for educational purposes, but also also as a citizen, as a person in the world, as a human being with a body that needs to be healthy. How do you interpret information that you get? Because we know that young people do use it for health information, but maybe not always accurate. So again, I think getting digging into some of those questions is really important. Lots of great tips for parents.

Amanda Lenhart, head of research at Common Sense Media, really appreciate your time and your insight today. Thank you so much for being here. Thank you, Leanne, it was a pleasure. Thanks.