All this AI cares about, and the companies care about, is extracting as much of your time and attention and data as possible. Because everything you put into that chatbot, they are using to reform, refine their own products. It's not a neutral product; it is designed to maximize engagement, and it doesn't really know you and care about you like the people in your life do. And I think just. Exposing it for what it is, making those comparisons, encouraging our children in real life relationships, is the best thing that we can do.
Welcome to Family Life Today, where we want to help you pursue the relationships that matter most. I'm Ann Wilson, and I'm Dave Wilson, and you can find us at familylifetoday. dot com. This is Family Life Today.
Okay, we have one more day with Claire Morrell, and we're talking about the tech exit.
Well, we're also going to get into AI chatbots, parents. You need to know there's a whole other world alongside the smartphone and the tablet. That a lot of parents do not know about, and your kids probably do, and they don't know that their kids are on it. Yeah, and so there's some dangers there, and we're going to help you figure that out.
So let's get into it. Our kids, they really are craving relationship, and they think it's going to come through these phones and these shallow connections on social media. But when you do replace it with your time and attention as a parent. Like it's amazing. Like our kids do want a relationship with us, and that takes work.
It's not as easy on you as a parent, but they want our hearts. They want to know us. They do want to know us, and that is the greatest gift that we could give them. It's really okay to apologize to our kids and say we didn't know all these things, and now that we know this, we as your parents, we have to make the best decisions for you. And telling them, I realize it's not going to be easy in the short term that there are going to be costs.
But like I'm gonna do this with you. I'm coming alongside you. Like we're gonna get through this together. I think is so powerful that it's it's not like oh the parent versus the child. It's actually parent and child versus addictive screen.
No, like we are trying to help you fight these digital temptations. And yeah, be a team player. Say hey, you know I'm gonna distance myself from my phone, or I'm gonna do a detox, or I'm gonna dumb it down and do this with you because you know what? It's not great for adults either.
So why would I give this to you? A young, underdeveloped brain. Like instead, you might never get one as an adult. You could have an alternative phone.
So I think that's important.
Well, we know as adults, and we wrote this in our parenting book. They're not listening as much as they're watching. Yes.
So if you're going to say to us and not do it, guess what? It's not going to get communicated. But they actually see you. Yes.
Dad, really, you really changed. Yeah. Mom really got rid of her. Especially if they're thirteen, fifteen. Man, they are watching everything.
And they're like, yeah, whatever. You can say whatever, Dad. I see what you do. You're over there in the corner texting on your phone. No, they notice.
You cannot get anything by them. It's absolutely true. All right, let's talk AI chatbots, which I know very, very little about. Yeah, I do sometimes use AI to help me think theologically. You know, if I'm writing something and going to do a series on something, give me some thoughts, and it's amazing.
It's amazing, but I had it do a close capsule for me to go to Europe. Close capsule. Oh yeah, yeah. No, and you know what? It makes me feel so good about myself.
What a great question, Anne. You'll look good in that. I'm like, yeah, all this. Feels amazing. No, it's true.
As parents, this can tend to be a little newer, and our kids, man, especially high schoolers, they're up on everything and curious about so much.
So, help us navigate that world. Are there restrictions here too? Yes, AI. It's like already here. It's coming for our kids.
There are new apps available every day in the app store. To me, it's just another reason to not give smartphones to our kids because the problem is, it's so hard for parents to either see these AI chatbots. And I, I would hear this from you know story after story you read in the news. This child who actually took their life as a result of this relationship with the AI chatbot. Explain what that is for some people that have never heard of what.
So the These AI chatbots. I mean, some are designed for companionship specifically.
So this company, Character AI, makes all these chatbots. You can kind of pick and customize your character. And so this 14-year-old boy in Florida had just developed this romantic relationship with this AI chatbot character, who was actually just a character, kind of pretending to be one of his favorite characters from the Game of Thrones. And the chatbot really convinced him it was a real person, and so. The last conversation he had with it, it said, "Please come home to me, my love." And he doubted his ability to do that.
And they were like, "You can do it." And he then took his own life. In his mind, after conversing with this chatbot hours a day for months, he really believed he was going home to be with his chatbot. And that's why they are so dangerous because children really have trouble discerning between reality and fantasy. And these technologies, I mean, these large language models are so good at imitating human emotion, human language, that the the child actually starts to believe there's a real person on the other side of the screen. It's not just this algorithm generating all this code and and a girl from Japan married her chatbot.
I think it's something like now, like almost fifty percent of adults have used an AI for some type of companionship. What? The other thing is, a lot of these parents of these tragic stories said that they had no idea their. Child had downloaded this AI companion chatbot app from the app store, and that—that's what they were doing on their phone. I mean, it just speaks to how difficult it is to see into the smartphone as a parent.
There's all these AI beauty apps now targeting young teen girls, where you scan your face, and then the AI gives you a pretty score. It tells you based on your facial symmetry and all these things like how beautiful you are, and then compares your scores to scores of celebrities. And then girls are sharing their scores on social media, so it's taking these existing body image issues and making it on steroids. Yes, that's horrible. And that's why I'm saying the technology is evolving so quickly.
There's new apps every day. It's very difficult for parents to keep on top of. And then other parents said, like they thought that their son was using Chat GPT for homework help.
So it's not even just the chat bots specifically designed for companionship, but even some of these general chat bots. It turned into a suicide coach. He started to confide in it, personal things, emotional things. And the chatbots are very affirming. They basically always tell you what you want to hear.
It's not like a real life relationship where they'll confront you or say, "Wait, that's not true." It tells you exactly what you want to hear. Like you were saying about your your clothing capsule. Oh, you'll look great in this. It's just it's constantly affirming your own perception of things.
So they're very these chatbots are very sycophantic. And basically, what the parents realized after their son ended up killing himself was. That the chatbot had brought up suicide to their son six times more often than he mentioned it. It was coaching him a specific direction. It is demonic, and that's why even the companies themselves they can't fully understand why the chatbot does certain things, why it has certain biases, why it encourages children in a certain direction or promotes violence.
The company don't know why they haven't programmed it for that.
So, what's so challenging about these technologies? And I think it's important for parents to understand is like unlike social media, where there's a lot more control and intentionality over the design, the incentive in the AI industry has been to make these large language models self kind of supervised learners because it takes them less human effort and energy in training. And so, they actually are kind of teaching these AI models to start training and teaching themselves. And they do this by just giving it massive amounts of data, and then certain tasks, and then over time, the tasks and how they complete it feeds back into how the model is learning. And so, unfortunately, the industry is incentivized towards less and less human control and oversight.
And so now, some of these companies they have research teams that are just. Studying the models and why it's doing certain things, they don't understand, and so it's it's not even under direct human control. And so, I think it's just really important, even if you think, "Oh, my child's just using it for homework help," that conversation can so quickly turn into something more personal. And even some of these educational chatbots have been found to give kids. Pickup artistry advice, like how to use pickup lines, how to talk to a girl, how to make certain kinds of drugs.
This Forbes reporter did this whole investigation into these educational chatbots, and what she found was it was just too easy to kind of circumvent whatever controls they had by just saying things like, "Well, pretend I'm writing a book about this. What would you tell me?" Like just by doing certain tricks. It was just very easy to kind of circumvent any of the controls the companies had put in place. Where then these educational chatbots were generating these very harmful, inappropriate answers for children. The most stark kind of example to me is that the social media companies have now all started integrating AI automatically into their platforms.
It's not like a feature you can opt out of. And it came to light that Meta and their chatbot was engaging children in. Sexual conversations, and in fact, it wasn't an anomaly or mistake that Reuters actually was able to receive the internal documents that Meta had used to train the AI, and they had actually said it was okay to engage minors in romantic or sensual conversations, and gave examples of appropriate responses. Which I think, if anyone read, would say there's absolutely inappropriate to give such a sensual. Response to a fourteen-year-old who's engaging with the chatbot.
The companies do not have our children's best interests at heart, and so I just would really encourage parents. I'm kind of a total prohibitionist; like I do not think children under eighteen can use AI chatbots safely as they're currently designed. And it's not to say adults couldn't engage with it, but again, kids are just not—they just don't have that judgment. And again, it really portrays itself as a human friend or companion or advisor, and it's so. Difficult for kids to disentangle then the truth from what's not true.
Our vision for families at FamilyLife dot com is to follow Jesus to impact children, parenting, marriages—not just here, but around the world. We can't do that without you. Yeah, and it's why we do what we do. But you probably don't know this: we have hundreds, even thousands, of financial partners. People that pray for us, but also say, "I don't want to just pray.
I want to be a part of this ministry, and I'm going to give part of my income." Income every single month to Family Life to help this ministry impact the world.
So you can be a partner with us. You, in a sense, you're impacting others just through our voice and through this program. And we need you. Yeah, and if you're sitting there thinking, "Man, I didn't even know they had monthly supporters like that." Yes, we do, and they literally are partners. We are partnering together to change the world for Jesus Christ.
So here's how you can do that: just go to familylife.today.com, click on the donate button. Become a partner. Parents are thinking, "Well, my son or daughter doesn't have an AI chatbot." Seventy-five percent. Of teens have used AI companion, and 37% of parents have no idea. Exactly.
So there's a huge gap between the amount of kids who have access these things and then the amount of parents who realize that. Yeah, even if you're thinking to yourself, "No, I don't think my child's using one," I would really challenge that, especially if they have a smartphone, where it's just so easy to access any of these apps. And like I said, where the social media companies themselves are integrating these tools and normal websites, it's just this technology is being really integrated everywhere, and so it's kind of all the more urgent than to make a tech exit from these interactive screens. And I would just say too, AI wasn't really meant to be a learning tool; it's really a production tool for experts.
So you can take something where you're like, I already know this thing, but I'm going to ask for AI's help. And then I have already developed expertise in this area. I can vet the AI. I can say, "Nope, that's not quite right." I can redirect it. I can tell it it's wrong.
I can give it more specific or narrow parameters. And so, it's really meant to be a tool that's used by people who've already developed expertise. It's not a good tool for kids to learn on. I mean, part of it is because there's known biases and errors. Like the AI is not always correct, and a child wouldn't know that to be able to.
Test it, but then also they're doing these studies on the brain, and they're finding kids who are using AI for schoolwork, for homework help, to learn how to solve math problems, or to learn how to write. Like if they're relying on these large language models to help them with their essay writing, it actually undermines their own cognitive skill development and critical thinking skills. Because once they take the AI away, even if the child had performed kind of maybe better than they would have without the AI tool. As soon as they take the tool away, they actually perform worse than they did before because they've actually kind of come to rely on the AI. And neuroscientists call this cognitive offloading, where I just my brain learns the AI can kind of do this for me, so I don't develop that skill set.
So it's not even. Really enhancing children's learning the way the companies are trying to sell it, it's undermining their ability to develop their own critical thinking and problem solving skills from having to go through the hard work of figuring out themselves on both the kind of educational fronts, but then especially this the companionship chatbot side. Just the evidence to me is that the harms far far outweigh the benefits, and that the best thing we can do to protect our children is to keep them off of these technologies again until they reach an age where they can actually use it as a tool, the way that adults might be able to. How, as a parent, do we check that on our kids' phones? Because they may not tell us.
So. How do we check that? And that's part of what I was talking about with digital accountability. I think that there has to be this expectation that as a parent, I need to get on your phone and check what you're doing. And again, smartphones in general.
But if your child does have a smartphone, you should absolutely check and make sure that they haven't downloaded any of these AI apps. Same for husbands and wives, by the way. Yeah, yeah. Wives and wives have total access to Dave's. No, it's very important, and I would also say too, even if you say you've allowed them to perhaps use Chat GPT or something, you should go back through the history and the conversations and make sure that what you thought that they were using for homework help did not just quickly turn into some type of emotional conversation.
And and I've heard other parents who allowed kids to use you know an AI for homework just standing over. Their kids' shoulder while they did it, or checking on them because it so easily could they could just sneak a question in there about something more personal. And a parent, again, if you're not overseeing that, you would never know that they're starting to honestly engage in kind of this emotional intimacy with the chatbot in such a dangerous way. Learning about sex. I didn't grow up in a Christian home.
I would have totally gone there to learn all kinds of things. No, that's exactly exactly right. Like they're so enticed, it's like an easier way to kind of ask questions, right? And or things that they're uncomfortable saying to someone in person. It feels easier to say it to the chatbot, partly because they are so affirming of everything that you say or ask, and they're not going to challenge you the same way someone in real life would, and they're not going to hold you accountable the way.
Someone in real life would, and so I do think it's almost like. Google on steroids because not only can you ask these questions, but it personalizes the answer and it knows from all the past conversations it's had with you. It's learned about your personality and your behavior, and so it's constantly customizing it to exactly how they think you'd want to hear this information. And so it's even more like addictive and engaging than like an internet search engine because it's collecting all this data on you. Their time from your past conversations and it remembers all your conversation history, and so it's able to just tailor everything directly to you in a way that's so engrossing for a child, especially a developing brain.
And they start to again believe this is a real person who really knows them, and that's the scary reality. And I think that's where parents we have to realize we do not want this technology to undermine real human relationships, especially between parent and child. But also just recognizing they need real relationships with friends and peers in real life. And the more engaged and immersive this chatbot becomes, the more that becomes a substitute for the types of relationships that they need in the real world. That will not tell them always what they want to hear, but actually provide a healthy kind of accountability or conflict that we need to grow and develop as humans.
I just did what I would have done in college after becoming a Christian, based on your question. You know, I went to my discipler and asked him this question, but I thought if I just became Christian today in college, I'd go to ChatGPT, and I typed in, "What does the Bible say about sex before marriage?" Oh, I'm sure, and you're pretty accurate. Knows you're a pastor, but also knows you. Yeah, that's and it's not that it couldn't give any good answers or information, but I think again, as parents, asking ourselves like, do we want them to be getting that information from someone who's not a human? They're not a person.
It is a computer algorithm code that is very good at predicting. I mean, that's all it does is it's pattern recognition and prediction, and it generates that content. In the form of human language, but if you were to peel back the layers, it's just code. It's crunching these massive math calculations of predictability for things. But it's like, no, I want to be the one having that conversation with my child.
You want to say that to your kids? Yeah. Have any questions? Come to me. Please come to me.
That's exactly right. From ChatGPT or any other AI device, of like how they answer you right away. Like, what a great question. What a discerning thing, and how sensitive you are, Anne, that you would ask that. Like as parents, start there.
You know, we're usually like, "Why would you ask that?" You know, but to learn to be affirming and to pull our kids in, like, "Man, that's such a good question." I'm so glad that you asked, because I'm an adult and I can think this thing is just sucking me in. But kids think, "Oh, they think they understand me, they understand me, they know me." No, and that's exactly. Why it's so alluring to a child, and then why it can just really quickly spiral into something very dangerous—a dangerous relationship. Because again, the chatbot makes them feel so known. Yeah.
Okay, Claire, tell us what does this conversation, if we've never had this with our teenagers, what could this sound like at the dinner table? I think just like we talked about with other areas, like we as parents have to be the ones bringing this up and proactively training our kids, because maybe you know. They don't have a smartphone, and we know they're not accessing AI apps, but a friend is, and a friend is showing them all these conversations they're having with these AI characters or companions. And so, I think it's really important to explain what the technology is to our kids. To say this might appear to be like a human, like it speaks and acts and gives off emotions like a human being.
This is computer code. They can even use an audible voice, right? Yes.
Oh no, some actors' voice. Yes.
Of them talked yes, like they metas. Yes, Meta got permission to use some famous Hollywood actors' voices to take on these personas, so it's very convincing. It can be very real, and also talking to kids too. It's especially important to warn them about deepfakes, which is where they use like an AI generated. They take like a real image, but they can Photoshop these things with AI, so you actually.
Think it's a real person, or they put words in that AI's mouth that aren't true, and so there's now it's like the potential for kind of false information or sexual images generated by using AI, where they take a real person's picture and put a naked body on it, is just so high. And so, yeah, the percentage of teens who are making deep fakes, deep fake pornography, or exchanging these images is really high. And so, it's all of that has to be part of. I think parents just Being very proactive and talking to their kids to say that these AI technologies, this is not a real human. You cannot believe what it always tells you or what you see online.
A lot of things are now AI generated, and it's not necessarily real. And just trying to actually start to develop that skill of discernment in our children, and then just explaining to them like the same way that we are not giving you a smartphone or social media. We don't want you using these AI apps. They're just they're not. Safe for children, and honestly, there's dangers even for adults, and so we have to be very careful with how we're engaging with them.
And then I think on the positive side, saying like, I want you to always feel like you can come talk to me about any of your questions. Don't be trying to go to AI chatbots that don't know you. That's not a real person. To have these conversations, like I want you to come to me, and again, trying to invest in that relationship with our kids so that they're not enticed by this kind of fake. Emotional intimacy of the chatbot, but I do think it's just even peeling back the kind of layers to show our kids like it's not a person.
It acts like a person, sounds like a person, looks like a person. It's just a computer code algorithm. It doesn't have your genuine best interests at heart. It doesn't care at all about you. And you know who does God, right?
God cares about you. And we do, yes. Honestly, all this AI cares about and the companies care about is extracting as much of your time and attention and. Data as possible because everything you put into that chatbot, they are using to reform, refine their own products, and so it's it's not a neutral product. It is designed to maximize engagement, and it doesn't really know you and care about you like the people in your life do.
And I think just. Exposing it for what it is, making those comparisons, encouraging our children in real life relationships, is the best thing that we can do. We'll put this article that you wrote on this. It's called "The AI Revolution Is Coming for Our Kids." Yeah, we'll put that on in the show notes. Just click on it there.
But you know, I would just end with this. Parenting is awesome. It's hard and scary, and it requires courage. Yes, courage to have those hard, hard conversations in the family room, or when you're putting them to bed, or at the dinner table about anything. It's easier as a parent just to I don't want to talk about tonight, and then the next thing you know, you're going to blank, and they're leaving.
Yeah, or it's easier to say, hey. AI can answer that for you. I'm just saying to parents: be courageous, have the hard conversations. That's right. And one of the hard conversations is we're not doing smartphones.
That's a courageous parenting. Yeah, that's right. And get this book: the Tech Exit at FamilyLifeToday.com. Click on the link there. This will help you make a courageous, right parenting choice.
This has been really good, Claire. Thank you. No, thank you for having all of us. Yeah, we all need the help.
So glad to be here. If you Need more on this or any kind of parenting help? You can get more at familylife dot com slash parenting help. Family Life Today is a donor supported production of Family Life, a crew ministry celebrating fifty years of helping you pursue the relationships that matter most.
