The Decibel - Stressed-out parents are turning to ChatGPT for help

Episode Date: August 6, 2025

Parents seem to be turning to ChatGPT for advice more and more – but what do robots have to offer when it comes to such a core human relationship? What guardrails are necessary when trusting artificial intelligence with shaping the development of our children?Contributing columnist Amberly McAteer weighs in after ChatGPT’s advice actually got one of her daughters to stop stealing from the other.Questions? Comments? Ideas? E-mail us at [e-mail]

I don't know, I was really torn about using it because I didn't know if I wanted to be part of its betterment. Amberley Maceteer is a contributing columnist who writes about parenting for the globe. She recently had to confront her own reservations about generative AI. There's a lot of time and energy and money right now going into generative AI to literally get the bot

to learn how to write and sound like a real person. person, which I mean is not great for a columnist. So, you know, I was just kind of avoiding making eye contact with the robots in any real way. Amberley eventually did look at the robots after she found out that a lot of her mom friends were using it for parenting advice. But is AI actually giving good advice? To test it, I fed it a scenario with Amberley, who is not just a parenting columnist, but also a mother of two. Okay, allow to use, allow this time. Okay, we're going to restart that.

Failed to start voice mode. Wow, after all of that, what, okay. Hey there, what's on your mind? Okay. So we have a parenting scenario here that we're going to run through chat GPT. I'm just going to read it out loud. My five-year-old kids sometimes acts out, and my partner and I are always responding with consequences

that we know we won't follow through on. Like we'll say, if you don't do X, we won't go to Y. Knowing full well, we are going to go to Y no matter what. So how do we gently discipline our child without falling down the empty threat rabbit hole? It's really common to feel that way. Setting clear and consistent boundaries is key. One approach is to focus on positive reinforcement

and natural consequences. For example, if they don't pick up their toys, they might not be able to find them when they want to play next time. When you set a consequence, try to ensure it's something you can follow through with. Also, try to focus on talking through feelings and encouraging good behavior with praise. It's all about balance and showing them that you're there to guide them. I mean, I don't hate it. Today on the show, Amberley discusses what she learned from using chat GPT for advice with her kids.

I'm Madeline White, and this is the decibel. from the Globe and Mail. Hi, Amberly. Thanks for joining us. Hey, Maddie. This is great. Thanks for having me. So we are going to get into your own experience using Chat GPD for parenting advice. But to start, Amberly, how much of a trend is this really? Like, are parents actually doing this? Yeah, absolutely they are.

I mean, in the last two to three months, it's all I hear is Chat GPD. Really? You know, in my mom chats, everyone's answer of late to my question. questions for guidance. You know, I'll often, like, put these questions like, what on earth are we making for dinner? Guys, I cannot make spaghetti again. Or, you know, how are we encouraging five-year-olds to read or, like, give me some tips, moms, what are we all doing? And so many of these answers of late have just been, I don't know, I just asked chat GPT. And it figures it out for me. I mean, you're a parenting columnist. How do you feel about this? You know, I think on one hand, the technology is there.

And I think that there is a real feeling that if we don't embrace it, we'll get left behind. I feel like I've read this analogy before that, like, people are worried. It might be the same way that people viewed, I don't know, cars. And people are like, what's wrong with a bike or the advent of the internet? And people are like, wait a second, what about encyclopedias? So we're like, it is technology that's making human life and parenting a little. bit easier, but it also kind of feels like we're at the precipice of maybe something bad. Or at least different, like very different than say the car.

Like one reader commented on my column that our laziness is the harbinger of our demise. Wow. It's like it's all so convenient that we just can't help ourselves and we're just like on the train to robot takeover land and we can't help it because they're writing our emails for. And how great is that? So there's this real tension here. Like you acknowledge that it is a new technology that it could have some use. It is having some use in some people's lives.

And yet the end result of that use is still very much unknown. And that's worrying. Huge question. And maybe a little worrying, you know. So why don't we get into it then? And let's talk about your experience specifically, Amberly. Tell me that story. So I had actually just been reading on my phone that, you know, there was this big news story that Bill Gates predicts that

these applications are going to replace teachers and doctors in our everyday life within 10 years. We're talking about large language models. Yeah, absolutely. The generative AI. Yeah, the technology behind chat GPT and all of that. You know, and I just thought 10 years, like that seems rather soon to be not going to a doctor. So I was kind of thinking about that. And then at the same time, I looked down and my daughter, she was almost five, she's grabbing this doll.

out of her sister's hands, and it's just a scene that I've seen, like, so many times, and it's so infuriating. You know, you've got this kid who knows better, or I think, but she just can't help herself, and she grabs the doll, and her sister's wailing, and I know kind of logically what to do, but when you see it happen, like, 300 times in a row, and you just, you crimes you down. So this time, I asked chat GPT, which, you know, I was a little bit hesitant, but I was like desperate times, you know. So what did it say? So it gave me this advice that I needed in this exact moment. It said, you're a traffic light. They just turn red and everyone knows what that means. Oh, how like, sage. I know, so profound. Traffic light. And then it added, that's your

vibe. Wow. And I was like, okay, Mr. Robot. Like, how dare you know me so well, sir? But it's exactly kind of like the energy shift. Did that make sense? Like that kind of confidence boosts that I needed. I was like, you know what? I don't need to go in with these like wild actions and like yelling at her. I just have to be like, hey, that's not okay. We're going to try that again.

And then the robot also told me that if I needed to use like going to her room as a consequence, it's not a punishment, but it's like a resetting. So if we need to have some time away from the toy to, you know, reset and reframe. So like some space. Some space. But it's not like, I'm so mad at you. And I can't deal with you doing this to your sister for the 300th time. It just gave me this like, you know, I was the only adult in the house.

But then all of a sudden, like I had another adult on my phone who was like, hey, you got this. And you're a traffic light, man. So did it work? It did. What do you mean it did, actually? Which I was like, so what happened? Like, how'd your daughter respond? She kind of understood in that moment.

Like I just said, that's not how we do things. I calmly took the doll. I calmly gave it back to her sister. And I said, you know, that's not how we do things in this house. We can take some space. And then, you know, the two of them went back to their playing kind of together, kind of separately, areas. And then, like, the sorcery of it all.

Maddie, is that it hasn't happened again. No. Actually, the robot, the chat GPT has actually, like, for now anyways, it's been, I don't know, several weeks, hasn't happened again. Wow. I've seen her ask for the toy. I've seen her ask if she wants to do a trade. Like, hey, I've got this Anna. Do you want to, or do you want to play together?

Mm-hmm. But there hasn't been this, like, sort of power move where I'm taking that from you. Huh. So, okay. Sarsery. Yeah, that is. There's some sort of mysticism happening here. How did you feel about the fact that it was artificial intelligence that gave you decent, or dare I say, very good, it seems, advice on something so human as parenting your own child? Yeah, I mean, what shocked me is that it said that's your vibe because it felt like it actually knew what I needed.

I think a lot of the advice, like obviously the traffic light thing existed on the internet. Yes. But it pulled that and knew that's what I needed. And knew how to talk to you. And it knew how to speak to me in a way that didn't feel at all robotic. Which, again, great and also terrifying. So did you have this moment when she handed the doll back and didn't freak out, didn't cry? Were you kind of stunned, like in that moment?

I don't know if stunned because, you know, I think the world of her. But I also think that it gave me time to reflect on like how my previous actions didn't really de-escalate the situation at all, right? That happens, and then it's like, there's no room for apology or there's no room to reframe it or to redo it. There's just mom coming in, like, why don't you know better? Let's go to your room and like a real, like, elevated tension, whereas I'm a traffic light and, no, I can do better.

And that I'll be the gentle guide that I think at some point in our conversation, chat GPT said be like the safe wall that's like sturdy safe and something that she can push back against that I'm not going to escalate yeah it's very matter of fact yeah as I'm listening to you kind of explain this amberly it seems like in some regard that what actually happened here is not so much that the advice worked on your daughter per se but more it seems to have worked really well with you essentially absolutely I mean the chat GPT was actually quite chatty. And in a situation where, you know, I'm home alone a lot with my kids, it felt like it was a

one-on-one experience where I was like, hey, can you help me with this? And it actually did, which was great, also spooky, but it knew me in a way that knew what I needed and the advice that would actually resonate. We'll be right back. Okay, so that's how you used it to help your daughter. Earlier, you mentioned that other moms you know are using it as well. Can you give me a sense of like what kind of questions they're asking, chat GPT? Well, more and more, I hear that it's for advice, which I don't know, may or may not be concerning.

One mom friend told me that it helped her fix her fraught relationship with her sister and has given her, like, really good advice that she's used to mend or at least maintain that. And then another mom, a new mom, just told me that it helped her get her baby on a sleep schedule that actually worked because there's a lot of flexibility that you can have when you're not just looking at like a chart online somewhere, but you're saying, oh, the baby only nap for 20 minutes, now what? Right. You know, and new moms.

Yeah, exactly. And personal to what you're looking for. So, I mean, right now, this app is everywhere. Yeah, it sounds like they're using it both in terms of their own personal life and in terms of parenting advice. But on the parenting side of things, I do wonder, though, how different this actually is because, you know, our parents had parenting books. Us millennials have Instagram reels for better or worse. And, you know, AI is, of course, just feeding off these materials to generate the advice that it's giving you. So how does this or why does this maybe feel?

different to you? Well, I think the great benefit and also maybe the danger is in the proximity, the closeness, because you open this app and it literally says, ask me anything. And you literally ask it anything. And when you do, you feel like you're getting the right answer. You know, it's a closer, more intimate experience than watching a TikTok video and then sort of thinking, huh, does that apply to me? Yes, no. This is like your bestie, you know, saying to you, it's going to be okay. And this is what we should do. And, you know, it's a little bit more personalized and close. And quick, too, right? Yeah, absolutely. You don't have to search around. You don't have to, like, if you're using an old school book, you don't have to,

like, rifle through the pages. It's just there. It's just there. And it has read that book. Yeah. Probably if that book's online. You know, it's read the entire internet. And it's matching the best advice to you. So maybe, I don't know, maybe the traffic light analogy was in some book 10 years ago or somebody posted it to Twitter a long time ago or, you know, but it knew from its reading of the entire internet that that would be helpful for me. That's very interesting point you're making because it's, you don't know the source of these things.

And so I want to talk a little bit. I know you've talked to researchers about this issue as well. So, you know, when you told them your experience, for example, how did they make sense of it? Yeah. So I spoke to two Canadian researchers who've actually just been looking at AI's use in parenting for moms. And they told me that AI is, or at least chat GPT, is just one big recycling machine. So again, it's read the internet and it's spitting out what it thinks or guessing what you want to hear.

But they told me that the danger is kind of in the type of person that's using it. So some people will get their answer, and then they'll say, okay, I want to dig a little deeper and use that information as one of many sources. But some people will just get the answer and go, oh, okay, problem solved. Right now. There's many different people. And then the other thing is that they told me that the closer you are to the margins, the less effective chat GPT will be for your question. You know, if you're asking something very specific or that it doesn't know, say my kid had. autism or was nonverbal.

Right. And it gave me this like, be the traffic light advice. Like, that's not necessarily the most useful thing. Okay. But it doesn't know that. So it can feel very specific and really useful. But as one researcher told me that, you know, it's like a high risk scenario often. Sure.

The stakes are really high here. In parenting. Or if you're asking, like, what's this rash on my kid? Or why does my kid, you know, have a list? Or why does, you know, it can be all kinds of things. And just having like a generalized recycling machine answering you is not that useful. Well, in that point, you brought up, you know, a couple health things there, like a rash or something. Are there areas where researchers say parents should not turn to chat GPT for?

I think health is the big one. I mean, they kept giving me the analogy of like WebMD. So sometimes that's a good thing and a good source to kind of check. Sometimes it can be completely off. And at least right now, I mean, with all credit to Bill Gates, like, it's not replacing medical professionals yet. And so health is that one area where you probably shouldn't just like ask it and then at least believe that that's the one and only answer. It could just be a jumping off point. Right.

Okay. So, you know, we've been talking about some of the challenges or limitations here that researchers are seen with this trend. But are there ways in which it can be useful to parents outside of, say, getting behavioral advice? Yeah. So one area that I have really been using it for is I've sort of labeled it my boring co-parent. You know, the mental load of motherhood, Maddie, is I know something I've spoken about before, but I just will never shut up about because you cannot possibly feel the weight of the mental load of parenting until you're truly under it. It's not just like the soccer signups are knowing who needs a raincoat, but it's actually just thinking about that stuff all of the time, like just always doing a scam all of the time.

And so it's really useful for chat GPT to make me a meal plan. You know, I send it a picture of my pantry and it can be like, this is what's for dinner tonight. If you want to make this for dinner tomorrow night, you're going to need these three things to go get. And it just takes a little bit of the edge off. Yeah. You know, to your point earlier, it can tell me very quickly the things that I need. Like it'll say, I just go, what toddler sandals are the best for summer and also under $40? And it's like, these are the best reviewed.

It's right here. Here's the link to buy it. Right. I don't have to worry about that so much. It doesn't take up so much of my brain. So it has really helped with the mental load a little bit. Well, yeah. On something like the meal plan example, which I think is very relatable for a lot of us, parents and non-parents, it helps a little bit with decision fatigue as well, right?

Yeah, that's exactly what it is. You know, the number of times that, you know, you have to think about, just the act of thinking about what to feed your kids is like, it's just all the time. I mean, people joke that motherhood is just basically being a snack giver outer, but it really is. and it's just helpful to have like a robot do sort of like the boring least human things about parenting. Like it's, that's useful and I appreciate it. I do think we can give mothers and parents in general a little bit more credit than that we give to a vending machine. But fair enough. It's a lot of snacks, Maddie.

I don't know what to tell you. Okay. I believe you. I wonder in like a big picture way, especially now with the benefit of time, because it's been a little bit since this happened to. Did this experience change your trust in your own instincts as a parent? I wouldn't say change. I would say underscored, you know, that I was like, you know what? I can solve a problem and it helped me solve that problem, but I sort of knew it was there

all along. But I was so frustrated with like the repetition of the toy stealing. I was just bringing this energy into it that wasn't helpful. And I can see that now because now I'm just a little bit more traffic lady, and I'm a little bit less more like, ragey mom. And I feel like, I don't know, it's helped us. So where do you think this goes from here? Like, what does the future look like in terms of parenting and generative AI? So I think it is so helpful right now.

And that's sort of the conundrum of it, because it's so good. But then, you know, there's people like the godfather of AI who, the University of Toronto Proff, Jeffrey Hinton. And he won the Nobel Prize. And then he literally had this Frankenstein moment where he was like, this could be bad. Like his quote is that a number of possible bad consequences could happen, particularly the threat of these things getting out of control. And I'm like, man, you just invented this thing. And it is really helpful for me in this period of my life.

And I know it's helpful for so many people. But you're like looking at it in a big picture way. And I don't know, it's a little bit terrifying that this is the world in which I think we'll always know now and our kids will always know. Yeah, I think navigating these unknown spaces, especially in these very human places, like parents connecting with children, young children who, you know, are still learning language, still learning emotional regulations. That is a very scary place. And then you throw in something as powerful and as unregulated currently as generative AI. And here we are. And I don't know.

I just feel like parenting, you know, it comes with a lot of headaches and challenges and a lot of hard things, but there's also so much joy. And, you know, I put all of my heart into raising these little living beings. And even if I have a little help from things that aren't human, you know, it feels honestly like parenting really is like the most human thing I can do. Well, Amberly, thank you so much for coming here and chatting with me today. Thanks, Betty. Thanks for chatting with the thoughts. That was Amberley McIteer, a contributing columnist for the globe who writes about parenting. That's it for today. I'm Madeline White. This episode was produced by Tiff Lamb.

I produced the show with Mihal Stein and Ali Graham. David Crosby edits the show. Adrian Chung is our senior producer, and Angela Pichenza is our executive editor. Thank you for listening. Thank you.