Mobicip interview series for Safer Internet Day 2026 · Part 7 of 8
What Neuroscience Tells Us About AI
Watch the recording on YouTube
Summary
Anitha asks how technology is shaping a child’s brain and behavior. I say chatbots are still new for children, but AI has long run the recommendation algorithms on platforms like YouTube, TikTok and Instagram, and what scares me most is children giving up their sense of agency over their own thinking.
Key points
- It’s easy for AI to become a crutch, because its answer will always sound better than what a child could produce by doing a little research and writing a paragraph.
- If a child asks Alexa instead of a parent, it can weaken the bond with the parent and the parent’s standing as a source of knowledge.
- As a parent, I’m especially worried about generative AI producing media while a child watches. There is no episode to finish, so “come to dinner when the episode’s over” stops working.
Speakers
- John Zoltner Guest
- Anitha Swaminathan Interviewer, Mobicip
Transcript
The wording is as spoken, with punctuation and obvious errors corrected. See “How this transcript was made” above.
[00:00:00]Anitha Swaminathan: So since your current work also focuses on developmental neuroscience, how do you think the technology is shaping the child’s brain and behavior?
[00:00:23]John Zoltner: AI chatbots used generally by children around the world are relatively new, but of course AI was running the algorithms that children used for a lot of social media platforms, or platforms like YouTube and TikTok and Instagram. And we know that children’s attention spans have diminished over, I think it was, from 2004 until 2024 (editor’s note: should be [2016]). There was one study that showed a diminishment from almost two minutes to 47 seconds to pay attention to a specific topic. But yeah, so we’re seeing children’s attention spans, ability to concentrate, being affected. And what really scares me about AI is that of children kind of giving up the feeling that the agency around thinking. So if you think about a child growing up and from a young age using technology to ask a question and receive a response, in a way that’s fine and it could be a tool that can teach and they could learn a lot from that. But it’s very easy for that then to become a crutch, to become, you know, the answer you’ll get from the AI will always be better than or sound better than what you’ll come up with from doing a little research and writing a paragraph. So very easy to just turn to the AI because you know it’s much better. Also with parents, I think I was having a discussion with a friend the other day about when you have a young child, the child is constantly asking questions and that’s part of, you know, the developmental phase they’re going through, until they hit kind of young teenage years when they start asking questions of their friends or monitoring their social environment more for those answers. But the parents’ response to those questions won’t be as accurate and as well presented as an AI’s response. So why ask your mom and dad something when you can ask Alexa? And if that’s happening, I think that it reduces a bond that you have with your parents. It reduces your faith in your parents as, you know, a source of truth or knowledge. And maybe the parent was never the best choice of truth or knowledge, but it’s very important to have that interaction, that and that dynamic with a parent as you’re growing up. So I think that somehow takes away or can take away from the relationship and the authority of the parental figure. And also, as I was, you know, one of the things as I was thinking about creating AIChildSafety.org that really affected me was thinking about, I’m a parent, it might be obvious, I have a seven-year-old now, and you know it’s pretty common experience to be, say, your child’s watching a cartoon, watching Netflix or an episode, and you say, okay, when that episode’s over, come to the dinner table, we’re going to eat dinner. With generative AI very soon it won’t make sense to say episode, when the episode is over, because the media will be generated as the child is watching. So there will be no end to the episode. There will be, if it’s on and the child is watching, it’s going and it’s creating based on some input that it received from the children and some input from millions of other children and keep creating that storyline. So there’s not a season of 12 episodes that are dropped that you might binge during that season. There’s an always constantly running season of a story that, you know, is there, and you know that if you go and you turn on the iPad it’s there and it’s going to run from where you dropped off and it’s going to run until someone makes you close that iPad. And that, I think, is really dangerous.