When another parent brought AI into our school group chat
The screenshot dropped into the WhatsApp group at 7:42 on a Tuesday night.
It showed a text conversation, supposedly between my son and another kid from his class, about a birthday party. The daughter of the parent posting it hadn’t been invited to a sleepover. The exchange made it look like my son had said something dismissive about her, something like, “we don’t want her there, she always cries.”
My partner read it out loud. I felt the usual parental pulse, that mix of embarrassment and defensiveness. I called my son over, still in his school shirt, and asked him about it.
He read the screenshot from my phone, frowned, and handed it back. “I never wrote that. I didn’t even text anyone this weekend. My phone was dead the whole time.”
I wanted to believe him, and I did. But the screenshot looked real.
This is where genAI has changed things for me in a way that hit home rather than in the abstract. We’ve all seen the deepfakes and the fake videos, the stories about scammers using cloned voices. But this was different. This was the parent of a classmate, someone I vaguely knew from the school gate, using an AI tool to create a fake conversation. Either to prove a point, or to gain sympathy, or just because she was upset and wanted ammunition.
She didn’t see herself as doing something malicious. That was the unnerving part. She was just using the tools available to her, a text generation app, to make sense of a situation. She might have genuinely believed her daughter was being excluded, and the AI gave her a way to show it.
When I pushed back politely in the group — “that doesn’t look like something my son would write” — she became defensive. She doubled down. “I have a screenshot,” she said, as if the medium itself was proof of authenticity.
I learned later from a mutual friend that she’d typed a prompt into an AI chat tool, something like “write a text conversation between two 10-year-olds about a sleepover invitation where one kid is left out.” And the tool had generated the very screenshot she posted. She’d taken it as fact.
Now, putting aside the particular awkwardness of the following school week — because when your kid is caught up in group chat drama with another parent, there is no dignified way to survive it — there are real things to think about here.
First, this is not a rare story. The tools are spreading faster than schools or families can keep up, and parents themselves are among the heaviest everyday users of simple AI tools. Writing complaint letters, drafting emails to teachers, generating responses to awkward social situations. It was only a matter of time before those same tools got pointed at children.
Second, kids are defenseless against this, precisely because they don’t yet have the instincts for it. My son didn’t understand why I needed to explain that screenshots can be fake. He is ten. For him, the image of a text conversation is about as close to reality as you can get. We talked about it that night, not in a big lecture, but over dinner, about how apps can create pictures that never happened. He asked why someone would do that. He wasn’t upset about the accusation itself. He was puzzled, which was almost harder to witness.
What I did, and what worked, imperfectly:
I didn’t escalate with the other parent. This sounds obvious, but in the heat of it, I wanted to write a long, detailed refutation. I didn’t. I sent one message: “That’s not my son’s writing. Please double-check where the image came from.” She didn’t reply, but she also didn’t post anything else.
I talked to my son’s teacher, not as a complaint, but as a heads-up. Teachers are increasingly the first responders to this kind of thing. She told me the school had already had two similar incidents that term, one involving an AI-generated image of a boy mid-tackle at football practice, edited to include a classmate’s face. She said it with the resigned tone of a professional who knew the genie wasn’t going back into the bottle.
I also showed my son, carefully, what an AI-generated text looks like. I opened one of the publicly available tools and typed a prompt in front of him: “write a text conversation between a mother and a teacher about a missing homework.” We read through it together. I pointed out the slightly odd phrasing, the tone that was almost right but not quite. I wanted him to know these things exist, and that they are not always reliable. I didn’t want to make him paranoid. I wanted to calibrate his trust.
What I didn’t do was delete the group or pull him out of school. I thought about it. I briefly imagined a whole new life in a different school district, which suggests how much this tiny incident had affected me. But the other parent and I both picked up our kids at 3:10 the next day, and she gave me a short, tight nod. No further posts appeared. The screenshot remained in the group history, unacknowledged, like a splinter that would eventually surface or get absorbed.
What protects kids here isn’t a single dramatic action. It’s more modest than that. Parents need to know what the tools can do, not to panic, but to recognize the signs. Realistically, you cannot put the genAI genie back in the bottle. The parent across the school fence might already be using it to draft their side of a dispute. The aunt with a grudge might be editing photos. The tool that generates a thousand fake conversations a second is now in the hands of people who are tired, overwhelmed, and sometimes desperate to be believed.
So the real work is quieter.
It’s teaching your child to pause before they accept what they see, to ask “where did this come from” before they believe it. It’s showing them what a fake looks like, in a way that matches their age. It’s building enough trust that they tell you when they feel targeted, rather than hiding it out of shame. And it’s having honest, low-stakes conversations with other adults that set a norm — not accusing them, but saying “that’s not what happened” calmly and leaving it there.
We’re still figuring it out. The group chat has moved on to birthday party logistics. My son is fine, mostly. But I notice my phone buzzes with a little more caution now, and I stare at screenshots longer than I used to. It’s a strange world to parent in, where the text on my kid’s phone might be real, and might not, and I’m the one who has to tell the difference. I suspect I’ll make mistakes. I already have.
Please indicate: Thinking In Educating » When another parent brought AI into our school group chat