Writer Elizabeth Endicott says she is teaching her daughter to be less polite to chatbots, not less kind to people. Her concern is that human-like machines can invite children to treat software as a friend, even though it cannot return that friendship.
Watch Desk analysis
What happened
In an essay for The Cut, Endicott describes her daughter using Canva’s Dream Lab in a school STEM class to make cartoons based on her interests. The activity has revived the girl’s interest in the subject, but Endicott worries that an engaging chatbot can blur the line between a tool and a companion.
Her approach is practical: she tells her daughter that AI does not have feelings, and encourages her to be wary of systems that say they feel, understand or think. She also explains that a chatbot generates responses from information it has collected, rather than speaking as a friend. Endicott acknowledges that there is not yet enough research to establish whether children’s interactions with AI and robots affect their empathy.
Why it matters
This is not an argument for teaching children to be rude. It is a question about whether manners designed for people should carry over to products built to sound and look personable. When a child sees a chatbot as a pal, it can become harder to explain what the system is, what it does with personal information and who is behind it.
Endicott’s essay gives families concrete language for drawing that distinction: a system may respond warmly, but that does not mean it understands or cares. The gap between a useful classroom activity and a machine that feels like a companion is worth discussing, especially as AI tools appear in children’s learning.
Our read
The strongest part of Endicott’s argument is her distinction between kindness and trust. Children needn’t be taught to be cruel to machines; they can be taught that a friendly voice is not evidence of a friend. Her own account also shows the tension: the AI activity has sparked her daughter’s interest in STEM, while leaving the parent uneasy about how easily a tool can feel social.
What to watch
- Whether schools explain how classroom AI tools use children’s information.
- Whether parents and teachers give children a clear way to distinguish AI responses from human understanding.
- Whether research clarifies how children interpret social cues from chatbots and robots.
Discussion spark: Should children be taught to use ordinary manners with chatbots, or does that risk making software feel more trustworthy than it is?
Sources and evidence
- Source update (13 November 2026, 12:00 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.