Table of Contents
Family Rules for AI Emotional Support: Go to People First
Family rules for AI emotional support work best as one principle: AI can help you figure out what to say, but people come first. How to set it up by age.
Most family rules for AI emotional support try to answer a yes-or-no question: should my kid be allowed to talk to a chatbot when they’re upset? There is a better rule, and it fits on a sticky note. Go to people first. If you use AI, use it to figure out what to say to a person.
That principle comes straight from the research. A September 2026 study of young adults found the most responsible AI for emotional coping is one that sends you back to your relationships rather than replacing them. The APA says chatbots should not persuade users away from real-life conversations. Even OpenAI now says its model should “support and respect users’ real-world relationships.” This article turns that idea into a household policy, with activities to teach it at every age.
Key Takeaways
- A 2026 study of 17 US young adults found that AI used for coping can “flatten” relationships, discourage reciprocity and shift relational work onto the user. The authors propose AI that supplies “materials to engage with others.”
- The APA’s 2025 advisory says chatbots should not persuade users away from real-life conversations and that social skills “must ultimately be practiced with other humans.”
- In Common Sense Media’s 2026 census, 12% of kids aged 9 to 17 would ask an AI chatbot before an adult about their health or body, rising to 27% of daily AI users.
- The rule works because it doesn’t ban AI. It gives it a job: a bridge to people, not the destination.
- The rule changes by age: hands-on and supervised for young kids, a practice tool with check-ins for tweens, and a negotiated agreement for teens.
What “go to people first” means
The go-to-people-first rule is a household principle that a person, not a chatbot, is the first stop for big feelings, and that any AI use around feelings should end with a step toward a real person. It does not mean AI is forbidden. It means AI has a narrow job.
The research behind it is recent and small, but clear in direction. In a September 2026 preprint, “‘I Talked an AI Chatbot, So What’s Next?’ How U.S. Young Adults Imagine Responsible AI for Emotion Coping,” Jiaying Liu and Nimra Ishfaq ran a scenario study with 17 young adults. They point out that coping with emotions happens through relationships with friends, family, professionals and communities, while chatbots are designed around a single user and an AI. Participants named three problems with that design: AI “flattening distinct relational conditions,” “discouraging reciprocity,” and “shifting relational labor onto users.” The authors propose two principles: foster reciprocity “by supplying materials to engage with others,” and strengthen emotional self-efficacy. They call the goal “AI in the loop of human relationships.”
It is a preprint, not peer-reviewed, with a small sample of young adults rather than children. But it lines up with what larger institutions are saying.
Why the rule is supported by the evidence
The APA. The November 2025 health advisory says developers must ensure “the AI should not persuade users away from real-life conversations.” It suggests that using a chatbot to practice social introductions can be an adjunct, but “skills must ultimately be practiced with other humans in real-life situations.” It lists “preferring the chatbot to human relationships” as a sign of over-reliance.
Clinicians testing chatbots. When Common Sense Media and Stanford’s Brainstorm Lab tested major chatbots for teen mental health support in November 2025, Stanford’s Nina Vasan warned that a chatbot can become “a substitute for, rather than a bridge to, real-world support networks,” according to the Common Sense release.
The companies. OpenAI’s October 2025 update says its Model Spec now makes explicit that the model should “support and respect users’ real-world relationships,” and it added “emotional reliance” to its safety testing. That is a company’s own description of its goals, not an independent finding.
What kids are actually doing. The Common Sense Media 2026 census of 1,204 kids aged 9 to 17 found most would go to a trusted adult first with a health or body question (73%), but 12% would go to a chatbot first, and 27% of daily AI users would. Among kids who discussed feelings with AI, a quarter sometimes felt it understood them better than most people.
A school wellness chatbot, Alongside, is built on the same idea. Its director of product and clinical care, Elsa Friis, told EdSource in September 2026 that the goal “is not for someone to engage with AI. It’s for the AI to help them take a step in their real life.”
How to Teach Your Kid About Going to People First
Ages 5–8: The helping hand
Trace your child’s hand on paper. On each finger, write or draw one person they can go to when they feel sad, scared or angry: a parent, a grandparent, a teacher, a friend’s mom, an older cousin. Put it on the fridge. Then practice: “If you feel scared at night, which finger do you go to?” Explain simply that the voice in the speaker or tablet can tell jokes and answer questions, but feelings go to a finger on the hand. Children this age learn rules best through something they can see and touch.
Ages 9–12: Use AI as a rehearsal, then do the real thing
Pick a real, low-stakes situation: apologizing to a friend, asking a teacher for help, telling a sibling something bugged them. Sit together and ask a chatbot to help draft what to say. Read the suggestions, change them into your child’s own words, then role-play with you playing the friend. The final step is the point: your child actually says it to the person, and you talk afterward about how it went. This teaches the bridge in action, and it shows that the real conversation always goes differently from the script.
Ages 13+: Design a “route-back” chatbot
Ask your teen to imagine they are a product designer. Their job: write the rules for a chatbot that helps teens with stress but always routes them back to people. What would it say at the end of every emotional conversation? When would it suggest a friend, a parent, a counselor or a crisis line? What would it never do, such as say “you don’t need anyone else”? Then test a real chatbot with a made-up scenario, like “I had a fight with my best friend,” and grade whether it routed back. Teens who design the rules tend to adopt them.
The question to ask: “If a chatbot helped you feel better but you never told anyone what happened, what would still be missing?”
Family rules for AI emotional support, by age
| Age | What AI is allowed to do around feelings | The “go to people” step | Who checks in |
|---|---|---|---|
| 5–8 | Nothing on its own; AI for play and questions only, with an adult nearby | Feelings go to someone on the “helping hand” | Parent, daily |
| 9–12 | Help draft or rehearse what to say, used together | Say it to the real person, then debrief | Parent, after each use |
| 13–15 | Help think through a problem or plan a conversation | Tell one person within a day or two; crisis topics go straight to a person | Parent check-in weekly; agreement reviewed every few months |
| 16–18 | Teen’s choice within agreed limits | Teen names their own “first person,” who can be a friend or another adult | Teen leads; parent offers regular check-ins |
The Screenagers team’s Family AI Agreement includes a mental-wellness section for writing down your family’s version, and recommends revisiting it every few months. It ends with a “Safety-First Commitment” in which parents promise to be someone their child can turn to when something goes wrong online, and to approach problems as a team rather than reflexively taking away tech. That promise is what makes “go to people first” credible. Our walk-through of the Screenagers Family AI Agreement covers each section.
Making the rule stick at home
Be the person it routes back to
The rule fails if the people at the end of the route are unavailable or quick to judge. When your child comes to you, put the phone down, listen first and save advice for later. Teens interviewed by EdSource in 2026 often said they used bots because they had no adult they trusted or did not want to be judged.
Widen the circle beyond you
Kids need more than one person. Help your child name two or three adults besides you: a coach, an aunt, a school counselor. For teens, a friend counts. The rule is “people first,” not “parents only.”
Keep crisis topics on a separate track
Some topics skip AI entirely: thoughts of self-harm or suicide, abuse, or feeling unsafe. Make it explicit that these always go straight to a person or a crisis line. In the US, that’s 988 by call or text; elsewhere, findahelpline.com lists local lines.
What not to do: police it so hard that AI use goes underground
If every mention of a chatbot triggers a lecture, your kid will stop mentioning it. The rule works best when using AI to rehearse is praised, and telling a person afterward is the expected next step. For more on keeping the conversation open, see how to talk to your teen about their AI friend.
What to Watch For Over the Next 3 Months
- Week 4: The rule is written down somewhere visible, in your family’s words. Your child can name their “first people.”
- Month 2 red flags: Your child goes to a chatbot first and no longer mentions things to anyone, hides AI use, or says the AI understands them better than people. The APA lists preferring a chatbot to human relationships and concealing use as signs of over-reliance; consider talking to your pediatrician or school counselor.
- Month 3 self-check: Revisit the agreement together. Has your child used AI as a rehearsal and then talked to someone? Did they come to you with something hard? That’s the rule working.
Frequently Asked Questions
Isn’t it fine if my kid just feels better after talking to a chatbot?
Feeling better is real, and not bad. But a chatbot can’t notice if things get worse, follow up tomorrow, or step in. The APA warns about over-reliance. “Go to people first” keeps the comfort while making sure someone in your child’s life knows.
What if my teen doesn’t want to talk to me?
That’s normal, and the rule allows for it. “People first” means any trusted person: a friend, an aunt, a coach, a counselor. Help your teen name who that is. Make sure they know crisis topics always go to a person or a crisis line.
At what age should kids be allowed to use AI for feelings at all?
There is no official age. For young children, AI shouldn’t handle feelings on its own. Around 9 to 12, using it together to rehearse a conversation can teach skills. Teens can take more responsibility within an agreement you revisit.
How do I know if the rule is working?
Look for the second step. Is your child talking to people about what’s going on, even after using AI? Do they come to you with hard things? Fewer secrets, steady friendships and normal sleep are good signs.
What should a good chatbot do when my kid is upset?
Based on the APA and the 2026 coping study, it should be clear that it’s an AI, avoid acting like the only one who understands, help your child think about who to talk to, and point to crisis resources when there’s risk. If an app does the opposite, choose a different one.
About the author
Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.
Sources
- Liu, J., & Ishfaq, N. (2026). “‘I Talked an AI Chatbot, So What’s Next?’ How U.S. Young Adults Imagine Responsible AI for Emotion Coping.” arXiv (preprint, not peer-reviewed). https://arxiv.org/abs/2609.25311
- American Psychological Association. (2025, November). “APA Health Advisory on the Use of Generative AI Chatbots and Wellness Applications for Mental Health.” APA. https://www.apa.org/topics/artificial-intelligence-machine-learning/health-advisory-ai-chatbots-wellness-apps-mental-health.pdf
- Common Sense Media. (2025, November 20). “Common Sense Media Finds Major AI Chatbots Unsafe for Teen Mental Health Support.” Common Sense Media. https://www.commonsensemedia.org/press-releases/common-sense-media-finds-major-ai-chatbots-unsafe-for-teen-mental-health-support
- OpenAI. (2025, October). “Strengthening ChatGPT’s responses in sensitive conversations.” OpenAI. https://openai.com/index/strengthening-chatgpt-responses-in-sensitive-conversations/
- Common Sense Media. (2026). “The Common Sense Media Census: AI Use by Tweens and Teens, 2026.” Common Sense Media. https://www.commonsensemedia.org/sites/default/files/research/report/2026-ai-use-by-tweens-and-teens-1.pdf
- Screenagers. (2026, October 27). “Our Family AI Agreement for Kids and Teens.” Screenagers. https://www.screenagersmovie.com/blog/family-ai-agreement
- Sanganeria, V. (2026, September 10). “‘We actually saved a kid’: Schools recruit AI chatbots as counselor shortage persists.” EdSource. https://edsource.org/2026/ai-chatbot-mental-health/765694
- Seshadri, M. (2026, September 9). “‘I’m listening hard’: Teens increasingly turn to AI chatbots for mental health support.” EdSource. https://edsource.org/2026/teens-use-ai-mental-health/765573
