A practical, printable contract families can use to set clear expectations around how children interact with artificial intelligence at home — covering time limits, approved tools, privacy, homework honesty, and parental oversight.
A Family AI Safety Agreement is a written contract — signed by parents and children together — that defines the rules, boundaries, and expectations for using AI tools in the home. Just as families have historically set guidelines for screen time, internet use, and social media, an AI agreement extends that same principle to a technology that can generate text, images, code, and answers on demand.
Think of it as a shared understanding rather than a punitive document. The goal is to encourage open conversation, teach responsible habits, and give kids a framework for making good decisions when Mom or Dad is not looking over their shoulder. When children help write the rules, they are far more likely to follow them.
AI is no longer a future concept — it is embedded in search engines, homework helpers, messaging apps, and even games children already use daily. Without explicit boundaries, kids can inadvertently share personal information, rely on AI to do their thinking for them, or encounter content that is inaccurate or inappropriate. A written agreement turns vague “be careful online” advice into concrete, agreed-upon rules.
Children may not realise that prompts they type can be stored, logged, or used to train models. An agreement sets clear lines on what is off-limits.
Without guidance, the line between “using AI as a tutor” and “having AI do the assignment” blurs quickly. The agreement makes that line explicit.
AI can be compulsive — instant answers for every curiosity. Time limits and purpose rules prevent over-reliance and keep kids thinking independently.
The process of negotiating the agreement builds a channel for kids to come to parents when something confusing or uncomfortable happens online.
A thorough Family AI Safety Agreement addresses six core areas. Each family will weigh these differently, but every section below should appear in some form.
Specify how long AI tools may be used per day or per session, and during which hours. For example: “AI helpers may be used for up to 30 minutes on school nights, only between 4:00 PM and 7:00 PM, and only after homework is attempted independently first.”
List the specific AI tools children are allowed to use (e.g., a parent-approved chatbot, a school-sanctioned learning platform). Anything not on the list requires a parent conversation first. This prevents kids from wandering into unvetted or unmoderated tools.
Define the difference between learning from AI and copying from AI. A good rule: AI may explain a concept, quiz you, or give feedback on work you already did — but it may not write the assignment for you. If a teacher would be surprised to learn how the work was produced, it probably crosses the line.
Parents reserve the right to review AI conversations, check browser history, and discuss anything that looks concerning. This is not about spying — it is about coaching. Frame it as: “We will check in periodically because our job is to help you learn to use these tools well, not to catch you doing something wrong.”
State clearly what happens if the rules are broken. Consequences should be fair, proportional, and — most importantly — agreed upon in advance. Examples include a temporary pause on AI access, a conversation about what went wrong, or loss of a related privilege. Avoid surprises; the agreement itself is the warning.
Below is a ready-to-use sample agreement. Read through it with your child, adjust any wording to fit your family, then print and sign it together. You can also copy the text into a word processor and customise it further before printing.
Use the button below to print just the agreement, or press Ctrl/Cmd + P and choose your printer.
Our family’s promise to use AI tools responsibly, honestly, and safely.
This agreement is made on (date) between (parent/guardian) and (child).
If I break this agreement, the following will happen:
I understand that this agreement exists to keep me safe and help me build good habits with technology. I agree to the rules above and will do my best to follow them.
Child’s signature:
Parent/Guardian signature:
Second Parent/Guardian (if applicable):
The conversation matters more than the document. If you simply hand your child a printed contract and ask them to sign, you will get compliance — not understanding. Instead, treat the agreement as something you build together. Here is an approach that works:
Ask your child what AI tools they already know about, what they think AI is good at, and what worries them. Listen before you lecture. You may be surprised by how much they have already encountered.
Instead of “don’t share your address,” explain: “Some AI systems store what you type, and we don’t know who might see it later, so we keep personal details private.” Kids follow rules they understand.
Give your child real input on time limits, which tools are approved, and what consequences seem fair. If they feel ownership over the agreement, they will respect it far more than one imposed from above.
Share age-appropriate stories: an AI that gave a confidently wrong answer, a chatbot that asked for personal details, or a student who got in trouble for submitting AI-written work. Concrete examples stick.
End the conversation by saying: “If anything ever feels weird, confusing, or scary when you’re using AI, you can come to me — no matter what it is, and you will not be in trouble for telling me.” That promise is the most important part of the whole agreement.
A single template cannot fit a six-year-old and a sixteen-year-old. Adjust the agreement’s depth, language, and restrictions based on your child’s age and maturity. The table below offers a starting framework.
| Age Range | Supervision Level | Key Modifications |
|---|---|---|
| Under 8 | Always supervised; parent operates the tool | No independent AI use. Keep it playful and educational. Focus on curiosity and basic “computers can be wrong” lessons. No agreement signature needed — rules are for parents. |
| 8–11 | Supervised with short independent sessions | Simple language, short rules list (5–7 items). Only parent-approved, kid-friendly tools. No account creation by child. Parent reviews all sessions. Emphasis on privacy and “ask first.” |
| 12–14 | Periodic check-ins; more independence | Full agreement in place. Child can suggest approved tools. Introduce homework integrity rules in detail. Parent reviews weekly. Discuss misinformation and emotional manipulation risks. |
| 15–17 | Trust-based with spot checks | Agement reads more like a shared understanding than a list of rules. Address academic honesty in the context of college and career. Include guidance on AI-generated images, deepfakes, and digital reputation. Consequences focus on real-world impact. |
| 18+ (still at home) | Adult agreement; advice-based | Shift from rules to recommendations. Focus on privacy, data permanence, employment implications, and critical thinking. Keep the conversation open but remove parental enforcement. |
This page is designed to print cleanly. When you print, only the sample agreement will appear on the page — the navigation, background, and surrounding content are automatically hidden. Follow the steps below:
Want a pre-formatted PDF or editable version sent to you? We can provide customised templates tailored to your child’s age and your family’s specific concerns. See the section below.
Avondale.AI runs parent workshops on AI safety at home — practical, jargon-free sessions that walk you through building an agreement your kids will actually respect. Whether you have a curious seven-year-old or a tech-savvy teenager, get in touch to find out about the next workshop or to book a one-on-one session.
Contact Steve About Parent Workshops