Why an agreement works better than a ban
Many families feel unsure about AI chatbots and writing tools. Some ban them outright; others ignore them. Neither approach teaches much. Teenagers will meet these tools at school, in future work and on their phones, and they need habits that hold up when no adult is watching. A household agreement aims to build those habits by talking openly about what the tools are good for and where they go wrong.
An agreement is also fairer than a list of rules handed down. When a teenager helps write it, they are more likely to understand the reasons and to follow it. It gives the family a shared vocabulary for later conversations, so a question such as did you check that can refer back to something everyone has already discussed.
Starting the conversation with curiosity
Begin by asking how the teenager already uses AI tools, without judgement. Which tools, for what tasks, and what have they noticed about the answers? Parents can share their own uses and doubts too. Many teenagers know more about the tools than their parents, and many parents know more about how information can mislead. Both kinds of knowledge are useful.
Try one small activity together. Ask a chatbot a question about something the family knows well, such as a local place or a hobby, and look at the answer side by side. Notice what is correct, what is vague and what is simply wrong. A shared example makes the later agreement concrete rather than abstract.
Following the school's rules first
The household agreement should sit under the school's own policy, not compete with it. Ask the teenager to find and read the school's guidance on AI use, and on each assignment, the teacher's instructions. Some tasks may allow AI for brainstorming; others may forbid it entirely. The family agreement should say plainly that the school's rule wins whenever it applies.
This also models an important habit: checking what the rules are before acting. If the guidance is unclear, the agreement can say that the teenager will ask the teacher rather than guess. That habit protects the student and shows respect for the teacher's judgement about what the assignment is meant to teach.
Writing down helpful and unhelpful uses
Make two short lists together. Helpful uses might include asking for an explanation of a concept in simpler words, generating practice questions, or getting feedback on the structure of a draft the student wrote. Unhelpful uses might include submitting generated text as one's own, using a chatbot as the only source for a fact, or asking it for medical, legal or personal advice without checking with a qualified person.
Keep the lists specific to your household and revisit them. The point is not to cover every case but to make the reasoning visible. When a new tool or situation appears, the family can ask which list it belongs on and why, which is itself a useful AI literacy exercise.
A simple checking habit
The agreement should include one checking habit that everyone, including parents, uses. For example: before relying on an AI answer for anything important, find one independent source that confirms it, and note where it came from. If no source can be found, treat the answer as unverified and say so.
Practise the habit on everyday questions, such as opening hours, a historical date or a recipe substitution. Small, low-stakes checks build the reflex. Over time, teenagers start to notice which kinds of questions AI tools answer well and which they answer confidently but wrongly.
Privacy and personal information
Include a short privacy section. Agree what should not be typed into AI tools: full names and addresses, school details, photos of other people, passwords and private family matters. Explain that what is entered may be stored or used by the service, depending on its terms, and that the family cannot always know how.
Encourage the teenager to look at the settings and terms of the tools they use, and to talk about anything surprising. This turns privacy from a vague warning into a practical skill: knowing where information goes and choosing what to share.
Time, wellbeing and the human questions
AI tools can be absorbing, and some are designed to feel like a friend. The agreement can include a reminder that personal worries, health questions and difficult feelings are best discussed with trusted people, and that the family is open to those conversations. It can also include general screen-time habits that apply to all devices.
Avoid framing this as fear. The goal is balance: the tools can be useful, and people remain the right source for care, judgement and support. Saying this plainly helps teenagers place AI tools in perspective.
Two short examples of agreement lines
Concrete wording helps. One family might write: we use AI tools to explain ideas and generate practice questions, but schoolwork we submit is written by us unless the teacher says otherwise. Another might write: before we share an AI answer with someone else, we check it against one other source and say where the answer came from.
Neither line is perfect, and that is the point. Writing a line, testing it for a few weeks and then revising it teaches teenagers that rules about technology are judgements made by people, which they can question, improve and explain.
Signing, reviewing and updating
Keep the agreement to one page, sign it together and put it somewhere visible. Agree a date to review it, perhaps at the start of each school term. Tools change quickly, and so do teenagers' needs, so the agreement should change too.
At each review, ask what worked, what was ignored and what new situations came up. If a rule was broken, focus on what was learned rather than punishment alone. A household agreement is a teaching tool. Its success is measured by the habits a teenager carries into school and beyond, not by perfect compliance.