When ChatGPT first reached Pakistani classrooms, many schools reacted the way schools always react to a new disruption: with a ban. Notices went up, warnings were issued, and teachers were told to watch for AI-written work. Within months, most of those schools quietly discovered the same thing — students were still using it, just more secretly, and teachers had no guidance for the cases they caught. A ban that cannot be enforced is worse than no policy at all, because it teaches students that school rules are decorative.
The alternative is a policy that makes sense to students, teachers, and parents alike. It does not need to be long. One page, written in plain language, explaining what students may use ChatGPT for, what they may not, how work will be checked, and what happens when rules are broken, will do more than any detection tool. This guide shows how to write that policy and teach the skills behind it.
Panic Bans vs Sensible Policy
A panic ban fails for three reasons. First, detection is unreliable: AI detectors flag innocent work and miss clever cheating, so enforcement feels random and unfair. Second, the ban ignores the reality that students will meet AI in higher education and workplaces, where they will be expected to use it well. Third, it wastes a teaching opportunity — the schools that talk openly about ChatGPT are the ones whose students learn to use it critically rather than blindly.
A sensible policy starts from a different question: what do we want students to learn? If the learning goal is understanding a concept, then using ChatGPT to explain that concept is legitimate help, like a tutor or a textbook. If the learning goal is producing an original essay or solving a problem independently, then submitting ChatGPT's output as your own is cheating. The policy should be organised around learning goals, not around the tool itself, because the tool will keep changing while the goals stay the same. Share the policy with parents on WhatsApp in both English and simple Roman Urdu, so families understand the school is guiding students rather than simply policing them.
Allowed Uses: Where ChatGPT Genuinely Helps
Start by naming the uses you encourage. Brainstorming ideas before writing: a student stuck on an essay topic can ask for ten angles and then choose one to develop independently. Explaining difficult concepts: "Explain photosynthesis like I am in Class 8" often produces a clearer explanation than a dense textbook paragraph, and the student can ask follow-up questions without embarrassment. Practising: students can ask ChatGPT to quiz them, generate practice questions on a chapter, or check whether their summary covers the key points. Language help: students learning English can ask for simpler phrasings or corrections to their own sentences — with their original draft attached, so the AI edits rather than replaces.
Each allowed use has a condition that keeps learning intact: the student must do visible thinking. For brainstorming, they keep their notes showing which ideas they rejected and why. For explanations, they write a summary in their own words afterwards. For practice, they record which questions they got wrong and what they learned. When teachers can see the process, they can trust the product. This approach also fits naturally with class participation routines — a student who used AI to understand a topic can explain it aloud in class, which is the strongest proof of learning there is.
Not Allowed: Drawing the Clear Lines
Clarity protects students. State plainly that submitting AI-generated text, answers, or code as one's own work is not allowed, and neither is using ChatGPT during tests, exams, or closed-book assessments. Also ban entering personal information about oneself or others into public chatbots, and copying AI output without checking it — because unchecked AI text may contain errors the student will be marked down for anyway.
Be specific about grey areas, because grey areas are where most disputes arise. Is it allowed to ask ChatGPT to improve an essay the student wrote? Most schools now say yes, if the student keeps both versions and can explain the changes. Is it allowed to use AI for translation? Yes for understanding, but the final Urdu or English work should be the student's own words. Write these decisions down, give examples, and revisit them each term — a policy that never changes while the technology changes monthly will quickly become irrelevant. For BISE board exam preparation, remind students explicitly that board exams allow no AI access, so exam-term practice must increasingly happen without it.
Sample Policy Line
Plagiarism Rules That Are Actually Fair
The hardest part of a ChatGPT policy is enforcement. AI detectors should never be the sole judge: they produce false positives, especially for students writing in English as a second language, and a false accusation damages trust far more than one missed case of cheating. Treat a detector flag as a reason for a conversation, not a verdict. Ask the student to explain their work, show their drafts, or solve a similar problem in class. A student who learned the material can do this easily; a student who copied it cannot.
Consequences should be educational first. For a first offence, the usual fair response is redoing the work properly with guidance, plus a conversation with parents — not a zero and a public warning. Repeat offences can escalate, but the escalation ladder should be written in the policy and applied consistently, not invented in anger after a particular case. Keep records of each case so that patterns are visible and no student is punished on a single teacher's suspicion alone. Fairness is what makes students accept a policy instead of resenting it.
Teaching AI Literacy, Not Just Rules
Rules without teaching produce compliance at best and evasion at worst. Set aside two or three lessons per term for AI literacy, adapted by age. Younger students learn that AI guesses the next word rather than knowing facts, through simple demonstrations where the chatbot gives a confidently wrong answer. Older students learn to compare AI answers with textbook sources, spot invented references, and recognise biased or culturally mismatched output. A powerful classroom exercise is to give students a flawed AI essay and ask them to find and fix every error — it teaches critical reading far better than a lecture about honesty.
Teachers need training too. A single workshop where teachers try ChatGPT themselves, test its answers against their subject knowledge, and practise the "explain it back" check will do more than a dozen circulars. Schools that invest in this training find that teachers stop fearing ChatGPT and start using it as a teaching assistant — drafting lesson materials with AI tools, generating varied questions, and modelling good AI use in front of students. That modelling is the policy's best enforcement: when students see their teacher use AI critically and transparently, they learn the standard to copy.
ChatGPT is not going away, and the schools that thrive will be the ones that teach students to use it wisely rather than hide from it. A one-page policy, fair enforcement, and a few honest lessons each term are enough to start. The goal is not to catch every cheater; it is to graduate students who know the difference between using a tool and letting the tool replace their thinking.
AI Help: Muse AI
Muse AI is Meta's personal AI assistant, available in the Muse app. A teacher can ask it to draft a plain-language classroom policy on AI use, or to rephrase a difficult topic in simpler words for students who are struggling. Join Muse with invite code LTGSWD — in the Muse app, go to Settings > Redeem token and enter the code.
Disclosure: LTGSWD is the author's Muse invite code.