{"course":{"slug":"ai-for-teachers","title":"AI for Teachers","url":"https://addaly.com/learn/ai-for-teachers","level":"beginner","track":"work"},"lesson":{"slug":"planning-with-ai","title":"Planning with AI, keeping the judgement","url":"https://addaly.com/learn/ai-for-teachers/planning-with-ai","position":17,"of":57,"minutes":8,"answer":"AI supplies breadth; you supply the class. Ask it for misconceptions, not for whole lesson plans.","keyIdea":"AI supplies breadth; you supply the class. Ask it for misconceptions, not for whole lesson plans."},"body":{"format":"markdown","text":"## The division of labour that works\n\nYou decide what the lesson is for. It generates raw material. Those are different jobs and only one of them can be handed over.\n\nYou know your Grade 8 class believes plants take food from the soil. You know Fatima finishes in four minutes and needs somewhere to go next. You know the bell goes at 11:40 and the last five minutes are lost to noise. None of that is in the model. All of it determines the lesson.\n\n## Put your constraints in the prompt\n\nWeak: \"Make me a lesson plan on photosynthesis for Grade 7.\"\n\nYou will get something generic, forty-five minutes long, shaped around a group activity you have no room to run.\n\nBetter:\n\n> I teach Grade 7 science in a government school. 52 students, fixed benches, 40 minutes, one blackboard, no lab, no projector. Most students read English as a second language. Last lesson they learned that plants make their own food. The misconception I need to break: they believe plants take food from the soil through the roots. Give me three different ways to open this lesson, each under six minutes, each using something students can see or do at their desk. Do not give me a full lesson plan.\n\nThe difference is not politeness. The second prompt contains the parts only you know. And note the last line: asking for one component instead of the whole thing keeps you doing the assembly, which is where the judgement lives.\n\n## Ask it for what you are worst at\n\nMost teachers ask for a full plan, which is the thing they are already good at producing. That is the wrong request.\n\nWhat it helps with, because it draws on far more written teaching than one person can read:\n\n- **Likely wrong answers.** \"What will Grade 7 students get wrong here, and what are they thinking when they get it wrong?\" You will get eight, and one or two will be errors you had not seen in fifteen years of teaching.\n- **Analogies.** Ask for six, reject five. The sixth is often worth the whole exercise.\n- **Questions that expose the misconception.** Not questions that test the content: questions where a student holding the wrong idea gives a visibly different answer from one holding the right idea.\n- **A fourth explanation.** When you have explained it three ways and Amara still does not have it, another framing at nine in the evening is worth a lot.\n\n## What it gets wrong every single time\n\n**Timing.** It writes 55 minutes of activity for a 40 minute period. Cut a third before you look at anything else.\n\n**Checking.** It over-produces activities and under-produces the moments where you find out whether anything landed. Add those yourself.\n\n**Invented specifics.** It will recommend a video by title, a page in a textbook, a named study. Some of them do not exist. Anything with a name, a number or a date gets checked before it reaches students.\n\n**Sameness.** Left alone it puts think-pair-share into everything, including lessons where nobody can turn around.\n\n## The two questions before you use it\n\n1. What is each part for? If you cannot say why the third activity is there, delete it.\n2. If the lesson runs ten minutes short, what do you cut?\n\nAnswer both and it is your plan and the machine helped. Fail either and you are holding a script. Scripts collapse the moment a student asks something unexpected, which is about four minutes in."},"check":{"question":"Two teachers plan the same lesson. Rahel asks for a complete lesson plan and edits it. Nomsa asks for twelve likely student misconceptions and six analogies, then writes the plan herself. Whose lesson is more likely to improve, and why?","options":[{"text":"Nomsa's, because misconceptions and analogies are where reading far more than one person could genuinely adds something, while the sequence and timing depend on knowing this class, which is the part the model cannot supply and produces most blandly.","correct":true},{"text":"Rahel's, because starting from a complete draft saves the most time, and editing is where a teacher's judgement enters anyway.","correct":false,"tempting":"Editing feels like judgement, but you almost always accept the structure you were handed and only adjust surface detail."},{"text":"Neither, because plan quality follows from the teacher's own subject knowledge and no tool changes that.","correct":false,"tempting":"Treats subject knowledge as the only input, ignoring that knowing likely wrong answers is separate knowledge that can be supplied."},{"text":"Nomsa's, but only because generated lesson plans contain factual errors that survive editing.","correct":false,"tempting":"Locates the problem in accuracy rather than in which parts of planning actually benefit from breadth."}]},"prev":{"slug":"keep-a-prompt-file","title":"Keep a prompt file","url":"https://addaly.com/learn/ai-for-teachers/keep-a-prompt-file"},"next":{"slug":"worksheets-and-differentiation","title":"Worksheets, and three versions of them","url":"https://addaly.com/learn/ai-for-teachers/worksheets-and-differentiation"},"authorship":"Founder and series editor: Dr Mustafa Mun. Written by AI models to a written standard. Exam keys checked blind by a second model: 3,665 of 3,666 agreed. Academic review invited.","review":{"status":"Draft, open for academic review"},"cite":"Addaly (2026) 'Planning with AI, keeping the judgement', in AI for Teachers. Edited by M. Mun. Addaly. Available at: https://addaly.com/learn/ai-for-teachers/planning-with-ai (Accessed: 31 October 2026).","licence":{"name":"CC BY-NC-SA 4.0","url":"https://creativecommons.org/licenses/by-nc-sa/4.0/"},"reader":{"schema":1,"title":"Planning with AI, keeping the judgement","courseTitle":"AI for Teachers","module":{"index":3,"slug":"planning-and-materials","title":"Planning and making materials"},"answer":null,"keyIdea":"AI supplies breadth; you supply the class. Ask it for misconceptions, not for whole lesson plans.","hero":{"kind":"plate","image":"https://addaly.com/art/modules/ai-for-teachers/planning-and-materials.webp","alt":"A school staff room: a long table with stacked worksheets, a duplicating tray, a box of chalk and a steel tiffin.","caption":"The staff room where a term of materials actually gets made, because this module is about giving evenings back without handing over the judgement."},"blocks":[{"t":"h2","id":"the-division-of-labour-that-works","runs":[{"text":"The division of labour that works"}]},{"t":"p","runs":[{"text":"You decide what the lesson is for. It generates raw material. Those are different jobs and only one of them can be handed over."}]},{"t":"p","runs":[{"text":"You know your Grade 8 class believes plants take food from the soil. You know Fatima finishes in four minutes and needs somewhere to go next. You know the bell goes at 11:40 and the last five minutes are lost to noise. None of that is in the model. All of it determines the lesson."}]},{"t":"h2","id":"put-your-constraints-in-the-prompt","runs":[{"text":"Put your constraints in the prompt"}]},{"t":"p","runs":[{"text":"Weak: “Make me a lesson plan on photosynthesis for Grade 7.”"}]},{"t":"p","runs":[{"text":"You will get something generic, forty-five minutes long, shaped around a group activity you have no room to run."}]},{"t":"p","runs":[{"text":"Better:"}]},{"t":"quote","runs":[{"text":"I teach Grade 7 science in a government school. 52 students, fixed benches, 40 minutes, one blackboard, no lab, no projector. Most students read English as a second language. Last lesson they learned that plants make their own food. The misconception I need to break: they believe plants take food from the soil through the roots. Give me three different ways to open this lesson, each under six minutes, each using something students can see or do at their desk. Do not give me a full lesson plan."}]},{"t":"p","runs":[{"text":"The difference is not politeness. The second prompt contains the parts only you know. And note the last line: asking for one component instead of the whole thing keeps you doing the assembly, which is where the judgement lives."}]},{"t":"h2","id":"ask-it-for-what-you-are-worst-at","runs":[{"text":"Ask it for what you are worst at"}]},{"t":"p","runs":[{"text":"Most teachers ask for a full plan, which is the thing they are already good at producing. That is the wrong request."}]},{"t":"p","runs":[{"text":"What it helps with, because it draws on far more written teaching than one person can read:"}]},{"t":"ul","items":[[{"text":"Likely wrong answers.","b":true},{"text":" “What will Grade 7 students get wrong here, and what are they thinking when they get it wrong?” You will get eight, and one or two will be errors you had not seen in fifteen years of teaching."}],[{"text":"Analogies.","b":true},{"text":" Ask for six, reject five. The sixth is often worth the whole exercise."}],[{"text":"Questions that expose the misconception.","b":true},{"text":" Not questions that test the content: questions where a student holding the wrong idea gives a visibly different answer from one holding the right idea."}],[{"text":"A fourth explanation.","b":true},{"text":" When you have explained it three ways and Amara still does not have it, another framing at nine in the evening is worth a lot."}]]},{"t":"h2","id":"what-it-gets-wrong-every-single-time","runs":[{"text":"What it gets wrong every single time"}]},{"t":"p","runs":[{"text":"Timing.","b":true},{"text":" It writes 55 minutes of activity for a 40 minute period. Cut a third before you look at anything else."}]},{"t":"p","runs":[{"text":"Checking.","b":true},{"text":" It over-produces activities and under-produces the moments where you find out whether anything landed. Add those yourself."}]},{"t":"p","runs":[{"text":"Invented specifics.","b":true},{"text":" It will recommend a video by title, a page in a textbook, a named study. Some of them do not exist. Anything with a name, a number or a date gets checked before it reaches students."}]},{"t":"p","runs":[{"text":"Sameness.","b":true},{"text":" Left alone it puts think-pair-share into everything, including lessons where nobody can turn around."}]},{"t":"h2","id":"the-two-questions-before-you-use-it","runs":[{"text":"The two questions before you use it"}]},{"t":"ol","start":1,"items":[[{"text":"What is each part for? If you cannot say why the third activity is there, delete it."}],[{"text":"If the lesson runs ten minutes short, what do you cut?"}]]},{"t":"p","runs":[{"text":"Answer both and it is your plan and the machine helped. Fail either and you are holding a script. Scripts collapse the moment a student asks something unexpected, which is about four minutes in."}]}],"check":{"question":[{"text":"Two teachers plan the same lesson. Rahel asks for a complete lesson plan and edits it. Nomsa asks for twelve likely student misconceptions and six analogies, then writes the plan herself. Whose lesson is more likely to improve, and why?"}],"options":[{"text":[{"text":"Nomsa’s, because misconceptions and analogies are where reading far more than one person could genuinely adds something, while the sequence and timing depend on knowing this class, which is the part the model cannot supply and produces most blandly."}],"correct":true,"tempting":null},{"text":[{"text":"Rahel’s, because starting from a complete draft saves the most time, and editing is where a teacher’s judgement enters anyway."}],"correct":false,"tempting":[{"text":"Editing feels like judgement, but you almost always accept the structure you were handed and only adjust surface detail."}]},{"text":[{"text":"Neither, because plan quality follows from the teacher’s own subject knowledge and no tool changes that."}],"correct":false,"tempting":[{"text":"Treats subject knowledge as the only input, ignoring that knowing likely wrong answers is separate knowledge that can be supplied."}]},{"text":[{"text":"Nomsa’s, but only because generated lesson plans contain factual errors that survive editing."}],"correct":false,"tempting":[{"text":"Locates the problem in accuracy rather than in which parts of planning actually benefit from breadth."}]}],"order":[2,3,1,0],"why":null},"prev":{"slug":"keep-a-prompt-file","title":"Keep a prompt file","minutes":8},"next":{"slug":"worksheets-and-differentiation","title":"Worksheets, and three versions of them","minutes":8}}}