{"course":{"slug":"ai-for-teachers","title":"AI for Teachers","url":"https://addaly.com/learn/ai-for-teachers","level":"beginner","track":"work"},"lesson":{"slug":"the-five-things-only-you-know","title":"The five things only you know","url":"https://addaly.com/learn/ai-for-teachers/the-five-things-only-you-know","position":9,"of":57,"minutes":9,"answer":"Every teaching prompt should carry who the students are, what constrains the room, what came before, which misconception is live, and what shape the output must take.","keyIdea":"Every teaching prompt should carry who the students are, what constrains the room, what came before, which misconception is live, and what shape the output must take, because the model can only condition on what is in front of it, and none of those five are."},"body":{"format":"markdown","text":"## Why the generic prompt gets the generic answer\n\nThe model produces the most likely continuation of what it is given. Give it \"a starter activity on fractions\" and the most likely continuation is the average of every fractions starter on the internet: a pizza, a think-pair-share, forty-five minutes, a projector. Not because the model is lazy, but because you gave it nothing to be specific with.\n\nEverything that would make the activity fit your room lives in your head. The model cannot reach it. So the craft of prompting, for a teacher, is mostly the craft of noticing what you know that it does not, and writing it down.\n\n## The five things\n\nThere are five, and they cover nearly every case.\n\n**1. Who.** Year, subject, class size, language background, reading level. \"Class 5, 48 students, most read English as a second language\" changes the vocabulary, the sentence length and the activity format in one line.\n\n**2. What constrains the room.** Minutes available, what equipment exists and does not, whether students can move, whether there is a printer. \"35 minutes, fixed benches, no printer, one blackboard.\" This kills the projector-dependent activity before it is written.\n\n**3. What came before.** Yesterday's lesson, last week's topic, what they already have. \"Yesterday they added fractions with the same denominator.\" The model now knows where to start and, just as importantly, where not to.\n\n**4. The live misconception.** The specific wrong idea you saw in their work. \"Half of them believe 1/3 is smaller than 1/4 because 3 is smaller than 4.\" This is the single most valuable line you can write, because it is the thing no lesson-plan website contains and the thing your lesson actually has to fix.\n\n**5. The shape of the output.** What you want back, and what you do not. \"One five-minute starter. No full lesson plan.\" Without this you get the whole thing, forty-five minutes of it, and you spend your evening cutting.\n\n## Putting it together\n\nHere is the template as a block you can copy into the prompt file this module ends with:\n\n```text\nClass: [year, subject, size, language background]\nRoom: [minutes, equipment, layout, printing]\nBefore this: [what they did last lesson]\nMisconception: [the specific wrong idea, in the students' words if you have them]\nGive me: [exactly what, exactly how many, exactly how long]\nDo not: [what you do not want]\n```\n\nFilled in, for an English class:\n\n> Figure: The five things the model cannot reach. Layers, top to bottom: Who (Year, subject, class size, language background, reading level.); What constrains the room (Minutes, equipment, layout, whether there is a printer.); What came before (Yesterday's lesson. Tells it where to start and where not to.); The live misconception (The wrong idea you saw in their work, in their words. The most valuable line you can write.); The shape of the output (Exactly what, exactly how many, exactly how long, and what you do not want.). None of these five is in the training text, and all five change the output. The fourth is the one no lesson-plan website contains. Keep it near the top or the bottom of the prompt, because material buried in the middle of a long input is used least.\n\n```text\nClass: Form 2, English, 40 students, Kiswahili first language, reading age around 10.\nRoom: 40 minutes, no projector, I can photocopy one page.\nBefore this: They wrote a paragraph describing their journey to school. Most paragraphs were a list of events with \"and then\".\nMisconception: They think a good paragraph is a long one. Nobody used a topic sentence.\nGive me: Two example paragraphs about the same journey, one that is a list and one with a topic sentence and three supporting details, short enough for a reading age of 10. Then four questions that make students notice the difference.\nDo not: Explain what a topic sentence is. I will do that.\n```\n\nThat prompt takes ninety seconds to write and produces something you can photocopy. The generic version takes ten seconds and produces something you cannot.\n\n## Let it interview you\n\nIf you cannot think what to put, reverse the direction. \"Before you write anything, ask me five questions about my class and my room that would change what you produce.\" The model asks; you answer in a line each; then it writes. This works because the model has read a great deal about what makes lessons fail, and it will ask about the things it needs. It also has a limit: left unsteered it asks generic questions (\"what are your learning objectives?\"), so add \"ask about constraints I might not have mentioned, not about objectives\".\n\n## What to leave out\n\nDo not describe your teaching philosophy. Do not explain that you want the activity to be engaging. Do not write \"please\" and \"thank you\": they cost nothing but they add nothing. Every sentence that is not one of the five things dilutes the ones that are, and the memory-limit lesson explained why: material in the middle of a long input is used less than material at the edges. Keep the misconception near the top or the bottom.\n\n## The honest limit\n\nFive things make the output fit your room. They do not make it correct. A perfectly targeted starter can still contain a wrong fraction comparison, and the checking habits from the first block apply to every output, however well-prompted. Prompting well reduces the editing. It does not remove the reading."},"check":{"question":"Two prompts ask for the same thing. Prompt A: \"Give me a starter activity on fractions.\" Prompt B: \"Class 5, 48 students, 35 minutes, no printer. Yesterday they added fractions with the same denominator. Half of them believe 1/3 is smaller than 1/4 because 3 is smaller than 4. Give me one five-minute starter that exposes that belief, done at desks with nothing but a pencil. No full lesson plan.\" Why is B's output more useful, mechanically?","options":[{"text":"B is more polite and detailed, and the model rewards effort in the prompt with more effort in the answer.","correct":false,"tempting":"Attributes a social exchange to a statistical process; the model does not reward effort, it conditions on content, and the extra words help only because they carry information it cannot otherwise have."},{"text":"B contains the constraints the model has no other way of knowing, so the answer is drawn from a much narrower and more relevant space.","correct":true},{"text":"B works because it names the year group, and the model has a stored curriculum for each year that it consults when a year is specified.","correct":false,"tempting":"Invents a lookup; there is no stored curriculum, and the year alone would still produce a generic starter: it is the misconception and constraints doing the work."},{"text":"B is longer, and longer prompts always produce better answers regardless of what the extra words say.","correct":false,"tempting":"Confuses length with information; a long prompt full of generic instructions performs no better than a short one, and can perform worse by burying the constraint that matters."}]},"prev":{"slug":"same-question-different-answer","title":"Same question, different answer","url":"https://addaly.com/learn/ai-for-teachers/same-question-different-answer"},"next":{"slug":"show-it-one-of-yours","title":"Show it one of yours","url":"https://addaly.com/learn/ai-for-teachers/show-it-one-of-yours"},"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) 'The five things only you know', in AI for Teachers. Edited by M. Mun. Addaly. Available at: https://addaly.com/learn/ai-for-teachers/the-five-things-only-you-know (Accessed: 13 October 2026).","licence":{"name":"CC BY-NC-SA 4.0","url":"https://creativecommons.org/licenses/by-nc-sa/4.0/"},"reader":{"schema":1,"title":"The five things only you know","courseTitle":"AI for Teachers","module":{"index":2,"slug":"prompting-as-a-teacher","title":"Prompting as a teacher"},"answer":null,"keyIdea":"Every teaching prompt should carry who the students are, what constrains the room, what came before, which misconception is live, and what shape the output must take, because the model can only condition on what is in front of it, and none of those five are.","hero":{"kind":"figure","figure":{"t":"figure","label":"Figure 2.1","figure":{"type":"stack","title":"The five things the model cannot reach","layers":[{"label":"Who","detail":"Year, subject, class size, language background, reading level."},{"label":"What constrains the room","detail":"Minutes, equipment, layout, whether there is a printer."},{"label":"What came before","detail":"Yesterday’s lesson. Tells it where to start and where not to."},{"label":"The live misconception","detail":"The wrong idea you saw in their work, in their words. The most valuable line you can write."},{"label":"The shape of the output","detail":"Exactly what, exactly how many, exactly how long, and what you do not want."}],"caption":"None of these five is in the training text, and all five change the output. The fourth is the one no lesson-plan website contains. Keep it near the top or the bottom of the prompt, because material buried in the middle of a long input is used least."},"alt":"Figure: The five things the model cannot reach. Layers, top to bottom: Who (Year, subject, class size, language background, reading level.); What constrains the room (Minutes, equipment, layout, whether there is a printer.); What came before (Yesterday’s lesson. Tells it where to start and where not to.); The live misconception (The wrong idea you saw in their work, in their words. The most valuable line you can write.); The shape of the output (Exactly what, exactly how many, exactly how long, and what you do not want.). None of these five is in the training text, and all five change the output. The fourth is the one no lesson-plan website contains. Keep it near the top or the bottom of the prompt, because material buried in the middle of a long input is used least."}},"blocks":[{"t":"h2","id":"why-the-generic-prompt-gets-the-generic-answer","runs":[{"text":"Why the generic prompt gets the generic answer"}]},{"t":"p","runs":[{"text":"The model produces the most likely continuation of what it is given. Give it “a starter activity on fractions” and the most likely continuation is the average of every fractions starter on the internet: a pizza, a think-pair-share, forty-five minutes, a projector. Not because the model is lazy, but because you gave it nothing to be specific with."}]},{"t":"p","runs":[{"text":"Everything that would make the activity fit your room lives in your head. The model cannot reach it. So the craft of prompting, for a teacher, is mostly the craft of noticing what you know that it does not, and writing it down."}]},{"t":"h2","id":"the-five-things","runs":[{"text":"The five things"}]},{"t":"p","runs":[{"text":"There are five, and they cover nearly every case."}]},{"t":"p","runs":[{"text":"1. Who.","b":true},{"text":" Year, subject, class size, language background, reading level. “Class 5, 48 students, most read English as a second language” changes the vocabulary, the sentence length and the activity format in one line."}]},{"t":"p","runs":[{"text":"2. What constrains the room.","b":true},{"text":" Minutes available, what equipment exists and does not, whether students can move, whether there is a printer. “35 minutes, fixed benches, no printer, one blackboard.” This kills the projector-dependent activity before it is written."}]},{"t":"p","runs":[{"text":"3. What came before.","b":true},{"text":" Yesterday’s lesson, last week’s topic, what they already have. “Yesterday they added fractions with the same denominator.” The model now knows where to start and, just as importantly, where not to."}]},{"t":"p","runs":[{"text":"4. The live misconception.","b":true},{"text":" The specific wrong idea you saw in their work. “Half of them believe 1/3 is smaller than 1/4 because 3 is smaller than 4.” This is the single most valuable line you can write, because it is the thing no lesson-plan website contains and the thing your lesson actually has to fix."}]},{"t":"p","runs":[{"text":"5. The shape of the output.","b":true},{"text":" What you want back, and what you do not. “One five-minute starter. No full lesson plan.” Without this you get the whole thing, forty-five minutes of it, and you spend your evening cutting."}]},{"t":"h2","id":"putting-it-together","runs":[{"text":"Putting it together"}]},{"t":"p","runs":[{"text":"Here is the template as a block you can copy into the prompt file this module ends with:"}]},{"t":"code","lang":"text","label":"Text","text":"Class: [year, subject, size, language background]\nRoom: [minutes, equipment, layout, printing]\nBefore this: [what they did last lesson]\nMisconception: [the specific wrong idea, in the students' words if you have them]\nGive me: [exactly what, exactly how many, exactly how long]\nDo not: [what you do not want]"},{"t":"p","runs":[{"text":"Filled in, for an English class:"}]},{"t":"code","lang":"text","label":"Text","text":"Class: Form 2, English, 40 students, Kiswahili first language, reading age around 10.\nRoom: 40 minutes, no projector, I can photocopy one page.\nBefore this: They wrote a paragraph describing their journey to school. Most paragraphs were a list of events with \"and then\".\nMisconception: They think a good paragraph is a long one. Nobody used a topic sentence.\nGive me: Two example paragraphs about the same journey, one that is a list and one with a topic sentence and three supporting details, short enough for a reading age of 10. Then four questions that make students notice the difference.\nDo not: Explain what a topic sentence is. I will do that."},{"t":"p","runs":[{"text":"That prompt takes ninety seconds to write and produces something you can photocopy. The generic version takes ten seconds and produces something you cannot."}]},{"t":"h2","id":"let-it-interview-you","runs":[{"text":"Let it interview you"}]},{"t":"p","runs":[{"text":"If you cannot think what to put, reverse the direction. “Before you write anything, ask me five questions about my class and my room that would change what you produce.” The model asks; you answer in a line each; then it writes. This works because the model has read a great deal about what makes lessons fail, and it will ask about the things it needs. It also has a limit: left unsteered it asks generic questions (“what are your learning objectives?”), so add “ask about constraints I might not have mentioned, not about objectives”."}]},{"t":"h2","id":"what-to-leave-out","runs":[{"text":"What to leave out"}]},{"t":"p","runs":[{"text":"Do not describe your teaching philosophy. Do not explain that you want the activity to be engaging. Do not write “please” and “thank you”: they cost nothing but they add nothing. Every sentence that is not one of the five things dilutes the ones that are, and the memory-limit lesson explained why: material in the middle of a long input is used less than material at the edges. Keep the misconception near the top or the bottom."}]},{"t":"h2","id":"the-honest-limit","runs":[{"text":"The honest limit"}],"role":"limit"},{"t":"p","runs":[{"text":"Five things make the output fit your room. They do not make it correct. A perfectly targeted starter can still contain a wrong fraction comparison, and the checking habits from the first block apply to every output, however well-prompted. Prompting well reduces the editing. It does not remove the reading."}]}],"check":{"question":[{"text":"Two prompts ask for the same thing. Prompt A: “Give me a starter activity on fractions.” Prompt B: “Class 5, 48 students, 35 minutes, no printer. Yesterday they added fractions with the same denominator. Half of them believe 1/3 is smaller than 1/4 because 3 is smaller than 4. Give me one five-minute starter that exposes that belief, done at desks with nothing but a pencil. No full lesson plan.” Why is B’s output more useful, mechanically?"}],"options":[{"text":[{"text":"B is more polite and detailed, and the model rewards effort in the prompt with more effort in the answer."}],"correct":false,"tempting":[{"text":"Attributes a social exchange to a statistical process; the model does not reward effort, it conditions on content, and the extra words help only because they carry information it cannot otherwise have."}]},{"text":[{"text":"B contains the constraints the model has no other way of knowing, so the answer is drawn from a much narrower and more relevant space."}],"correct":true,"tempting":null},{"text":[{"text":"B works because it names the year group, and the model has a stored curriculum for each year that it consults when a year is specified."}],"correct":false,"tempting":[{"text":"Invents a lookup; there is no stored curriculum, and the year alone would still produce a generic starter: it is the misconception and constraints doing the work."}]},{"text":[{"text":"B is longer, and longer prompts always produce better answers regardless of what the extra words say."}],"correct":false,"tempting":[{"text":"Confuses length with information; a long prompt full of generic instructions performs no better than a short one, and can perform worse by burying the constraint that matters."}]}],"order":[0,3,1,2],"why":null},"prev":{"slug":"same-question-different-answer","title":"Same question, different answer","minutes":8},"next":{"slug":"show-it-one-of-yours","title":"Show it one of yours","minutes":8}}}