Copy-ready prompt templates for studying, essay writing, exam prep, and understanding hard concepts — without doing your homework for you.
"Help me understand photosynthesis" produces a Wikipedia summary you'll forget in an hour. "Ask me questions about photosynthesis one at a time and tell me what I get wrong" triggers active retrieval — the study technique with the highest effect size in cognitive science research. The prompts below are designed around learning techniques that actually work, not passive reading.
The most effective study prompts share four qualities: (1) Socratic questioning — "Don't explain this concept; instead, ask me questions that help me figure it out myself." (2) Level calibration — state your current understanding so the AI doesn't over-explain or under-explain. (3) Active recall — ask for quizzes, flashcards, and gap-fill exercises rather than summaries you can read passively. (4) Connection requests — "Connect this idea to something I likely already know from everyday life." Passive reading of AI-generated summaries produces much weaker retention than prompts that force you to retrieve and reconstruct information.
Use AI for process, not product: (1) Outline stage — "Here is my thesis. Suggest 5 structural approaches I could take; don't write any prose." (2) Argument review — "Here is my argument. What's the strongest counterargument? What evidence would challenge my thesis?" (3) Clarity editing — "Edit for clarity only. Don't change my arguments or add new ideas. Mark every change with [edit]." (4) Citation check — paste your claims and ask which ones need a citation. Your ideas and arguments stay yours; the AI sharpens the expression and logic. This approach passes academic integrity checks because the content originates with you.
The study-guide prompt that works: (1) Paste the source material (lecture notes, textbook section, syllabus). (2) Specify format: "Create a study guide with: (a) the 10 most testable concepts as bullet definitions, (b) 5 common misconceptions and why they're wrong, (c) 10 short-answer practice questions with model answers, (d) a one-page summary diagram in text form." (3) Add "Flag any claims in the source that are contested or simplified for introductory purposes." The misconceptions section is the most valuable — knowing what students commonly get wrong is more useful than restating what's right.
Three prompts for hard concepts: (1) The analogy chain — "Explain [concept] using an analogy from everyday life. Then explain where that analogy breaks down." Knowing the limits of an analogy is as important as the analogy itself. (2) The prerequisite map — "What 5 concepts do I need to understand before this one makes sense? For each, give a one-sentence definition." (3) The rubber duck — "I think [concept] means [your current understanding]. What's wrong or incomplete about my understanding?" Correcting a wrong model is faster than building from scratch because it gives the AI a specific target to fix.
Exam prep strategy with AI: (1) Retrieval practice — "Ask me 10 questions on [topic] one at a time. After I answer each, tell me what I got right, what I missed, and the most important thing to remember." (2) Past-paper simulation — paste an old exam question and ask for a marking rubric, then evaluate your own answer against it. (3) Weak spot targeting — "I understand X and Y well but struggle with Z. Ask me 5 increasingly difficult questions on Z." (4) The night-before summary — "Summarize the 20 most important facts about [topic] in bullet form, ordered from most to least likely to appear on an exam." Avoid AI-generated notes as primary study material — use them to supplement active retrieval.