AI POWERS
Examples of better questions to ask AI
Five practical before-and-after examples for learning, comparing, planning, researching and creating with AI, with a check for each answer.
How to adapt an example
A useful question gives the AI a task, a little relevant context and a result you can inspect. The examples below are starting points: replace the situation, audience and constraints with your own.
Do not paste personal or confidential details just to make a prompt sound specific. If a real name or document is unnecessary, use a fictional version.
Learn a difficult idea
Too broad: “Explain statistics.” Better: “I am new to statistics. Explain the difference between correlation and causation using a simple everyday example. Then give me two short questions to check whether I understood.”
Check whether the example actually illustrates both concepts. If a definition or number matters for your course, compare it with the teaching material or an original source.
Compare two options
Too broad: “Which one is best?” Better: “Compare these two commuting options for a journey of 12 km, three days a week. Use travel time, total cost, reliability and weather as criteria. Show assumptions and tell me what information is missing.”
Supply your own current prices and timetable rather than accepting invented figures. Decide which criterion matters most to you before choosing.
Plan a small project
Too broad: “Make me a plan.” Better: “I have four weeks to prepare a small community event for 30 people. Draft a sequence of tasks, identify dependencies and suggest one checkpoint each week. Leave venue cost and availability as unknowns.”
Confirm real deadlines, permissions and costs with the people concerned. Treat the proposed plan as a draft that you can adjust.
Research a changing fact
Too broad: “What is the latest rule?” Better: “I am looking for the current official rule for this specific situation in my location. List the facts I need to establish, suggest primary sources to consult and flag anything you cannot verify without live access.”
Open the official source yourself. Check publication and effective dates, jurisdiction, exceptions and whether a newer version replaces the text. For a consequential decision, seek qualified advice.
Create a usable first draft
Too broad: “Write an email.” Better: “Draft a friendly 120-word email inviting colleagues to a 20-minute meeting about the project timeline. State the purpose, ask for one preparation item and leave date and location as placeholders.”
Read the final text for tone, factual claims and unintended commitments. Replace every placeholder and obtain any needed approval before sending.
Sources for further reading and checking
These independent resources help you find a reference. Check the language and scope: a French resource may not describe rules in another country. These are not sources automatically consulted for generated answers.
- CNIL — IA générative : questions-réponses
Questions and answers on using generative AI and protecting data.
Language: fr · Scope: France / UE - NIST — Generative AI Profile (AI 600-1)
A generative AI risk profile for identifying and managing risks.
Language: en · Scope: US - UNESCO — Guidance for generative AI in education and research
Guidance on generative AI in education and research.
Language: en · Scope: International
Link check: 2026-10-06 · These links are reading resources. The answer service does not consult them automatically. A checked link does not guarantee every document is current.