AI Applications
Senior 2 · ICT
AI Applications
Lesson 4 of 4
22 min
At 9 p.m. before a test, an AI tool can explain a topic again in simpler words, quiz you, or turn your notes into flashcards. That makes it a powerful study partner. But the same tool may invent a source or a statistic, and it sounds just as confident when it is wrong. Good students use AI to think more, not less.
| Use | Example prompt |
|---|---|
| Explain simply | "Explain how a router differs from a switch for a 13-year-old, with one everyday example." |
| Quiz me | "Ask me 5 questions on network media, one at a time. Wait for my answer, then tell me whether I am right and why." |
| Summarise | "Summarise these notes in 6 bullet points: [paste your own notes]." |
| Flashcards | "Make 10 question-and-answer flashcards from these notes." |
| Feedback | "Here is my paragraph. Point out two weaknesses, but do not rewrite it." |
| Translate | "Translate 'network' into Kinyarwanda and give an example sentence." |
Worked example
the quiz-me method. Byiringiro is revising network devices. He asks an AI tool for 5 questions, one at a time, and answers each before he sees the feedback. He gets 3 out of 5 right, notices that both mistakes were about hubs, and rereads that part of his notes. This is active recall: pulling an answer out of your own memory. It makes learning stick far better than rereading. The AI did the asking; Byiringiro did the thinking.
Some AI tools answer only from what their model learned during training; others search the web and show links to their sources. Either way:
Worked example
checking a claim. An AI tool writes: "Zipline began delivering blood in Rwanda in 2019 from a centre in Huye." Byiringiro checks an encyclopedia article on Zipline. It says deliveries began in October 2016 from Muhanga, and a second centre opened at Kayonza in 2018. The AI was wrong about both the year and the place. He uses the checked facts and cites the article, not the chatbot.
When your teacher allows AI, keep a short record: the tool's name, the date, your prompt, and how you used the answer. For example: "Used a chatbot on 12 March 2027 to generate 5 practice questions on network devices; I checked all the answers against my notes." This is honest, and it shows that the final work is yours.
Tip
Tutor, not ghost-writer. Let AI explain, question and give feedback, but do the answering and the writing yourself. For facts, the rule is trust, but verify: every fact needs a source that you have opened yourself.
Write a "quiz me" prompt for revising Scratch block categories.
Solution (one good answer): "Ask me 6 questions, one at a time, about which category different Scratch blocks belong to. Wait for my answer, tell me whether I am right, and explain any mistake. Do not show the answers before I reply."
An AI answer gives this reference: "Habimana, J. (2021). Networks in Rwanda. Kigali University Press." You search and cannot find any such book. What has probably happened, and what should you do?
Solution: The reference is probably a hallucination. Do not cite it. Look for a real source that supports the claim; if you cannot find one, leave the claim out.
Which is better study practice, and why? (a) asking AI to write your homework answers; (b) asking AI to quiz you, then checking your mistakes in your notes.
Solution: (b). Active recall builds memory and shows you what to revise. (a) is dishonest and leaves you unprepared for the exam.
Write an AI-use record for this: you used a chatbot to simplify an explanation of fibre optic cable for your revision notes.
Solution (example): "Tool: [chatbot name]. Date: [today]. Prompt: 'Explain fibre optic cable simply for an S2 student.' Use: I read the explanation, checked it against the lesson on network media, and wrote my notes in my own words."
Summary
Sources and credits
The study prompts, worked examples and practice items were written by Ganzaa.
This lesson adapts CC BY-SA material and is shared under CC BY-SA 4.0.
