Explain AI
The information an AI learns its patterns from is called this.
What is training data?
Misinformation is false information shared by mistake. This term describes false information spread on purpose to mislead.
What is disinformation?
Name two visual clues that an image might be AI-generated.
Any two of: distorted hands or fingers, garbled text, odd backgrounds, overly smooth skin, inconsistent shadows or lighting
Your showcase must state this about your AI use.
What are which tool(s) you used and how you used them?
ChatGPT, Gemini and Claude are all examples of this type of AI tool, which produces new text in response to a prompt.
What is a chatbot (or generative AI / large language model)?
Netflix suggests a show you've never heard of by comparing your viewing habits with millions of other users. This is the type of AI tool behind it.
What is a recommendation algorithm?
A hiring AI trained mostly on past résumés from men starts ranking women lower. This is the root cause.
What is biased training data?
Uploading a photo to find where else it appears online, and where it first appeared, is called this.
What is a reverse image search?
Name three things you should never type into a chatbot.
Any three of: full name, address, school, passwords, photos of other people, friends' private messages
The instruction or question you type into an AI tool to tell it what you want is called this.
What is a prompt?
Unlike a calculator, which follows fixed rules, a generative AI chatbot creates text by doing this.
What is predicting the most likely next word based on patterns?
A scammer clones your grandparent's voice from a short online clip and rings asking for money. Name the risk and one way to protect yourself.
What is a deepfake/voice-cloning scam? Protections: agree a family code word, hang up and call back on a known number
Opening new tabs to see what other sources say about a claim or its author, instead of staying on the page, is called this.
What is lateral reading?
A friend says "Just copy the AI's answer, no one can tell." Give two reasons this is a bad idea.
Any two of: it isn't your own work (plagiarism), AI can be wrong, you don't learn anything, it breaks school rules, you can't explain it if asked
Tools like DALL·E, Midjourney and Adobe Firefly turn a written description into this.
What is an image (AI image generator)? Bonus: it may contain errors like distorted hands or garbled text
Name two everyday tools that use AI, apart from chatbots.
Any two of: spam filters, autocorrect, Google Maps traffic predictions, face unlock, photo tagging, voice assistants, fitness tracker alerts
An AI image generator trained on artists' work without their permission raises this legal and ethical issue.
What is copyright (ownership of creative work)?
A viral photo shows a shark swimming down a flooded street. Give two steps you'd take to check if it's real.
Any two of: reverse image search, look for the original source, check reputable news outlets, check the date and location, look for AI clues
After an AI writes a first draft, name two things you should do before you submit it.
Any two of: fact-check with reliable sources, edit in your own words, check for bias, acknowledge the AI use
Before using an AI tool for schoolwork, name two things you should check about the tool itself.
Any two of: what it does with your data or privacy policy, age restrictions, whether your school allows it, whether it's reliable, whether it cites sources
In one sentence, explain why an AI can sound completely confident and still be wrong.
What is: it predicts likely-sounding words from patterns and doesn't understand or check whether they're true?
Name two ways relying on AI for every answer could harm your own learning.
Any two of: skills don't develop, you can't spot mistakes, less critical thinking, you can't explain the work yourself, less creativity
A chatbot gives you a statistic with a source link, but the link doesn't exist. Name this problem and explain why you must verify even when the AI sounds sure.
What is a hallucination? AI generates plausible-sounding text and can invent facts or sources
Which is the most responsible use of AI: (a) having it write your reflection for you, (b) having it create practice quiz questions and checking the answers in your textbook, or (c) pasting a friend's private message into it for advice? Justify your choice.
What is (b)? You use AI as a tool, verify its accuracy and keep the thinking your own, while (a) replaces your work and (c) breaches your friend's privacy
You ask two different AI tools the same question and get different answers. Explain what this tells you about relying on a single AI tool, and what you should do next.
What is: AI tools can be wrong or have different training data and biases, so you shouldn't trust one tool alone? You should check both answers against reliable sources