Go from "I've never really used AI" to confidently using it for real, useful things — in about an hour. No jargon. If you can send a text message, you have every skill this needs.
| # | Module | Takeaway |
|---|---|---|
| 1 | What AI actually is (and isn't) | A prediction machine, not a mind |
| 2 | Meet the chatbots | Pick one — they're all fine to start |
| 3 | Your first conversation | Talk to it like a smart colleague |
| 4 | The 4-part prompt formula | Role · Task · Context · Format |
| 5 | Five real-life uses today | Emails, planning, learning, summaries, drafts |
| 6 | Beyond text — images & voice | The same skill applies everywhere |
| 7 | Limits, mistakes & staying safe | Trust but verify; guard your private info |
| 8 | Your 7-day practice plan | One small task a day builds the habit |
You know how your phone suggests the next word when you type? Type "Happy" and it offers "birthday". That's a tiny prediction machine — it has seen millions of sentences and knows what usually comes next.
Modern AI, the kind behind ChatGPT, is that same idea scaled up enormously. It was trained on a gigantic amount of text — books, articles, websites — and learned the patterns of how words and ideas fit together. When you ask it something, it predicts, word by word, what a good answer looks like. That's the whole magic trick.
Great at: drafting, summarising, explaining, translating, brainstorming, rewriting. Unreliable at: precise facts, very recent events, maths done "in its head", and anything where being wrong has serious consequences. A great assistant — a terrible oracle.
Three quick myths: "AI is conscious" — no, it's brilliant pattern-matching. "It'll be obviously wrong when it's wrong" — sadly not. "I'm too late to learn this" — you're actually early.
The AI world looks crowded, but for beginners it's a handful of chatbots — and they're much more alike than different. ChatGPT (OpenAI) is the famous one. Claude (Anthropic) is known for careful, thoughtful writing. Gemini is Google's. Copilot is Microsoft's, built into Windows and Office. All have free versions, and all work through the same simple text box.
Sign up with your normal email and a strong password — and treat the chat box as semi-public. Don't paste in passwords, bank details or other people's private information (more in Module 7).
The number-one habit shift: with Google you type keywords ("weather Manchester tomorrow"). With AI you talk in full, natural sentences, like briefing a bright colleague.
Type just "dinner ideas" and you get something generic. Instead try: "I've got chicken, a red pepper and rice to use up tonight. I have twenty minutes, cooking for two adults and a fussy seven-year-old. Give me three ideas, easiest first." Same tool — the difference was entirely in how you asked.
Keep the conversation going. You don't have to get it right first time — just reply. "Make option two vegetarian." "Now write me a shopping list." Each reply builds on the last. Refining isn't failure; refining is the technique.
The heart of the course. Every great prompt has up to four parts — you won't always need all four, but knowing them means you'll never stare at the box wondering what to type.
Weak: "Write a speech for my dad's birthday." Strong: "Act as a warm, funny speech-writer (role). Write a 2-minute 70th birthday speech for my dad (task). He's a retired postman, mad about fishing, famous for terrible puns, and hates being the centre of attention (context). Keep it light, end on one sincere line, give me two versions (format)."
1. The difficult email. "Write a polite but firm email to my landlord: the boiler's been broken nine days, reported twice, I want a repair date this week — keep it professional." Drafted in seconds.
2. The planner. Meals for the week on a budget; a three-day itinerary for someone with dodgy knees; a revision timetable. AI takes your constraints and builds a plan — then reshuffles instantly when "Tuesdays are no good".
3. The patient teacher. "Explain pensions like I'm fifteen." Then "simpler." Then "give me an example with real numbers." It never sighs, never judges, never tires of you asking.
4. The summariser. Paste a long email chain or article: "summarise this in five bullet points and tell me if there's anything I need to act on." Twenty minutes of reading becomes two.
5. The first-draft machine. CVs, complaints, speeches, tricky texts. The mindset: AI writes the first draft, YOU write the final draft. It breaks the blank page; you add the truth and personality.
Everything you've learned applies beyond text. Image generators work on the same principle: describe what you want in detail, then refine. "A watercolour fox reading a book under a lamp, cosy, children's-book style." Don't like it? "Make it night-time. Make the fox older." Same conversation skill, different canvas.
Voice works both ways: you can talk to most chatbots out loud (brilliant while cooking or walking), and AI voices can read text aloud or narrate a video.
Remember Module 1: AI predicts likely-sounding answers. Sometimes it predicts something that sounds perfect and is simply false — a made-up statistic, a book that doesn't exist, a legal rule from the wrong country. The industry calls these hallucinations. They're not rare, and they're not flagged.
Privacy, three rules: never paste passwords, PINs or full financial details; be careful with other people's information; and check your tool's settings — most let you turn off using your chats for training with one toggle.
Skills stick through use — one small task a day, five minutes each: