TutorialsFull Visual Sets with AI, No Design Skills Needed
One five-part formula, 3 hands-on runs and 8 templates — turn posters, article covers and slide visuals into publish-ready images in 20 minutes.
You do not have a design problem. You have a brief problem. When you order work from a designer, you say what to draw, where it happens, what mood it should have, how big it is and what words go on it. AI needs exactly the same five answers — give it all five and the image appears.
This lesson follows one complete path: take a real community reading event, produce the poster, the article cover and the slide visual in 20 minutes, and walk away with eight fill-in-the-blank templates.
What you take home
| 1 formula | 3 images | 8 templates | 1 brief sheet |
|---|---|---|---|
| Subject ＋ scene ＋ style ＋ composition ＋ text | Poster, article cover and slide visual, produced live | Ordered by your work week, ready to copy | Fill seven rows, join them into one prompt |
The promise: a live hands-on session (3 images produced on the spot) plus template files (8 prompt templates) plus homework.
Problems you probably recognise
- A poster is due and you freeze the moment you open a design tool — an hour of template browsing, and the layout breaks the moment you change the size;
- Article covers are stitched together from screenshots, or you dig through a stock library for half an hour and paste in whatever you find;
- Your deck is all text and you want one image to say "efficiency", but the stock search runs until quitting time without a single on-topic result;
- AI finally produced something, but the date has a wrong digit or the title has a typo — and you only notice after it is published;
- The same campaign needs three sizes (portrait poster, landscape cover, vertical WeChat image) and you start from zero every time you change the ratio.
The problem is never that you cannot design. It is that you described a picture but never stated the purpose, the ratio and the words that must appear on it — so AI guesses, and the guess looks nice but is unusable.
What you will learn
Understand first: four terms and six scenarios
Plain-language explanations of four concepts — text to image, prompts, reference images, text rendering; then six high-frequency scenarios you will face next week, with what AI does in each. Finally the heart of the course: the five-part formula (subject ＋ scene ＋ style ＋ composition ＋ text) and the one line that doubles your success rate.
Hands-on: one story, three images
Using a community reading event as the through-line, run three prompts live — a 3:4 portrait poster, a 16:9 article cover, and a slide visual produced in two steps (let AI write the prompt first). Each comes with the full copy-ready text, the demo point to watch, and how to fix one thing at a time.
Take it home: eight templates and five red lines
Eight prompt templates ordered by your working week (Monday promo → Wednesday poster → Thursday cover → Friday deck), copy and fill; a seven-row visual brief sheet; a tool-selection table organised by scenario rather than ranking; and five red lines, especially reading every word on the image before publishing.
Who it is for
Course structure
Three lessons plus one assignment, moving from "understand the four terms and six scenarios" to "produce three images live", and finally to "turn your role's recurring image needs into a repeatable workflow".
Understand first: four terms ＋ the five-part formula
Text to image, prompts, reference images and text rendering explained plainly; six high-frequency scenarios matched one by one; then the five-part formula taken apart — what each slot answers, which one people forget, and how to fill it so you do not get a wasted image; plus the one-line technique and the safest two-step path for beginners.
Hands-on: one story, three images
Around the case "a community reading event on 26 October", three prompts in sequence: run A produces a 3:4 poster (with the re-roll demo), run B produces a 16:9 article cover (with the thumbnail check), run C takes two steps — let AI translate the theme into visual language, then generate a 16:9 slide visual. It closes with the editing rule: change only one thing.
Take it home: eight templates and five red lines
Eight templates in the real order of your working week, each in full, with fill-in slots; a seven-row brief sheet that becomes a prompt once completed; a tool-selection table organised by scenario; and five red lines.
Course materials
This course comes with two supporting files:
- Course handout: the full original text, including the four concepts, six scenarios, the five-part formula, the three hands-on prompts, eight templates, the blank brief sheet, five red lines and the homework.
- Case brief: the Community Reading Event brief (fictional demo data) with the event name, date, venue, theme colour and the exact words that must appear on the images. Both the live session and part one of the homework rely on it — copy the prompt and you can reproduce the results.
Download:
- AI图文海报汇报配图-课程讲义.md (course handout)
- AI图文海报汇报配图-读书分享会信息单.md (case brief)
The principle
One line: AI produces the image; you pick it, fix the text and sign off. AI is the designer, you are the client.
Five red lines:
- AI makes things up with total confidence: dates, times, prices and phone numbers must be checked by you — read every word on the image before it goes out;
- Never feed confidential or sensitive information: internal data, unpublished plans and other people's portraits are off limits — what you send cannot be taken back;
- The output is not the final draft: it carries your name and your responsibility — a typo or a wrong date published is on you;
- Do not misuse portraits or brand logos: no real celebrity faces, no other companies' logos — copyright and portrait rights apply;
- Label AI-generated images: when publishing publicly, mark them as AI-generated rather than letting them pass as real photography.
Assignment: Turn a Pile of Records into a Retrospective You Can Present
Three stages in one assignment — reproduce a full GRAI run on the course materials, run the whole thing on a real project of your own with a time comparison, then fix it into your team's retrospective SOP. All three, or it does not count.
Understand First: Four Terms and the Five-Part Formula
Learn the four concepts, match yourself to six scenarios, then memorise the heart of the course — subject, scene, style, composition, text. Miss one and the image drifts.