GONDAL AI / BEGINNER GUIDES

Make Your First AI Image & Video: A Hands-On Lesson

Reading about AI images is like reading about swimming. This lesson is the pool: you will describe a picture in words, watch it appear, fix what is wrong with it the way a photographer would, then bring it to life as a short video — step by step, with the exact words to use.

What you need

Any AI tool that makes images — most major AI chats now do, and several free image tools exist (see the tools directory). No drawing skill, no camera, no experience. About thirty minutes and a bit of curiosity.

Step 1 — The formula for a good image prompt

An image prompt is not a wish; it is a brief for a very literal artist. The formula that works:

Subject + details + setting + light/mood + style.

Weak: “a bakery”
Strong: “A small cosy bakery at dawn, a baker in a blue apron holding a tray of fresh croissants, warm golden window light, wooden shelves with bread, photographed like a magazine photo, shallow depth of field”

Every added detail removes a guess the AI would otherwise make for you — usually wrongly.

Copy-paste prompt — your first image. “A cheerful street market stall in Saskatoon in autumn, wooden crates of red apples and pumpkins, an elderly vendor smiling and handing a paper bag to a child, soft morning light, gentle mist, photorealistic, warm colours.”

Generate it. Then look at it like a critic: what is wrong? Hands, text on signs, and extra objects are the usual suspects — which leads to the real skill:

Step 2 — Fix it like a photographer, not a gambler

Beginners re-roll the dice hoping for luck. Creators direct. Change ONE thing per try, in words: “same image, but move the vendor to the left,” “brighter, midday sun instead of mist,” “remove the text from the sign,” “closer up, just the hands exchanging the bag.” Three directed fixes beat thirty random generations, and you learn what the tool hears.

Step 3 — Turn the still into motion

Many image tools and video tools can animate a still image: upload your picture and describe the movement. Keep it small — big motion breaks; small motion charms:

“Gentle camera push-in, the vendor smiles and nods, steam rises from a coffee cup, leaves drift past.”
Five to ten seconds is plenty. You are making a moment, not a movie — yet. (When you are ready for the full video workflow — script, voice, captions — our Your First AI Video walkthrough is the next lesson in spirit, and the studio's daily bread.)

Step 4 — Captions and a title

A clip without words is a riddle. Add one line of on-screen text that makes a scroller stop: the most surprising or useful sentence about the scene. Then ask your AI chat for five title options and pick the one a friend would click.

Step 5 — Post it honestly

Rights: check your tool's terms for commercial use if this is for a business; personal posts are fine on mainstream tools. Never generate images of real people without their consent, and never pass off an AI scene as a real photograph of a real event.
Labels: several platforms ask you to mark AI-generated content — follow the platform's current rule when you post. Honesty travels well.
Keep the prompt. Save the winning words with the image. Your prompt library is your style, growing.

Check yourself

My prompt used the formula: subject + details + setting + light + style.
I fixed problems by directing changes one at a time, not by re-rolling.
My clip's motion is small and believable.
It has a caption line, a title I chose myself, and I know my tool's rules for using it.

Stuck on a prompt? Ask the chat on this page: describe the picture in your head and say “turn this into a detailed image prompt using the subject + details + setting + light + style formula.” That completes the AI Foundations course — revisit any lesson from the courses page, and when you want a system for posting every day, the product shelf is where the studio keeps its own.