Can you create a complete short video story using AI while holding only your phone? This is what Ashraf Salim, owner of the PRO GRAPHIC YouTube channel, tried in an episode he began with an AI guide and then created a story titled “He Had Little.” Here, we summarize his method step by step and the tools he chose for each stage.
Disclosure: This episode is part of the Daleel AI Partner Program, and Ashraf prepared the explanation and experiment in his own independent style.
Tired of juggling ten tabs? ToolSuite bundles the AI workflow tools power users rely on — in one place.
Try ToolSuite NowThe Tools I Used
- ChatGPT: To write the story and divide it into scenes, describe the characters, then turn each scene into a ready-made prompt for the video tool.
- Google Flow: Google’s AI video scene creation tool, powered by Google Veo models. I used it to generate every shot in the story.
- InShot: A mobile video editing app, where I combined the shots into one video.
Step 1: Start with a Clear Idea in ChatGPT
Ashraf asked ChatGPT for a short human-interest story about a young man who has very little but gives everything he has. His advice: the clearer the prompt, the better the assistant understands the type of story and the tone you should maintain.
Step 2: Break the Story into Scenes
Instead of asking for a complete video all at once, ask for the story to be divided into consecutive scenes: what we see in each shot, what movement it contains, and how the narrative transitions to the next scene. This way, the big idea is transformed into small parts that are easy to work on.
Step 3: Establish the Characters’ Appearance
Request consistent descriptions for each character (the young man, the mother, the elderly man, and the shop owner), then reference images for them. These images are not the final scenes, but rather references that help keep the character’s features consistent from one shot to another.
Step 4: Turn Each Scene into a Detailed Prompt
He returned to the chat and asked for each scene to be phrased in English for the video tool: who appears, what they do, how the camera moves, and what the lighting is like, for a scene lasting about ten seconds. It is not enough to write “a young man in a room”; specify what the camera should capture and how the action develops. Review the text before copying it so that it does not add elements that are not in the story.
Step 5: Generate the shots in Google Flow one by one
In Google Flow, open a project and paste the prompt for the first scene, then review the result and compare it with the idea: Did the required details appear? Is the movement clear? Does the character resemble its reference? Do not move on to the next scene before you know what worked and what needs improvement. If you do not like the shot, adjust the description and try again. The idea is to treat generation as short stages that you review, not as a single button that creates a finished film.
Step 6: Assemble the Scenes in InShot
Place the clips on the timeline in InShot and arrange them according to the sequence of the story, adjust the beginning and end of each shot and the transitions between them, and review the pacing of the scenes before exporting. The result at the end of the episode: a complete short story made from the phone.
Important Note from the Episode
The AI Index guide introduces you to tools and helps you get closer to them, but it does not create images or videos itself. The presence of a tool in the guide does not mean it is free; plans and limits vary from one tool to another, so review each tool’s page before you begin.
For more tutorials on creating content with AI from your phone, follow the PRO GRAPHIC YouTube channel.

Comments
0No comments yet.
Please log in to comment.
Comments are available for members only. Sign in to participate in the discussion, or create a new account for free.