How to Make a Movie Poster About Yourself in ChatGPT: A Simple Prompt for a Personal Brand

Every October the same format surfaces in themed channels: ChatGPT builds a horror movie poster starring you, without a single uploaded photo. Here is what actually makes this trick work, rather than just look striking in a screenshot.

What makes it work

ChatGPT can store facts about you between conversations through Memory and pull them into a new request without you reminding it. If this setting is on and enough conversations have piled up, the model already knows your work habits, recurring topics, and even the small quirks you mentioned in passing.

When you ask it to build a movie poster "about you", the model doesn't draw at random. It takes the accumulated facts, picks a conflict and a mood from them, and GPT Image builds the visual on that basis. The longer and livelier your chat history, the more accurate the result.

A working prompt

Here is wording that gives a predictably more accurate and more surprising result than an abstract request with nothing from your chat history behind it:

"Make a movie poster about me, using what is already known about me from our conversations. Imagine that my life already exists as a film in the genre of a psychological thriller or mystery, and this poster is for its premiere. Decide on your own the genre, the mood, the main inner conflict, and what frightens or haunts the hero. Come up with a title, short and a little unsettling, like a real film. Don't explain your train of thought, just show the poster."

The key part here isn't the visual description: we directly ask the model to determine the conflict and the mood from what it knows. If you remove that part and just ask to "draw a horror movie poster", you get a random picture with no link to you, ordinary stock horror art.

What to do if there is little history

If you rarely write to ChatGPT or Memory is off, the model has nothing to build a personal conflict from. In that case, add five to seven concrete facts to the request itself: what you do, what repeats in your work every week, one real worry, and one habit that people around you notice first.

That way the model gets the same input that Memory would have given it, just directly in the text of the request instead of from chat history.

Where to use the result, not just show it once

A one-off striking poster is quickly forgotten in the feed. Working uses: a cover for a series of posts about your personal path in the profession, a visual for a story with the question "what kind of film would you get", or a reason for a carousel where each slide reveals one point of the "plot" that the model came up with.

If the poster turned out truly accurate in mood, that is the signal the request was built right: the model didn't guess, it pulled something real out of the accumulated context.

Bottom line

This trick rests on two things: Memory that really remembers something about you, and a prompt that asks the model to find the conflict itself, not wait for a ready description from you. Without the first component, a detailed list of facts in the request itself helps. Without the second, you get a random picture with no link to you.

I write about breakdowns of similar tricks for personal content and visuals in the channel «Растём с ИИ» ("Growing with AI"). A related piece: how to make a business portrait in ChatGPT from an ordinary selfie. If you need a whole visual content system for your niche, you can start with a free consultation.