Skip to content
AI for Beginners

Can ChatGPT create videos?

Beginner4 min readBy25 September 2026Facts checked 25 September 2026

No. ChatGPT cannot make video clips, and OpenAI closed its separate video app, Sora, in 2026. ChatGPT can still help you plan a video, write the script, and create still images. This guide explains what changed and how to use AI video responsibly.

New here? Before this guide, you might want to read How to use ChatGPT: a beginner's guide.

No, ChatGPT cannot create videos. At the time of writing it produces text and still images, but not video clips. OpenAI did have a separate video tool called Sora, but closed it in 2026. ChatGPT can still help with most of the work around a video: ideas, scripts, shot lists, and images.

In this guide, you’ll learn what ChatGPT can and cannot do for video, what happened to Sora, and how to use AI video tools responsibly.

Why can’t ChatGPT make videos?

ChatGPT is built mainly to work with words. It can also create still pictures from a description, which is known as text-to-image. Video is a much heavier job: a few seconds of footage is many images that must stay consistent with each other, and generating it takes a great deal of computing power.

OpenAI tackled video through a separate product rather than inside ChatGPT, and that product no longer exists.

What happened to Sora?

Sora was OpenAI’s AI video tool. You typed a description and it made a short clip. It first launched for paying ChatGPT users in December 2024, and a standalone Sora app followed in 2025.

In March 2026, OpenAI announced it was shutting Sora down. The Sora app and website closed in April 2026, and OpenAI has also wound down the version developers used to build it into their own products. Reporting at the time pointed to the high cost of running it. OpenAI has not added video creation to ChatGPT in its place.

If you made videos with Sora, OpenAI’s help centre has an article on the discontinuation that explains exporting your work and when the data is deleted. Check it soon if you have anything you want to keep.

What can ChatGPT do for your video?

Quite a lot, as long as you make the footage somewhere else. ChatGPT can help you:

  • Brainstorm ideas and angles for a video
  • Write a script or voiceover, timed to a rough length
  • Plan a shot list or storyboard, scene by scene
  • Create still images for thumbnails, title cards, or storyboard frames
  • Write titles, descriptions, and captions

Several other companies offer AI video tools if you want generated footage. Features and rules differ a lot between them, so read what each one allows before you start.

How can ChatGPT help you plan a video?

Here is a straightforward way to use it:

  1. Tell ChatGPT what the video is for, who will watch it, and how long it should be.
  2. Ask for three possible angles or openings, and pick the one you like.
  3. Ask for a script with timings for each section.
  4. Ask for a shot list that matches the script.
  5. Ask for any still images you need, such as a thumbnail.
  6. Film the video yourself, or use a dedicated video tool, and edit it together.

This prompt handles steps one to four in one go:

I'm making a [length] video about [topic] for [audience].
It will be shared on [where it will appear].

Please give me:
1. Three possible opening hooks
2. A script with rough timings for each section
3. A shot list describing what should be on screen during each part
4. A suggested title and a two-sentence description

Keep the tone [friendly / professional / playful].

How do you use AI video responsibly?

Realistic AI video is powerful, and that is exactly why it needs care. A fake video that shows a real person saying or doing something they never did is called a deepfake. Deepfakes have been used to spread false stories, run scams, and humiliate people.

Concerns like these followed Sora from the start. OpenAI had to restrict videos of well-known figures after complaints from their families and others.

A few habits keep you on the right side:

  • Ask first: never make a realistic video of a real person without their clear permission.
  • Label it: say plainly when a video is made with AI, especially if it could be mistaken for real footage.
  • Avoid harm: do not create anything designed to mislead, embarrass, or impersonate.
  • Check the rules: laws on AI video are changing quickly and differ between countries.

It also helps to get better at recognising AI video when you see it. Our guide to spotting AI-generated content covers the signs to look for.

Next steps

If you are new to ChatGPT, start with how to use ChatGPT. Since still images are something ChatGPT can make, our walkthrough on how to create images with AI is a good place to try its creative side.

Frequently asked questions

Can ChatGPT make a video from a text description?
No. At the time of writing, ChatGPT does not produce video clips. It can write a script, plan the shots, and create still images you could use, but the video itself needs a different tool.
What happened to Sora?
Sora was OpenAI's video-making tool. OpenAI announced in March 2026 that it was closing it, and the Sora app and website shut down in April 2026.
Can ChatGPT make images for a video?
Yes. ChatGPT can create still images from a description, which you could use as thumbnails, title cards, or storyboard frames in a video you put together elsewhere.
Is it legal to make an AI video of a real person?
It depends on where you live and what the video shows, and the rules are changing quickly. The safe approach is not to make realistic videos of real people without their clear permission, and never to pass one off as real.

Try it yourself

Put this into practice with a ready-made prompt.

CreativeBeginner

Image Generation Prompt Builder

Turn a rough idea into a rich, detailed prompt for an AI image generator, with the subject, style, mood, and composition spelled out so you get a better picture.

Use this prompt →

CreativeBeginner

Story Starter

Kick off a story from a genre and a premise, with an opening you can build on and a few directions the plot could take next.

Use this prompt →

ResearchBeginner

Brainstorm Partner

Generate a wide range of fresh ideas on any topic, then have the AI pressure-test the strongest ones so you leave with options worth pursuing.

Use this prompt →

See it in action

Creative ProjectsBeginner20-30 minutes

How to create images with AI

Turn a description in plain words into an image made by AI, then refine it until it matches what you had in mind. You will learn how to describe what you want clearly and how to steer the result rather than settle for the first attempt.

Works with Claude, ChatGPT, Gemini