No, not on its own. As of October 2026, ChatGPT has no built-in video generator. OpenAI discontinued Sora, its video model: the Sora app and website shut down on April 26, 2026, and the Sora API followed on September 24, 2026. But you can still make videos with ChatGPT. Apps inside ChatGPT, like Adobe and Descript, can produce and edit real video files, and ChatGPT itself is very good at the parts that come before the camera: the script, the storyboard, the shot list, and the prompts you feed a video tool.
This guide covers what happened to Sora, what ChatGPT can still do for video, the practical ways to get a finished video out of it, and five prompts you can copy today.
TL;DR
- ChatGPT can't generate video by itself as of October 2026. Sora is gone, and there's no OpenAI video model inside ChatGPT now.
- Apps fill the gap. Adobe for ChatGPT (Premiere and Express), the Descript app, the ChatCut plugin in the ChatGPT desktop app, and third-party GPTs like invideo's Video Maker can all produce video from a ChatGPT conversation.
- ChatGPT is still the best place to plan a video. Scripts, scene-by-scene shot lists, voiceover timing, captions and prompts for AI video generators are all things it does well.
- Free is possible but limited. The ChatGPT free plan handles the writing. Most video apps run on credits or paid plans for anything beyond a taste.
- If you want to edit footage you already have, read can ChatGPT edit videos instead.
What happened to Sora?
Sora was OpenAI's text-to-video model, and for a while it was the reason people searched "can ChatGPT make videos." OpenAI announced its discontinuation on March 24, 2026. The timeline:
- April 26, 2026: the Sora app and web experience shut down.
- September 24, 2026: the Sora API (sora-2 and sora-2-pro) shut down.
OpenAI's own summary is in the help center: What to know about the Sora discontinuation (opens in a new tab). The practical result is simple. If you open ChatGPT today and ask it to "make a video of a dog surfing," it can't render that clip itself. Older articles and videos that show Sora inside ChatGPT are out of date.
What ChatGPT can do for video today
Losing a video model doesn't make ChatGPT useless for video. Most of the work in a good short video happens before anyone hits record or render, and that's where ChatGPT is strong.
- Scripts. Hooks, full scripts, and multiple versions for different lengths (15, 30, 60 seconds).
- Storyboards and shot lists. A scene-by-scene plan with what's on screen, what's said, and how long each shot runs.
- Prompts for video generators. If you use a separate AI video tool, ChatGPT can write consistent, detailed prompts for each scene.
- Voiceover scripts with timing. Lines broken into timed chunks so they match your cuts.
- Captions, titles and descriptions. On-screen text, platform captions, hashtags and thumbnail text.
- Still images. ChatGPT can generate images, which you can use as storyboard frames or thumbnails. Some separate tools can animate a still image into a short clip, but that animation happens outside ChatGPT, and results vary a lot by tool.
One limit to know: ChatGPT can't watch a video or a YouTube link the way you do. If you want feedback on an existing video, give it a transcript and a few screenshots.
Ways to actually make a video with ChatGPT
Here are the realistic routes as of October 2026. Each one uses ChatGPT for the conversation and a video service for the actual footage or edit.
| Route | What you get | Cost notes |
|---|---|---|
| Adobe for ChatGPT (Premiere and Express) | Quick cuts, highlight reels, resizing for Shorts and Reels, animated designs | Some features need a paid Adobe plan |
| Descript app in ChatGPT | Edits and clips built from your own recordings, using Descript's tools | Needs ChatGPT Plus or higher; uses Descript media minutes and AI credits |
| ChatCut plugin (ChatGPT desktop app) | Generation, timeline edits, captions, voiceover, motion graphics, audio cleanup | Credit plans from $25/month for 100 credits (vendor pricing) |
| invideo Video Maker GPT | A third-party GPT that turns a prompt or script into a video | Check invideo's own plans for free vs paid limits |
| Write in ChatGPT, generate elsewhere | Script and prompts from ChatGPT, footage from a dedicated AI video or avatar tool | Depends on the tool; most charge credits |
Adobe for ChatGPT
Adobe for ChatGPT launched in August 2026. You call it with @Adobe in a chat and ask for things like a quick cut, a highlight reel, or a version resized for YouTube Shorts and Instagram Reels, using Premiere and Express behind the scenes. It can also make animated designs. This is the closest thing to "ChatGPT makes my video" from a big, established brand. Some features require a paid Adobe plan.
The Descript app
Descript has an official app in ChatGPT. It's best when you already have footage or audio, such as a talking-head recording or a podcast, and want it cleaned up and cut into something publishable. It needs a ChatGPT Plus plan or higher, and it uses your Descript media minutes and AI credits.
ChatCut
ChatCut is a plugin for the ChatGPT desktop app on Mac and Windows. It isn't available on the web or mobile versions. According to ChatCut, it handles generation, timeline edits, captions, voiceover, motion graphics and audio cleanup, and can export an XML file for a traditional editor. ChatCut says it installs on every ChatGPT plan, including Free, but the work itself runs on ChatCut credits.
invideo's Video Maker GPT
invideo offers a Video Maker GPT in the GPT Store. It's a third-party GPT, not an OpenAI feature, so the video is produced by invideo's service, under invideo's terms and plan limits.
Write in ChatGPT, generate in a dedicated tool
The most flexible route is to keep ChatGPT as your writer and planner, then paste its output into a dedicated video tool. Two broad categories are worth knowing:
- AI video generators create footage from a text prompt or an image.
- Avatar video tools turn a script into a video of a digital presenter speaking it.
Both categories change fast, so pick based on a short test with your own script rather than a feature list. The prompts later in this guide are written to work with either.
Can ChatGPT make videos for free?
Partly. The ChatGPT free plan can write every script, shot list, caption and prompt in this guide at no cost. The video itself is where money usually comes in:
- The new ChatGPT desktop app, with its Chat, Work and Codex modes, is available on every plan, including Free.
- ChatCut says its plugin installs on the Free plan, but rendering uses paid credits.
- The Descript app needs ChatGPT Plus or higher.
- Adobe for ChatGPT has features that need a paid Adobe plan.
- Most standalone AI video tools offer a small free tier or trial, then charge credits.
A realistic free workflow: plan the whole video in ChatGPT, film it on your phone, and edit it in a free mobile editor. If you're weighing a paid ChatGPT plan for the app integrations, our ChatGPT Plus free trial page explains what's actually available.
Can ChatGPT make a video from photos?
Not natively. ChatGPT can describe your photos, put them in order, write a caption or voiceover for each one, and suggest timing and transitions. To turn them into an actual slideshow or animated clip, you need a video app: one of the apps above, or a separate editor or AI tool that animates still images. A good approach is to upload the photos to ChatGPT, ask for a shot list with timing, and then follow that list in your editor.
Can ChatGPT do text to video?
Not by itself since Sora shut down. "Text to video" now means one of two things with ChatGPT: a third-party app or GPT inside ChatGPT that generates the clip, or ChatGPT writing the prompts that you then run in a separate AI video generator. The second route gives you more control over which tool you use, and the prompt pack below is built for it.
5 copy-paste prompts for making videos with ChatGPT
Replace the parts in square brackets. These work in the free plan. For more variations, try our free AI video editing prompt builder.
1. A 30-second script with a scene-by-scene shot list
Write a 30-second vertical video script for [business/topic] aimed at [audience].
Goal: [get followers / explain a product / drive sales].
Format the answer as a numbered shot list. For each shot give:
- time range (for example 0:00-0:03)
- what is on screen
- the exact spoken line
- on-screen text, if any
The first 2 seconds must be a strong hook. End with one clear call to action.2. A storyboard table
Turn this script into a storyboard table with columns:
Scene | Duration | Camera framing | Visual description | Audio or voiceover | On-screen text.
Keep each scene under 5 seconds. Flag any shot I could not film with just a phone.
Script: [paste script]3. A prompt pack for an AI video generator
I'm going to generate this video scene by scene in an AI video generator.
Write one prompt per scene. Each prompt should describe subject, setting,
lighting, camera movement, mood and duration in plain language.
Keep the main character and visual style consistent across every prompt
by repeating the same description word for word.
Video idea: [describe the video]
Style: [for example bright, clean, handheld, cinematic]
Number of scenes: [number]4. A voiceover script with timing
Write a voiceover for a [length]-second video about [topic].
Speak at about 2.5 words per second. Split the voiceover into timed lines
like this: [0:00-0:04] line text.
Tone: [friendly / confident / calm]. Avoid jargon.
Mark where a pause or a cut should happen.5. Captions, title and description
Here is my finished video script: [paste script].
Write:
1. Five title options under 60 characters
2. A platform caption for [Instagram Reels / TikTok / YouTube Shorts] under 150 words
3. Short on-screen caption lines, one per shot, max 6 words each
4. Five relevant hashtags
Keep the language plain and specific. No hype words.What about Claude?
Claude doesn't have a generative video model either. As of October 2026, Anthropic's closest feature is Claude Motion, a beta on Team and Enterprise plans that writes code to animate text, charts, shapes and images, then exports the result as an MP4. It can't produce realistic footage or people. If you're comparing the two assistants for video work, see can Claude edit videos.
Where ChatGPT falls short (and what to verify)
- No native video generation. Anything labeled "ChatGPT video" today is either a third-party app, a GPT, or ChatGPT writing for another tool.
- Apps change quickly. Features, pricing and plan requirements for Adobe, Descript, ChatCut and invideo are set by those companies. Check their pages before you commit.
- Credits add up. Many video apps charge per generation or per minute of media. Test with a short clip before running a full project.
- It can't watch your video. For feedback on footage, share a transcript and screenshots.
- Old tutorials still mention Sora. If a guide tells you to open Sora inside ChatGPT, it was written before the 2026 shutdown.
- Editing is a separate question. Cutting and cleaning up footage you already have is covered in can ChatGPT edit videos.
Frequently asked questions
Can ChatGPT make videos for free?
ChatGPT's free plan can write scripts, storyboards, shot lists, captions and prompts for free, but it can't render video on its own. Apps that produce video from ChatGPT, such as ChatCut, Descript and Adobe, generally use credits, a paid ChatGPT plan, or a paid plan of their own for most features.
Is Sora still available?
No. OpenAI discontinued Sora. The Sora app and website shut down on April 26, 2026, and the Sora API shut down on September 24, 2026. As of October 2026, there is no native OpenAI video model in ChatGPT.
Can ChatGPT turn photos into a video?
Not directly. ChatGPT can order your photos, write captions or a voiceover, and plan the timing, but you need a video app or editor to turn them into an actual video file. Some separate AI tools can animate still images into short clips.
Can ChatGPT edit videos?
Not natively, but apps inside ChatGPT can. Adobe for ChatGPT handles quick cuts, highlight reels and resizing, the Descript app works on your own recordings, and the ChatCut plugin in the desktop app does timeline edits and captions. See our full guide to whether ChatGPT can edit videos for details.
What is the best way to make a video with ChatGPT?
Use ChatGPT to write the script, shot list and captions, then produce the video in a dedicated tool. If you want to stay inside ChatGPT, use an app such as Adobe for ChatGPT, Descript or ChatCut. If you want the most control, paste ChatGPT's prompts into the AI video or avatar tool you prefer.
Plan the video in ChatGPT, then pick the right tool
Start with prompt 1 above and get a script and shot list you're happy with. That's the part that decides whether a short video works, and ChatGPT does it well on any plan. Then choose the app or tool that fits your budget to produce it.
If you want to get comfortable using AI assistants for real work like this, the free Claude Cowork course walks through practical workflows step by step, no coding required.