--- title: "Can you send ChatGPT videos? What it does with them, and what to do instead" url: https://canyouuseai.com/blog/can-you-send-chatgpt-videos/ summary: "OpenAI's upload help page lists no video formats, so send audio, a transcript or screenshots instead. Eligible subscribers can share live camera video in Advanced Voice on the mobile apps." published: 2026-10-15 updated: 2026-10-15 author: "Jack Stovell" publisher: "Adapt Progress Evolve Limited" language: en-GB --- # Can you send ChatGPT videos? What it does with them, and what to do instead OpenAI's upload help page lists no video formats, so send audio, a transcript or screenshots instead. Eligible subscribers can share live camera video in Advanced Voice on the mobile apps. Short answer, as of October 2026: OpenAI's help page on uploading files lists documents, spreadsheets, presentations, images and audio. Video formats are not on that list. Video shows up in one place only, and it's live: in the ChatGPT iOS and Android apps, eligible subscribers can share their camera or screen during an Advanced Voice conversation. So it helps to be clear which kind of "sending a video" you mean. ## What people mean by 'sending ChatGPT a video' People use the phrase for three different jobs, and most of the confusing advice online comes from mixing them up. 1. **Uploading a video file** (a clip from your phone, a screen recording, a webinar download) and asking what happens in it. 2. **Giving it the sound or the words**: an audio file, or a transcript of what was said. 3. **Getting ChatGPT to make or edit a video** for you. The official answer is different for each one, so take them separately. ## What OpenAI's help pages say about video, audio and image uploads **Video files.** OpenAI's file uploads page has a table of supported formats: DOCX, PDF and TXT for documents, spreadsheet formats, PPTX, and a list of audio formats. No video format appears. The audio row accepts "audio-only MP4" and "audio-only WebM", then adds that WebM and MP4 files identified as video are not supported by audio uploads. OpenAI's capabilities overview also covers files and audio and says nothing about uploading video. **Audio.** Audio uploads are available on paid plans and workspaces, and OpenAI says they are "not available on the Free plan at this time". Audio files can be up to 512 MB, and the file must contain "a valid, decodable audio stream". You can ask for a transcript and then ask questions about it. OpenAI also warns that transcripts may contain errors and speaker identification may be unreliable. **Images.** Image uploads are available on Free and paid plans, with a limit of 20 MB per image. Free users are limited to 3 file uploads per day. **Live video.** The ChatGPT Voice page says the newer **Live** option "does not support video or screen sharing". Eligible subscribers can still share live video or their screen in the iOS and Android apps by switching to **Advanced** in Settings, then Voice. This is a live camera feed during a conversation. It doesn't let you upload a recording. **Making video.** OpenAI says its Sora web and app experiences were discontinued on 26 April 2026, and the Sora API on 24 September 2026. None of the help pages we read describe another way to generate video in ChatGPT, so we won't claim one. If making videos is your goal, our page on [making YouTube videos](/questions/make-youtube-videos/) covers that. OpenAI adds that availability "may vary by workspace settings, region, client version, and selected model", so your account may differ from someone else's. ## Why you may see an upload button but get a poor answer One attach button handles every file type, so having the button tells you nothing about video support. Even when a file is accepted, two details on OpenAI's own page can quietly weaken the answer. First, transcripts made from audio may contain errors, so a summary can be confidently wrong about who said what. Second, outside ChatGPT Enterprise, documents go through text-based retrieval, which "extracts digital text from the file rather than its embedded images". If you paste screenshots from a video into a PDF or Word file, ChatGPT may never see them. Upload stills as image files instead. If an answer could have been written without seeing your footage, treat it as a guess. ## A fallback that works: transcript plus screenshots This is our own advice, a rule of thumb. OpenAI doesn't promise it, and we make no claims about how well it performs. The idea is to give ChatGPT the formats its help pages say it supports: text, audio and still images. 1. **Get the words.** Use an existing caption file or transcript if you have one. On a paid plan, you can export the audio track as an MP3 or M4A file and upload it, then ask for a transcript. Check the transcript against the recording before you rely on it. 2. **Pick the frames that matter.** Take screenshots only where the picture carries meaning the words don't: a slide, a chart, a product on screen, a hand-drawn diagram. Note the rough timestamp of each one. 3. **Tell ChatGPT what it's looking at.** For example: "This is the transcript of a 20-minute product demo. The four images are screenshots at roughly 2, 7, 12 and 18 minutes. Summarise the main points and say what each screenshot shows that the transcript doesn't." 4. **Check the result against the video.** Anything about what happened on screen between your screenshots is guesswork. The same habit of breaking a source into readable parts and checking the output applies to [summarising a PDF](/questions/summarise-a-pdf/). If your video is a recorded meeting, [taking meeting minutes](/questions/take-meeting-minutes/) covers that case, including when a transcript belongs in your organisation's approved tool instead of a consumer chatbot. ## Check before you upload: faces, voices and other people's footage Most guides skip this. Under UK GDPR, personal data is any information relating to someone who can be identified directly or indirectly, and the ICO's guidance sets out that definition. Its video surveillance guidance says recordings can "capture information about identifiable individuals and how they behave", and that this is likely to be personal data. Faces and voices are exactly that kind of information. The ICO also says that processing "in the course of a purely personal or household activity, with no connection to a professional or commercial activity, is outside the UK GDPR's scope", giving the example of taking pictures for your own enjoyment. A clip of your own children on holiday sits on that side of the line. A recording of colleagues, customers, pupils or clients is connected to work, and the rules apply. Then there is what OpenAI does with the upload. Its privacy policy for the UK and EEA says it collects content you upload, including files, images, audio and video, and may use that content to train its models. You can opt out by turning off **Improve the model for everyone** in Settings, then Data controls. Temporary chats are not used to train models but may be kept for up to 30 days for safety purposes. OpenAI says it does not use content from Business, Enterprise or Edu workspaces for training by default. Video clips from Advanced Voice are kept for 30 days and are not used for training unless you switch on **Include your video recordings**. Also note that deleting a chat does not delete a file saved in your Library. You have to remove that separately. A rule of thumb from us: if you would be uncomfortable explaining the upload to the person on screen, don't upload it. Crop or blur bystanders out of your screenshots first; [editing photos](/questions/edit-photos/) covers using AI for that kind of job. Voice recordings carry their own risks, which we cover in [copying someone's voice](/questions/copy-someones-voice/). ## What the sources do not settle - **Whether ChatGPT understands an uploaded video file.** No OpenAI page we read says it does, and the supported formats table leaves video out. That is the strongest evidence available. It doesn't tell you what any particular app version does when you try. - **YouTube and other links.** None of the pages we read say ChatGPT can watch a video from a link. - **Family clips and the household carve-out.** The ICO pages describe purely personal use in general terms. They don't address sending a personal video to an AI company, so we can't tell you exactly where that sits. - **How well the fallback works.** It's a sensible method, and we haven't measured it. - **Next month.** OpenAI updates these pages often. The uploads page was updated within the last day when we checked, so read it again before relying on any limit here. ## Sources - [OpenAI Help Center: Uploading files and audio to ChatGPT](https://help.openai.com/en/articles/8555545-file-uploads-faq) - [OpenAI Help Center: ChatGPT Voice](https://help.openai.com/en/articles/8400625-voice-mode-faq) - [OpenAI Help Center: ChatGPT capabilities overview](https://help.openai.com/en/articles/9260256-chatgpt-capabilities-overview) - [OpenAI Help Center: What to know about the Sora discontinuation](https://help.openai.com/en/articles/20001152-what-to-know-about-the-sora-discontinuation) - [OpenAI Help Center: Data controls in ChatGPT](https://help.openai.com/en/articles/7730893-data-controls-faq) - [OpenAI: EU (EEA, UK and Switzerland) Privacy Policy](https://openai.com/policies/eu-privacy-policy/) - [ICO: What is personal data?](https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/personal-information-what-is-it/what-is-personal-data/what-is-personal-data/) - [ICO: Video surveillance guidance, what does this guidance address?](https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/cctv-and-video-surveillance/guidance-on-video-surveillance-including-cctv/what-does-this-guidance-address/) - [ICO: A guide to the data protection exemptions](https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/exemptions/a-guide-to-the-data-protection-exemptions/) ## Questions people ask ### Can ChatGPT watch a video I upload? OpenAI's file uploads help page lists documents, spreadsheets, presentations, images and audio, with no video formats on the list. It also says MP4 and WebM files identified as video are not supported by audio uploads. The safest route is to upload the audio track or a transcript, plus screenshots of the moments that matter. ### Can ChatGPT summarise a YouTube video from a link? None of the OpenAI help pages we read say ChatGPT can watch a video from a link. Don't count on it. Paste in a transcript or upload the audio, add a few screenshots, then ask for the summary. ### Why won't ChatGPT accept my video file? Video is not among the supported formats on OpenAI's file uploads page. Audio uploads accept audio-only MP4 and WebM, but files identified as video are not supported, and audio uploads are not available on the Free plan. Exporting just the audio as MP3 or M4A is the usual fix. ### Is it safe to upload a video of other people to ChatGPT? Footage of people who can be identified is likely to be personal data under UK GDPR. Purely personal or household use falls outside the law, but anything connected to work or a business does not. OpenAI's privacy policy says it collects what you upload and may use it for training unless you opt out, so check your data controls and think about the people in the clip first. ### Can ChatGPT edit or make videos for me? OpenAI's help pages we read describe no current video-making tool in ChatGPT. OpenAI says its Sora web and app experiences were discontinued on 26 April 2026 and the Sora API on 24 September 2026.