Home · Personal · Writing & Creative · Photo & video
As of 13 August 2026, AI can turn a long video into social media clips.
Most people should do this themselves; the prompt is on this page.
Can you do it?
15 minutesto a draft.
30 minutesto something you’d act on.
Cost, all in£0/month
Skill needednone
Who has to check ityou
What the alternative costsA specialist video editor is the alternative; no price is stated in the supplied tool data.
If this goes wrong: the clips are dull, incorrectly captioned or poorly framed, so you replace them before publishing or take them down.
What to actually do
Do it yourself
The route this page recommends
A chat interface, no skill needed, and roughly 30 minutes until you can act on the result.
How to actually do it
- Open VEED or another AI video editor and upload the long video you own or have permission to use.
- Choose the target platform, such as TikTok, Instagram Reels or YouTube Shorts, and set the intended aspect ratio and maximum clip length.
- Paste the prompt and replace each bracketed instruction with your audience, aim, quantity, length, tone and branding details.
- Ask the tool to transcribe the video and propose the clips before committing to a final edit.
- Open the proposed clips and compare every cut, spoken quote and caption with the original video, correcting timestamps and words that are wrong.
- Check that faces, text and other important content remain visible after vertical reframing, and remove any music or footage you do not have permission to publish.
- Export the checked clips, write or edit the platform captions, and publish only the versions that match your intended audience and tone.
Prompt
I own or have permission to use the attached long video. Turn it into a set of social media clips for [PLATFORM]. The audience is [AUDIENCE], and the aim is [AIM]. Find the strongest self-contained moments, prioritising clear openings, useful or entertaining content, and moments that make sense without the rest of the video. Produce [NUMBER] clips, each between [MINIMUM] and [MAXIMUM] seconds where the source allows it. For each proposed clip, give me: 1. A short title. 2. The exact start and end time from the source video. 3. The first spoken line and the main point. 4. Why it could work for this audience. 5. A caption suitable for the platform. Do not invent words, events, claims or timestamps. Keep the speaker's meaning and wording. Flag any section where the audio, transcript or context is uncertain. Recommend vertical reframing where appropriate, keeping faces and important on-screen content visible. Suggest accurate burned-in captions, but mark words that need checking against the audio. Do not add copyrighted music, images or footage. Use this tone: [TONE]. Follow these brand details: [BRAND COLOURS, FONT OR OTHER RULES]. Return the clips in priority order and include a final checklist of anything I must verify before publishing.
Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.
Use a tool built for this
Second choiceHand it to a person
The distant thirdA person who owns the outcome does this end to end, worth it when the failure is dear.
What it gets wrong
- AI cannot reliably know which moment matters most to your audience without examples of what that audience responds to.
- It cannot replace your judgement about whether a clip makes the speaker look funny, harsh, misleading or out of context.
- It can mishear names, technical terms and overlapping speech, so captions and quotations need checking against the source.
- It cannot confirm that every person, song, image or piece of footage has permission for social publication.
- It does not guarantee that automatic reframing keeps the important person, product or on-screen detail in view.
Even on a YES, the friction has a name: taste, consent and privacy and context depth.
How we scored this
Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.
| Axis | Score (0–2) |
|---|---|
| Output | 2 |
| Inputs | 2 |
| Verification | 1 |
| Liability | 2 |
| Effort delta | 2 |
| Total | 9 / 10 |
The methodology and its thresholds are published in full.
FAQ
- Can ChatGPT turn a long video into short clips?
- It can help identify clip ideas, write timestamps and suggest captions, but a dedicated video editor is better for making and exporting the cuts. You still need to check the edits, captions and context against the original video.
- What is the best AI tool for making clips from a long video?
- VEED fits this task because it is a browser video editor with AI subtitles and cuts. Descript is another option when you want to edit audio and video through an AI transcript.
- Can AI choose the best parts of a video for social media?
- AI can propose moments with clear speech, strong openings or useful information. It cannot reliably judge your audience's taste or whether a short extract is misleading without your review.
- How do I make sure AI captions are correct?
- Play each finished clip while reading the captions and compare names, figures, technical terms and quoted words with the original audio. Correct the captions in the video editor before exporting or publishing.
Nearby answers
- Can AI create a UK passport photo from my photo?PARTLY
- Can AI detect whether a photo is fake?PARTLY
- Can AI find photos of a particular person in my library?YES
- Can AI identify where a photo was taken?PARTLY
- Can AI organise my photos by date and place?PARTLY
- Can AI remove background noise from a video?YES
Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.
The newsletter
AI news, new answers and product picks, straight to your inbox.