How to Make a Video with ChatGPT: Step‑by‑Step Guide for Beginners (2026)

Key Takeaways

• You can use the official invideo app from the ChatGPT store to generate videos from text prompts directly inside ChatGPT.

• Inside ChatGPT, the invideo app behaves like an AI production team: it turns your prompt into a script, breaks it into scenes, chooses or generates visuals, adds AI voiceover and music, and returns a complete draft video.

• You refine that draft in two layers: first by giving follow‑up instructions in ChatGPT (for structural changes), then by opening the project in invideo when you want full control over the timeline, captions, branding, and exports.

• This workflow is especially useful for marketers, small teams, social media managers, YouTube automation creators, and writers who want to turn articles or scripts into videos without learning traditional editing software.

To start generating videos with the invideo app inside ChatGPT, you first need to do this:

1. Open the ChatGPT apps section.
2. Search for “invideo”.
3. Install or enable the official invideo app.
4. Connect your invideo account if you already have one, so projects created from ChatGPT are saved to your workspace.

Before you ask for a video, give ChatGPT enough context to create something useful.

In your conversation, answer three questions:

Who is this for?

For example, “beginners in personal finance,” “new users of our SaaS product,” or “first‑time home buyers.”

What outcome do you want?

Do you want them to understand an idea, save the post, click through, sign up, or try a product?

Where will it live?

A 9:16 vertical Reel or Short, a 16:9 YouTube explainer, a short ad, or a website video.

This is what an example prompt will look like:

Help me create a 45-second vertical video for beginners in personal finance about ‘how to make your first budget’. The video will be an Instagram Reel. Use a strong hook, 2–3 simple tips, and one clear CTA at the end.

This tells ChatGPT and the invideo app how long the video should be, who it is for, and how it will be used.

Next, you need a script that can be turned into a video.

Ask ChatGPT directly:

Write a 45-second 9:16 script titled ‘How to make your first video with ChatGPT’. Start with a hook in the first 3 seconds, then explain 3 simple steps, and end with a one-line CTA to follow for more tips. Keep sentences short and easy to caption.

ChatGPT will generate a script that fits your length and platform. Read it once as if you were speaking it aloud. Edit any lines that feel off‑brand, too complex, or unnatural.

Short, simple sentences are easier for viewers to follow and easier for invideo to turn into readable captions and smooth pacing.

If you want more control over what appears on screen, follow up with:

Now break this into 6 scenes. For each scene, give the exact line of dialogue and one sentence describing what the viewer should see.

This scene breakdown gives the invideo app a rough storyboard to follow when it generates visuals and transitions.

Once the script looks right, you can ask invideo to generate the actual video draft from within the same chat.

For example:

Use this script to generate a 45-second 9:16 vertical video. Add bold on-screen captions, a neutral AI voiceover, and simple background music that fits beginners. Make the hook text large and easy to read on mobile.

From this request, the invideo app will:

• Turn the script into scenes.

• Select or generate visuals for each moment.

• Add an AI voiceover in your chosen language.

• Place captions on screen.

• Add background music and basic transitions.

When it is done, ChatGPT shows you a preview of the generated video and a way to open the underlying project in invideo.

You can refine the draft from inside ChatGPT by giving new instructions, such as:

Remove scene 3 and add a CTA slide at the end.

Change the voiceover to a female English voice.

Replace the city skyline visuals with home-office footage.

Make all captions bigger and higher contrast.

The app applies these changes, regenerates, and returns an updated video. This gives you a way to shape structure, pacing, and tone without touching a timeline.

Once you are happy with the generated version in ChatGPT, you can move into invideo for detailed editing and export.

By opening the project in invideo, you see the same scenes, captions, and structure laid out on a proper timeline. From there, you can:

• Tighten or reorder scenes for better pacing.

• Swap specific clips, images, or backgrounds.

• Apply your brand fonts, colours, and logos via invideo’s brand kits.

• Adjust caption style and placement for mobile readability.

• Add extra b-roll, overlays, transitions, or sound design.

• Export multiple versions (9:16, 16:9, 1:1) from the same base project.

Edits and exports here behave like any other invideo project and follow whatever limits apply to your plan.

The mental model is simple: use ChatGPT and the invideo app to jump from idea to watchable draft, then use invideo to take that draft through your normal workflow and out to your channels.

To get the most out of this setup, it helps to go in with clear expectations.

• Video quality will depend heavily on your prompt and script. Vague instructions tend to produce generic visuals.

• AI voiceovers are useful and fast but can still sound slightly artificial, especially in longer videos.

• Because many visuals come from stock or AI-generated assets, it is possible that other creators will use similar footage. So be specific.

• The ChatGPT app does not bypass invideo’s plan limits; free tiers can include watermarks, credits, and export caps.

• For high-end cinematic work, complex animation, or frame-perfect timing, dedicated editing tools still offer more precise control.

This workflow excels when it comes to speed and simplicity. It is ideal when you want to go from “I have an idea” to “I have a solid draft on a timeline” in minutes, not hours.

1. Do I have access to editing tools inside ChatGPT?

You have generation stools, not a full editor. You can adjust your instructions and ask the invideo app to regenerate drafts, but detailed control over scenes, timing, and branding lives in the invideo editor.

2. What if I also have access to Sora inside ChatGPT?

Sora is useful for generating raw clips and concept shots directly from text. The invideo app and editor are where those clips get turned into full videos with structure, captions, music, and platform‑ready exports. You can use ChatGPT to generate both the script and any Sora‑style footage you need, then let invideo assemble everything into a single project.

3. Is invideo inside ChatGPT only for faceless videos?

No. It works well for faceless videos, but you can also use it for avatar‑style content. For faceless videos, lean on b‑roll, stock footage, and bold text overlays driven by your script. For avatar‑style videos, use ChatGPT to write the script and structure, then let invideo generate scenes with AI voiceover, visuals, and pacing that match your narrative.

4. What if I just want to tweak one tiny detail when using invideo inside ChatGPT?

If it is about the overall generation, music style, caption size, and length, then try adjusting your instructions and regenerating with the invideo app in ChatGPT. If it is a precise cut, brand layout, or timing detail, open the project in invideo and change it directly on the timeline.