Fun social clips and sizzle reels are an incredible way to drive engagement for your brand. They're also, let's be honest, a massive pain to make. Animating anything from scratch usually requires skills most of us don't have, time we definitely don't have, and software we don't know how to use

But now you can just generate it in Descript.

Why you'd want to do this

Maybe you need a stylized product launch video but your budget is more "scrappy startup" than "Super Bowl ad." Maybe you want to experiment with different visual styles without hiring five different animators. Or maybe you just want to make something that looks expensive and polished without the expensive and polished part.

Whatever the reason, generative video in Descript means you can create professional-looking motion graphics without knowing how to animate. You just describe what you want, and it appears. It's honestly a little weird how easy it is.

How to create AI-generated video in Descript

Let's walk through making a stylized montage—something like a guy walking through an airport in different art styles. Normally this would be impossible for most of us. In Descript it's... un-impossible.

1. Generate your base image

Go into the AI tools panel and select Generate an image. Choose the AI model you want to use, then write your prompt. Here's where specificity matters—the more detailed you are, the better your result. Don't just say "airport scene." Say "businessman in navy suit walking through modern airport terminal, dramatic side lighting, wide angle, cinematic composition." Then, hit the magic wand icon to select the visual style you want, or add a new style to the prompt itself.

If you're staring at a blank prompt box thinking "I have no idea what to write," Underlord is actually great at helping you craft strong prompts. Just tell it what you want—"person walking through an airport"—and it'll come up with a really good prompt that you can paste in.

‎

2. Turn that image into video

Once you've generated an image that works, select it in the canvas. Choose "Turn it into a video." Your static image now becomes the reference point for video generation.

‎Next, type in your motion prompt—describe how you want the scene to move. Then generate the video. That's it. You now have an animated clip.

3. Make multiple variations

Here's where it gets fun. Create new compositions and repeat this process for each art style you want. Generate the same scene in watercolor, then cyberpunk, then film noir, then whatever else strikes you. Each one takes a minute or two.

4. Assemble your final piece

Once you have all your generated clips, create a new composition and drop them all in. Cut them together, adjust timing, add sound design. You now have a sizzle reel ready for social.

‎The whole thing happens in one place

This is the part that actually matters most. You generate your video clips and edit them in the same workspace. No exporting to After Effects. No juggling between five different apps. No "wait, which version did I export?"

Whether you're making a quick social post or a full product launch reel, your entire workflow stays in Descript. Which means less time managing files and more time actually making things.

What else Descript does

Generative video is new and exciting, but it's just one tool in Descript. The platform does every stage of video editing with AI-powered transcription, which means you can edit your video by editing text. It also includes AI tools like Studio Sound for cleaning up audio, Eye Contact to fix your gaze, filler word removal, and it's all collaborative, so you can work on it with a team if you want to.

Basically, it's designed to streamline content creation from recording to export. The generative video stuff is just the latest addition to an already pretty stacked toolkit.