Can ChatGPT control After Effects directly?

No. ChatGPT in a chat window has no way to see your composition or click inside After Effects' panels. It can write you an expression or a script, or describe which panel to open, but it isn't watching the comp while it does it, and it can't tell if the step actually worked.

Is there a ChatGPT plugin for After Effects?

What you'll find are panels and extensions that call an AI model from inside After Effects for a specific task, not an agent that holds the whole comp and plans a multi-step edit. That's a narrower thing than what people usually mean by "use ChatGPT to edit in After Effects."

Can ChatGPT write After Effects expressions?

Yes, that part works well. Describe the motion or behavior you want and ChatGPT can draft the expression or script; you still paste it in, run it on your layer and check the result, since it can't see whether it did what you asked.

How do I remove an object in After Effects with AI?

After Effects has its own Content-Aware Fill panel for object removal and Roto Brush for masking, with Mocha AE built in for planar tracking. ChatGPT can talk you through the steps but it isn't the one running them.

How do I get a ChatGPT-edited cut into After Effects?

Edit on a timeline the agent can actually read and write, then export a format After Effects' host project can use alongside it, or hand off through EDL, FCPXML or OpenTimelineIO for the cut itself. In Shoty, the agent behind the ChatGPT option edits the project directly over a local MCP server.

ChatGPT After Effects: What It Can Script, Not Click

Ask ChatGPT to "clean up the background in this After Effects shot" and you'll get a walkthrough of Content-Aware Fill, or an expression to paste onto a layer. You still open the panel, run it and judge whether it worked, because ChatGPT never saw the comp. Ask the same thing to an agent that's actually holding a timeline, and the fix happens, then the shot is back in place. The difference isn't the model, it's what's on the other end of the sentence.

Can ChatGPT actually drive After Effects?

Honestly, not on its own. ChatGPT in chat has no eyes on your composition and no way to press a button inside After Effects' window. The paths people try to bridge that gap:

Expressions and scripts. Describing the motion or behavior you want is something ChatGPT is genuinely good at turning into an expression or script. You still paste it onto the layer, run it and check whether the result matches what you asked for.
Plugins and panels inside After Effects that call an AI model for one task, a specific effect or a generated element. Useful for that task, but it isn't a timeline-aware agent planning a multi-step edit across a comp.
After Effects' own tools, run by a person: Content-Aware Fill for removing an object, Roto Brush for a mask, Mocha AE (bundled in) for planar tracking. ChatGPT can tell you which panel to open; it isn't the one opening it.

None of these give ChatGPT a comp it can look at, reason about and edit the way it reads and edits code. That's the gap "chatgpt after effects" searches are actually running into.

The route that gives it a real timeline

Shoty is a desktop video editor (Mac and Windows) with an agent chat built in. Pick ChatGPT in the model picker and the shape of the problem changes:

Shoty runs the official Codex CLI on your machine, signed in with your own OpenAI account. It never sees your token and charges no fee for the agent.
For each turn, Shoty opens a local MCP server and hands the agent tools: look at a frame, read and update the timeline, generate a shot, export. Every call shows up in the chat.
The timeline is a text format (.shoty), so the agent is reading and editing a structure instead of guessing at an expression's result or hoping a pasted script ran clean.
When the cut is done, Shoty exports it as EDL, FCPXML or OpenTimelineIO, formats a finishing project can pick up for the final grade.

The agent never touches After Effects directly in this flow. It edits a project on its own timeline, including the per-frame VFX work After Effects' own tools would otherwise handle by hand.

1. Do the fix in Shoty, not in After Effects

Ask for the object removal, the background swap or the mask while the project is still in Shoty.

Remove the boom mic that drifts into frame in shot 4. Track it through the whole clip and rebuild the ceiling behind it.

2. Name what must not change

Replace the background behind the dancer with a neon street at night. Keep her, her motion and the camera move exactly as they are.

That's the clip above: a sofa background swapped for a rooftop at dusk, the people and camera move untouched. Naming what stays locked is what keeps the model from repainting the whole frame.

3. Export when the cut is done

Export this project as an EDL.

Ask for FCPXML or OpenTimelineIO instead if your next stop reads one of those formats.

4. Finish wherever you already work

Grading, type and final delivery can still happen in your usual pipeline. Shoty's job ends at the cut and the AI shots.

Where this still needs you

Judging the mask and the fix. The agent tracks the object and runs the removal; whether the edges and the shadow hold up at full size is still your call.
Reviewing generated shots. Anything sent to a video model costs credits, so set the run mode to review if you want to approve shots before they render.
Fine motion work. After Effects' own Roto Brush and Mocha AE are built for hand-tuned per-frame tracking; for that level of manual control, that's still the right place to do it.

Download the free Shoty editor and connect ChatGPT from the model picker. For the Claude side of the same setup, see Claude Code After Effects, or for the Premiere Pro version, ChatGPT Premiere Pro.