Can ChatGPT control DaVinci Resolve directly?

No. ChatGPT in a chat window has no way to see your Resolve project or click inside its panels. It can write a script against Resolve's own Python/Lua API or describe which menu to open, but it isn't watching the timeline while it does it, and it can't tell if a step actually worked.

Can ChatGPT write DaVinci Resolve scripts?

Yes, that part works well. Resolve ships a Python/Lua scripting API, and describing what you want is something ChatGPT is genuinely good at drafting a script for. You still paste it into Resolve's console, run it and check the result, since ChatGPT can't see whether it did what you asked.

Is there a ChatGPT plugin or MCP server for DaVinci Resolve?

Community MCP servers that wrap Resolve's scripting API exist, built by third parties, not Blackmagic or OpenAI. Check what a given one actually exposes before pointing it at a real project. A plugin that calls an AI model for one Resolve task is narrower than an agent that plans and executes a multi-step edit.

How do I get a ChatGPT-edited cut into DaVinci Resolve?

Edit on a timeline the agent can actually read and write, then export a format Resolve imports: EDL, FCPXML or OpenTimelineIO. In Shoty, the agent behind the ChatGPT option edits the project directly over a local MCP server, and the cut leaves in one of those formats.

Does this need an OpenAI API key?

No, if you're using Shoty. It runs the official Codex CLI on your machine, signed in with your own OpenAI account, and never sees the token. OpenAI's own plan limits and policies still apply.

ChatGPT DaVinci Resolve: What It Can Script, Not Click

Ask ChatGPT to "remove the dead air from this Resolve timeline" and you get a script to paste into Resolve's console, or a description of which panel to open. You still run it and judge whether it worked, because ChatGPT never saw the timeline in the first place. Ask the same thing to an agent that's actually holding a timeline, and the cuts happen, then the project lands in Resolve as an EDL. The difference isn't the model, it's what's on the other end of the sentence.

Can ChatGPT actually drive DaVinci Resolve?

Honestly, not on its own. ChatGPT in chat has no eyes on your project and no way to press a button inside Resolve's window. The paths people use to bridge that gap, for Resolve specifically:

Resolve's own Python/Lua scripting API. Resolve ships this, so a script ChatGPT writes has something real to call into: add a clip, move a cut, trigger a render. Writing the script is something ChatGPT is genuinely good at; running it and checking the result is still on you.
Community MCP servers that wrap that scripting API so an agent can call it as a tool instead of writing raw script files. These are built by third parties, not Blackmagic or OpenAI, so check what one actually exposes before trusting it with a real project.
Resolve FX and Resolve's built-in AI tools (Magic Mask, Object Removal, Patch Replacer) do real work inside the app, some needing Resolve Studio, but they're tools you run yourself from the Fusion or Color page, not something ChatGPT is operating.

None of these give ChatGPT a project it can look at, reason about and edit the way it reads and edits code. That's the gap "chatgpt davinci resolve" 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 calling Resolve's scripting API one function at a time.
When the cut is done, Shoty exports it as EDL, FCPXML or OpenTimelineIO, the formats DaVinci Resolve, Premiere Pro or Final Cut Pro open for the finish.

The agent never touches Resolve directly in this flow. It edits a project Resolve can later read, which sidesteps the scripting API entirely for the part the agent is actually doing.

1. Do the cut in Shoty, not in Resolve

Ask for the rough cut, the VFX fix or the subtitle pass while the project is still in Shoty.

Cut the dead air out of the first five minutes, keep room tone under the cuts, and add burned-in English subtitles.

2. Name what must not change

The clip above swaps the background behind the people on the sofa for a rooftop at dusk, keeping them, their motion and the camera move exactly as they were. That's the instruction pattern for any VFX fix before export:

Replace the background behind the two people with a rooftop at dusk. Keep them, their motion and the camera move exactly as they are.

3. Export for Resolve

Export this project as an EDL for DaVinci Resolve.

FCPXML and OpenTimelineIO work too; Resolve reads all three. If you also want to keep scripting inside Resolve after the import, its Python/Lua API is still there and separate from this workflow.

4. Finish where you already work

Color and mix happen in Resolve like they always have. Shoty's job ends at the cut and the AI shots; Resolve's part of the job doesn't change.

Where this still needs you

Picking the beat. The agent finds every pause fast; which one actually lands is still an editor's call, and you can always grab the Shoty timeline and trim by hand.
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.
Round-trips. If notes from the Resolve side mean a shot needs another pass, that happens back in Shoty, not by re-scripting Resolve.

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