AI is the mortar, not the house.
Summary: AI in post-production editing has moved from novelty to daily workflow, and it is changing how creators work.
Let’s be real: we’re all-consumed and saturated with AI, almost to a point where it’s often unclear what online content is created by humans and AI. As my teammate Kiko Toledo explored the basics and myths behind AI in a recent blog post AI 101: Predictably Unpredictable, AI is not a single all-and-powerful solution, it’s a tool that is as good as its source, and it’s often wrong.
Is AI writing this blog? No. Is it copy editing this blog? Sometimes. But even with a copy edit pass that takes ChatGPT a cool two seconds to render, it often suggests a better word, but often it’s a word that’s just not what I would use or say, so it didn’t go in this post.
When you press pause on the ethics of AI, including ideals of intellectual property and human creativity, AI is a tool for digital creators, like a saw is for a carpenter. There are apps like Runway.ml and Eleven Labs that we evaluate and use in post-production editing that are flat-out amazing. These apps can transform a layman video editor into a wizard of color correction, upscaling video quality and mixing perfect audio. Yet AI is often wrong, and the human makes the final decision on accuracy of story, clips, scenes, color, audio, captions.
AI tools in our post-production editing workflow:
- Runway for video effects, color, and quality upscaling
- ElevenLabs for AI text to speech and voiceover
- ChatGPT for copy editing passes
Is AI taking creative jobs? You might not like the answer.
There are corners of the creative world that are quaking because of AI.
Apps are unlocking new ways to build animation and visual effects. Wonderkid @ErikdoesVFX of Norway hilariously recreated a $10M VFX scene from Star Wars in just a week using apps anyone can access. Could it be done without AI? No. Could AI have created this scene without a source? No. At the core, AI completes a task. It didn’t originate the love saga of Rey and Kylo Ren or how they came to battle light saber to TIE whisper in Star Wars: The Rise of Skywalker.
Even with humans behind the concepts, ideas that make a blockbuster movie tick, there are elements that are being stripped from the creative community. One clear example is voice acting. We use AI app Eleven Labs which generates a WAV audio file from a script (called text to speech). We choose the voice based on the needs of the video, and the result is decent. The upside to tools like Eleven Labs is you can create the VO and adjust last-minute script changes without needing to hire and re-record a voice actor.
Text to speech doesn’t start with AI. Voice actors submit their voices in return for a small royalty (about 3 cents per 1K words) with the promise that AI as a force multiplier, the voice actor will realize more usage and theoretically create more payout. I think it’s way too early to understand the financial impact within the creative community, but you’ll find a lot of lively exchanges on Reddit.
How AI Lowers the Barriers to Entry in Post-Production Editing
Just as high tech gave me the ability to edit short films in my kitchen twenty-some years ago, AI is powering creators in post-production editing with cost-effective apps. But behind the app, you need the idea, a cogent story and compelling narrative. In feature films, AI is enabling creators to build realistic crowd shots and monsters, but AI doesn’t make a movie a good movie, that’s all up to the talent of the creator.
Next up in our AI in production series, executive producer John McDonald shares the top 5 AI tools we use in post-production.
AI in Post-Production Editing FAQs
Does AI replace voice actors?
Not entirely. Tools like ElevenLabs generate voiceover from a script using text to speech, which lets us adjust last-minute script changes without re-recording. But voice actors still supply the source voices behind these tools, and it is too early to know the full financial impact on the creative community.
Can AI edit video on its own?
Not on its own. Apps like Runway can upscale video, correct color, and mix audio, but AI is often wrong. A human still makes the final call on story, clips, scenes, color, audio, and captions.
Is AI creating this content for us?
No. AI handles tasks like a copy edit pass, but it does not originate the idea, the story, or the narrative. That part still belongs to the creator.
Will AI replace video editors?
It is not on track to. AI in post-production editing speeds up repetitive tasks like upscaling, color passes, and rough voiceover, but it cannot supply the judgment, taste, and storytelling that make an edit work. The editor stays in the driver seat and owns the final cut.
How do I start using AI in post-production editing?
Start small. Pick one repetitive task, such as cleaning up audio or generating a scratch voiceover, and test a single tool on a low-stakes project. Keep a human review step on everything the tool produces, since AI is often wrong, and expand only once you trust the output.
Is using AI in post-production editing ethical?
It depends on how the tools are sourced and used. The two biggest questions are intellectual property, meaning what the AI was trained on, and fair compensation for the artists whose work feeds these tools, like the voice actors behind text to speech. A reasonable standard is to keep a human accountable for the final output and to favor tools that pay their source contributors.