Yes. AI can edit unscripted footage by analysing what was said, what is visible, and how the recorded moments relate to the goal of the video. A script can provide useful structure, but it is not required to create a first cut.
For an unscripted project, the editing brief becomes especially important. It should tell the agent what the video is for, who will watch it, which ideas matter most, the desired length, and the tone or pace to aim for.
For example:
Create a three-minute customer story about how the team reduced delivery time. Open with the result, keep the most specific examples, and use the workplace footage as B-roll.
The agent can review the interviews and visuals, group related material, remove repetition, select useful takes, and arrange the strongest moments into a coherent sequence. This can work for interviews, testimonials, documentaries, event recaps, behind-the-scenes videos, video podcasts, and other projects that were not recorded from a fixed script.
The invideo agent for editing watches uploaded footage rather than relying only on a transcript. It can create a First draft with or without a script, search for supporting visuals, and return the work on an editable timeline.
An unscripted first cut still needs editorial review. AI may organise material logically while missing a more interesting emotional arc, a piece of production context, or an intentional moment of ambiguity. It cannot know an off-camera fact that was never included in the footage or the brief.
The clearest workflow is to give the agent a specific editorial objective, review the structure it creates, and then refine the story on the timeline.