Google looks set to bring a planning mode to NotebookLM’s Video Overviews, a control that has surfaced inside a recent build of the research tool. The option lives in the customization menu, the same panel, reached via the pencil icon on the Video Overview tile, where users already pick a format, choose a visual style, and write a custom prompt. A new toggle would switch on the planning step.
The exact layout is still taking shape, but the behavior closely resembles the plan-then-build pattern familiar from coding assistants. Rather than sending a prompt straight to rendering, Gemini would first draft a plan for what the video should cover and pause to let the user edit and approve it before generation begins. For educators, researchers, students, and anyone turning dense sources into watchable explainers, that checkpoint matters: it adds editorial oversight over structure and pacing, and it heads off wasted generations on a clip that misses the point. Today, NotebookLM hands those calls to Gemini, acting as a silent creative director, with no place for the user to step in.
The toggle also hints at something larger under the hood. The capability aligns with Gemini Omni, the multimodal model Google introduced at I/O 2026 in May, which now serves as its default video engine and can generate explainer-style clips from a single prompt. Moving Video Overviews from the current Veo-based stack to Omni would fit Google’s push to consolidate text, image, and video into one “anything from anything” system, and a planning step is the kind of control Omni’s editing-first design naturally supports. Such a shift would likely accompany any broader upgrade to the video pipeline.
No timeline has been attached to the feature, and the planning mode remains in development for now.
Tip by @thomas_gmry
