Gemini Notebook lets you edit the plan before it builds your slides or video

Google has added a co-creation flow to Gemini Notebook, according to a post on its blog. A new Review toggle lets users check and change the plan for a Slide Deck or Video Overview before anything renders. Google says the rollout reaches all users globally over the coming week.

A plan you can edit before rendering

Until now, Gemini Notebook went from a prompt straight to a finished artifact. The new flow adds steps in between. A user asks in the chat for slides or a video overview, and Notebook proposes topics drawn from the user’s own sources. After the user confirms, they can steer the content further and guide the visual style and format.

The Review toggle adds a second checkpoint. With it on, the plan appears in the studio panel. For slides, the user clicks “Review outline notes”; for video, “Review producer notes”. There the user can give feedback, and Google’s examples are adding an executive summary at the start or swapping the order of a scene. When the plan looks right, the user selects “Proceed” and Notebook produces the final artifact.

Two shortcuts remain. A user who already knows what they want can say so in the prompt, and the notebook goes straight to the review flow. A user who wants a quick result can open the studio panel, click the artifact, leave the Review toggle off and let Notebook generate it as before.

Who gets it, and what Google leaves open

Google says the capabilities are “rolling out globally to all Gemini Notebook users over the upcoming week”. The post names no plan restrictions, prices, usage limits or supported languages, so it is unclear whether the review flow counts against any existing generation quota or behaves differently across tiers. Since the rollout is staged, the toggle may not appear for everyone immediately.

The post also gives no figures on how much the flow improves results. Google frames the change as a response to user feedback, but it publishes no data on how often people were unhappy with first-pass output, how long the extra steps take, or whether edits to the outline are reliably reflected in the render. Those claims are Google’s and have not been independently tested.

Another step in a week of study-focused updates

The change follows a similar pattern to Google’s previous Gemini Notebook post. On Oct. 16, Google described three study features: 60-second Video Overviews for explaining a single hard concept, Interactive Learning Overviews that combine summaries with slides, quizzes and flashcards, and new quiz formats including short-answer, multi-select and fill-in-the-blank. A separate Google post from the same day, shared on Bluesky, again pointed to three new exam-season features.

Read together, the updates show Notebook moving from one-shot generation toward guided, multi-step work. The 60-second videos make the output shorter, the learning overviews bundle several outputs, and the Review toggle gives the user control over the plan. In all three, the sources stay the base: Google describes the content as grounded in the user’s own materials.

For readers, the practical difference is the cost of a miss. A slide deck or video that comes out in the wrong order or misses a key point previously meant rewriting the prompt and generating again. Editing a text outline first is a cheaper place to catch that, and it matches how many people already work with human designers or editors, who share a plan before producing the final piece.

What the post does not establish is how detailed the outline and producer notes are, whether users can edit them freely or only send feedback to the model, and whether the visual style choices extend beyond a preset list. Google’s walkthroughs for slides and video are embedded in the post, but the text does not describe them in more detail.