Generative AI is impressively capable, and keeps getting better. But in some ways, it’s a little like trying to plan spaceflight with a paper map. Sure, models and tools like Gemini and Veo 3 can do practically anything. But getting exactly what you want out of them, with a relatively narrow focus, can be hit-or-miss. That’s why tightly scoped features like Gemini’s Audio Overviews work so well: you feed in your source material, and you get a polished podcast-style summary out the other end. Easy. Predictable. Useful.
Now, it looks like Google’s lining up another targeted experience for Gemini users, and this one could be a hit with parents and educators alike — AI-generated Storybooks (Source: Android Authority).
Age-appropriate image and video generation
Ideally, not just AI slop for kids
Evidence of the Create a Storybook and PDF uploading options. Courtesy of AssembleDebug via Android Authority.
The first hint of this new feature surfaced a couple of months ago in version 16.19.42.sa.arm64 beta of the Google app, buried alongside chips for Timeline and Mindmap (which already exist in NotebookLM). At the time, “Storybooks” was just a name — no functionality, no UI, just code strings pointing at something Google was tinkering with.
Fast-forward to today and things are starting to get a bit more concrete. Recent reports from TestingCatalog have uncovered references to a “Storybook” Gem in the web version of Gemini, and in AssembleDebug’s dive into version 16.26.64.sa.arm64 of the Android app, he spotted even more resources hinting at the feature’s evolution.
There’s now an animation suggesting what this Storybook experience might look like: AI-generated illustrations paired with narrative text, packaged together in a format clearly aimed at younger audiences. Think bedtime stories, but conjured up on demand with a little prompt and a dash of Google magic.
As for functionality, nobody’s quite sure yet. AssembleDebug’s reverse-engineering coaxed Gemini into showing a “Create a Storybook” chip, but only under certain conditions, like feeding the model an image as a prompt. Timeline and Mindmap options also appeared when uploading a PDF, though they’re still not functional within the Android app.
Interestingly, some Gemini users have already been experimenting with similar projects through Google’s own AI competition platform — including two tools literally named Storybooks and Storybook AI Bedtime Stories. These community-built experiments might not be directly linked to what Google’s building internally, but the thematic overlap is telling.
If Storybooks can harness even a fraction of the visual fidelity we’ve seen in Google’s Imagen and Veo models, it could be something genuinely delightful. We’re keeping an eye on it — and crossing our fingers that the next Gemini update gives us a real taste of this whimsical new feature in action.
