Gemini AI can turn prompts into picture books, but I still prefer Paddington
- Gemini’s Storybook feature lets you instantly generate 10-page illustrated storybooks
- You can pick art styles and themes
- The results can be cute but are far from the quality of beloved classics
If you have a kid who loves to hear about themselves in a story, Google’s Gemini AI has a new trick that could keep them happy for a long time. Gemini’s new Storybook feature lets you generate fully illustrated, ten-page storybooks with narration from a single prompt.
You describe the tale, the look you want, and any other details, and Gemini writes the story, creates images for each page, and reads it aloud within a few minutes.
Storybook, in some ways, just combines existing abilities like text composition, image generation, and voice narration. Still, by putting them into a single prompt system, it speeds up the final product enormously. If you don’t like certain details of the look or writing, you can simply adjust the book with follow-up prompts. You can even feed it a photo to shape the setting or characters.
The appeal for those who might feel they lack creative writing or drawing skills is obvious. No need to hire an illustrator or record voiceovers yourself. If your child wants a bedtime story about a shy dragon who finds confidence at music camp, you type that in, and within minutes, you’ve got a book with pictures, narration, and page-by-page structure.
This isn’t just for bedtime, either. Teachers can create customized stories to explain hard topics, perhaps teaching second graders about gravity with a friendly astronaut cat. Therapists could use storybooks to help kids talk through emotions using characters they connect with. Aunts and uncles can make personalized birthday stories with inside jokes and family pets.
What used to be a labor-intensive creative project is now something you can do on your phone during lunch break.
AI storytellers
And it is a notable shift from the standard template with a blank to fill in approach common to other AI tools. The narration even adapts to the tone of the story, with voices that can be whimsical, soothing, or dramatic, depending on what your story needs. Google is pitching the tool to busy parents, overworked teachers, and creative kids looking for a co-author and illustrator for their ideas.
I asked Gemini to make a story about my dogs going on an adventure in nature, sharing their names and describing their looks, and that’s about it. You can read and listen to the Gemini-created story here.
It did a remarkably good job, albeit with a very inconsistent look to the dogs from page to page and a somewhat dull story. And when I tried it again to see how it would perform with the same prompt, the dogs sometimes had more than four limbs, not exactly reassuring to a child looking forward to a story about their pets.
And while it’s theoretically possible that Gemini could write and illustrate a story better than the many classic and modern children’s books out there, or one more personally resonant than writing it yourself, I personally have doubts. This is a fun little trick, but the idea of skipping every bookstore, library, and box of crayons and pencils for an AI alternative that can’t always even make your dog look the same on every page feels like the exact activity I’d rather do myself. I’ll stick to asking AI for help organizing my kitchen and leave the bedtime stories to me.
Add comment