Google Unveils Whisk: An AI Tool That Turns Images into Creative Prompts

Google Unveils Whisk: An AI Tool That Turns Images into Creative Prompts

Google has introduced Whisk, a new AI experiment that uses images as prompts to generate unique visuals. Unlike traditional AI tools that rely heavily on lengthy text inputs, Whisk lets users provide images to define the subject, scene, and style of the output. Users can also mix and match multiple images for each category, opening up new avenues for creativity.

How Whisk Works

Whisk allows you to:

  1. Upload images for prompts like subject, scene, and style.
  2. Generate prompts automatically using a dice icon, where Google supplies AI-generated images as starting points.
  3. Iterate and refine by adding optional text inputs or tweaking the suggested image prompts further.

The tool produces both AI-generated images and the accompanying text prompts. If you’re happy with the result, you can favorite or download the image. To refine it, users can edit the prompts or ask Whisk to adjust details directly.

Google emphasizes that Whisk is not intended for “pixel-perfect edits” but rather for rapid visual exploration. While the results might occasionally miss the mark, the tool’s ability to iteratively refine images makes it a fun and interactive experience.

Powered by Imagen 3

Whisk is built on the latest version of Google’s Imagen 3 image generation model. Imagen 3, announced alongside Whisk, promises improvements in realism and creative interpretation. Google also introduced Veo 2, its updated video generation model, which better understands cinematography and reduces common AI errors like hallucinating extra fingers.

Experimenting with Whisk

Whisk’s intuitive approach makes it accessible for both casual creators and professionals. Whether you’re exploring ideas or refining specific visual concepts, the ability to use images as prompts sets Whisk apart in a growing sea of generative AI tools.

While image generation currently takes a few seconds, Google is positioning Whisk as an engaging tool for rapid experimentation, offering a playful balance of control and unpredictability.

Google’s Whisk and Imagen 3 are available now for exploration. The company’s next-gen tools, including Veo 2 for video generation, are rolling out gradually, starting with Google Labs’ VideoFX. With these advancements, Google continues to push the boundaries of AI-driven creativity.