Whisk proves you don’t need words to create

GOOGLE

Whisk proves you don’t need words to create

Google has launched Whisk, an AI tool that creates images using other images as prompts instead of long text descriptions.

It offers more control and flexibility when generating visuals.

With Whisk, you can upload images to guide the AI on the subject, scene, and style.

You can use multiple images to refine the results or add optional text for extra details.

If you don’t have images, a dice icon can generate AI-suggested visuals to get you started.

Here’s what you should know:

• Whisk lets you create visuals using images instead of just text.

• Combine images and optional text for more refined results, or use suggested visuals from Google.

• The tool is made for quick creative exploration, not detailed edits.

No words, no problem

Google says Whisk is designed for “rapid visual exploration” and may not always be perfect.

If needed, you can tweak the prompts to improve the results.

The tool uses Google’s new Imagen 3 model, which was announced at the same time.

Google also introduced Veo 2, an updated video generation model.

Veo 2 is better at understanding cinematic concepts and avoids common AI errors like adding extra fingers.

It will first be available on Google’s VideoFX through the Labs waitlist and will expand to YouTube Shorts next year.

This shows Google’s efforts to make AI tools more creative, flexible, and easier to use.

Whisk might just be the Canva of AI art.