Google Whisk uses images instead of text for AI photo generation

Google has recently announced Whisk, a new generative AI tool which allows users to create images by using other images as prompts rather than relying on detailed text descriptions. 

Whisk enables users to combine images for the subject, scene, and style, generating new visuals through an automated process. Google's Gemini AI model writes detailed captions for the uploaded images, which are then processed by its Imagen 3 image generation model. The result captures the essence of the input images while allowing for creative reinterpretations.

Google stated that early testers, including artists and creatives, have described Whisk as a creative tool rather than a traditional photo editor. 

According to Google, Whisk is designed for "rapid visual exploration" rather than precision editing, making it ideal for concept development and creative experimentation. Users can generate and remix ideas, such as creating enamel pins or stickers, and quickly test multiple variations before downloading their preferred results.

The experimental tool, released in the United States, is part of Google Labs' ongoing efforts to advance AI-based creative solutions. The tool is not yet available in Bangladesh.