Google Unveils Whisk: An AI Tool Powered by Image Prompts

Google has introduced Whisk, a new artificial intelligence tool that allows users to upload photos to create AI-generated images without needing text prompts. The tool enables users to combine elements of uploaded images, such as subjects, settings, and styles, into a single creative output.

Whisk is described as a “creative tool” rather than a professional image editor, designed for quick inspiration and playful experimentation. Users can also “remix” their results by tweaking inputs to create variations like plush toys, enamel pins, or stickers. While text input can be added to refine details, it’s not mandatory for generating images.

How Whisk Works

Whisk uses Google’s Gemini AI platform, launched in December 2023, alongside Imagen 3, the latest text-to-image generator from DeepMind, Google’s AI lab. When users upload images, Gemini generates captions, which Imagen 3 interprets to produce an image. Rather than creating exact replicas, the tool captures the essence of the input, allowing for creative reinterpretations.

However, this flexibility means the final output may diverge from the original prompt, resulting in differences like changes in height, hairstyle, or skin tone.

A Fun Approach to AI

“Whisk is designed to let users remix subjects, scenes, and styles in new and creative ways, offering rapid visual exploration rather than pixel-perfect edits,” said Thomas Iljic, Director of Product Management at Google Labs.

Google sees Whisk as an entry point for consumers to experiment with generative AI, focusing on creativity rather than accuracy.

Competition in the AI Space

The launch of Whisk underscores the competitive landscape of generative AI, with tech giants racing to release consumer-friendly tools. Google’s announcement follows OpenAI’s launch of Sora, a text-to-video generator, signaling a push to dominate the AI market.

Dan Ives, Senior Equity Analyst at Wedbush Securities, called Whisk a “flex the muscles moment” for Google, highlighting DeepMind’s role as a key asset. “AI tools like Whisk are part of Google’s treasure chest of innovations for 2025,” Ives said, referencing Google’s plans for future products, including a new Android operating system developed with Samsung and Qualcomm.

Availability and Future Plans

Currently, Whisk is available as a web-based tool through Google Labs for U.S. users and remains in its early stages of development. Google aims to expand and refine the tool based on user feedback while continuing to innovate in the AI space.

Whisk’s debut represents a broader push to integrate AI into everyday consumer experiences, blending creativity and technology in new, accessible ways.