Google debuts Whisk AI: An image generator with a quirky twist

Just a heads up, if you buy something through our links, we may get a small share of the sale. It’s one of the ways we keep the lights on here. Click here for more.

Google has unveiled Whisk, an AI image generator tool that simplifies creating visuals by focusing on images rather than text prompts. 

Unlike traditional AI image generators that require carefully crafted text inputs, Whisk allows users to drag and drop existing images into the tool. 

It then blends these inputs to produce entirely new images, making it far more accessible and intuitive for users who lack expertise in prompt engineering.

Google Whisk lets you combine multiple photos to make something new

Google announced this new AI tool in a blog post. At the core of Whisk’s functionality is Google’s Gemini AI, which works in a team with its advanced image-generation tool, Imagen. 

The process begins with Gemini’s automatic captioning, which analyzes and converts the uploaded images into detailed descriptive prompts.

These prompts are then fed into Imagen 3 through a step called generative remixing, where the AI produces visual outputs inspired by the input images rather than replicating them exactly. 

Users can leverage editable prompts to further refine the results, tweaking the generated suggestions to achieve desired variations.

Google emphasizes that Whisk captures the essence of an image instead of a pixel-perfect replica, which means outputs can differ significantly in details like proportions, tones, and scales.

Google describes it as ideal for rapid visual exploration, enabling artists, designers, and other creatives to iterate through numerous ideas in minutes.

Its flexibility and speed make it particularly useful for sparking inspiration, though it may not meet the needs of users seeking perfect, high-fidelity imagery.

However, it also raises questions about the future, especially as AI-generated images are increasingly used to fuel further AI creations. 

Currently, Whisk is available as an experimental project through Google Labs, where users can test the tool and provide feedback.

Google has not specified if or when it will release Whisk to the broader public, but its potential utility is clear. 

Got any thoughts on Google’s new AI image generator? Do you think you would use it? Tell us your thoughts in the comments, down below, or reach out via our Twitter or Facebook.