Google Launches Whisk AI to Remix Images with Visual Prompts

Google has introduced Whisk, a cutting-edge AI image generator that enables users to create custom images by using existing images as prompts. Unlike traditional AI tools that rely solely on detailed text descriptions, Whisk lets users plug in multiple images to define the subject, scene, and style of their desired output, offering a more intuitive and visual way to explore AI-generated art.

How Whisk Works

Whisk simplifies the image generation process by blending user-provided images with optional text prompts:

• Image-Based Prompts: Users can upload images to guide the subject, scene, and style of the output. Multiple images can be combined for each category.

• Text Enhancements: While not required, users can enter additional details via text to refine the generated image.

• Dice Icon for Suggestions: If users don’t have images, Whisk can randomly generate visual ideas using AI-generated images.

Once the user provides input, Whisk generates:

• AI-created images based on the provided prompts.

• A text prompt that explains how the image was generated, which users can edit for further refinements.

• Users can then favorite, download, or tweak the results by editing prompts or providing new details.

A Tool for Visual Exploration

In a blog post, Google acknowledged potential limitations: Whisk may miss the mark, which is why it lets you edit the underlying prompts. Google emphasizes that Whisk is designed for rapid visual exploration rather than detailed, pixel-perfect edits.

Powered by Imagen 3

Whisk runs on Google’s latest Imagen 3 image generation model, announced alongside the new tool. Imagen 3 builds on the success of previous iterations with improved accuracy and versatility.

Introducing Veo 2: Google’s Advanced Video Generator

In addition to Whisk, Google unveiled Veo 2, an upgraded video generation model with enhanced cinematic understanding. Highlights include:

• Fewer Visual Errors: Veo 2 reduces common AI generation mistakes, such as hallucinating extra fingers.

• Cinematic Awareness: The model demonstrates an understanding of cinematographic principles, making it ideal for video creation.

• Integration with VideoFX: Initially available through Google’s VideoFX (accessible via the Google Labs waitlist). Veo 2 will expand to YouTube Shorts and other Google products in 2025.

Why Whisk Matters

Google’s Whisk represents a shift in AI-generated content creation by putting visual prompts front and center. This approach makes image generation more accessible to casual users and creative professionals alike:

• Ease of Use: By relying on visual prompts, Whisk lowers the barrier to entry for users who may struggle with crafting detailed text descriptions.

• Iterative Creativity: Users can explore, tweak, and refine their ideas quickly, making it a valuable tool for brainstorming and experimentation.

Combined with Google’s ongoing advancements in video generation through Veo 2, the company is setting a new standard for how AI can empower creativity in both static and dynamic formats.

Looking Ahead

As Whisk rolls out, it’s likely to attract a diverse range of users, from hobbyists experimenting with AI art to professionals seeking a faster way to prototype visual ideas. The integration of Imagen 3 signals Google’s commitment to improving AI tools for creative exploration.

Meanwhile, the upcoming expansion of Veo 2 to platforms like YouTube Shorts positions Google as a leader in the rapidly evolving field of generative video. Together, these tools showcase how AI is reshaping the creative process across industries.

Editor’s Note: This article was created by Alicia Shapiro, CMO of AiNews.com, with writing, image, and idea-generation support from ChatGPT, an AI assistant. However, the final perspective and editorial choices are solely Alicia Shapiro’s. Special thanks to ChatGPT for assistance with research and editorial support in crafting this article.