Google is beginning a phased rollout of native AI image editing tools within its Gemini app, expanding the app’s capabilities beyond text-based interactions. The new feature allows users to edit both manually uploaded images and those generated within Gemini, using natural language prompts to guide changes.
This update enables users to make a wide range of visual modifications, such as altering backgrounds, inserting or replacing objects, and applying stylistic transformations—all through conversational instructions. Context from previous prompts is preserved, supporting iterative edits without requiring users to start over. For instance, a user could add accessories to a pet in one step, then follow up by changing the setting from a grassy field to a beach scene, with Gemini adapting accordingly.
Styling and personalization are also supported. Users can upload personal photos and ask Gemini to show variations—like different hair colors or environments—directly within the app. This multi-step editing approach is designed to feel intuitive, with the added benefit of integrating text and visual outputs into a seamless response. A common use case might include creating illustrated bedtime stories, where Gemini generates both narrative text and accompanying images based on the user’s input.
The rollout includes support for all available Gemini models, and users will be able to identify when image editing is enabled through an inline prompt within the app. While the core tools are powered by Google’s existing AI infrastructure, the company is emphasizing responsible deployment. All images created or modified using the native editor will include SynthID—Google’s invisible watermarking system—ensuring provenance and discouraging misuse. The company is also testing visible watermarks, such as a subtle “ai” label in the corner of generated visuals.
This move follows internal testing in Google’s AI Studio and builds on recent improvements in rendering capabilities, particularly with the Gemini 2.0 Flash model, which has shown better performance in handling image generation involving text and more detailed reasoning.
Availability is rolling out gradually, with support planned for users across more than 45 languages and most countries in the coming weeks. As the update progresses, Gemini is positioned to become not just a conversational assistant, but a more robust creative tool blending text, image, and AI-guided personalization.
