Soon, Google to add image editing features to Gemini app: What is new

The Gemini app will support AI-powered image editing with natural language prompts, style transformations and watermarking for authenticity

Google is introducing native artificial intelligence-powered image editing features to its Gemini app. The new tools will allow users to modify both manually uploaded and AI-generated images using simple natural language prompts. Edits can include background changes, object replacements and the addition of new elements. Users can continue making changes until they are satisfied with the final result, said Google.

 

The company stated that the context of previous edits will be preserved throughout the editing session. Style modifications will also be supported — such as placing a hat on a dog, changing a grassy field to a beach, or altering hair colour in a portrait.  ALSO READ: OpenAI rolls back GPT-4o update as AI goes 'sycophantic': What it means

 

 

Once available, this functionality will be accessible through an inline prompt within the app. It will be supported across all currently available Gemini models. Google said that this "intuitive, multi-step editing" will offer more contextual and seamless interactions by blending text and visual inputs. For example, users drafting a bedtime story in Gemini can request matching illustrations, or generate image-based, step-by-step instructional guides.

Watermarking for safety and authenticity

 

To address concerns around authenticity and misuse, Google stated that all images created or edited using Gemini's native tools will carry a SynthID digital watermark — invisible to the human eye. Additionally, Google is testing an optional visible watermark to further enhance transparency. 

Rollout timeline

 

The rollout of Gemini’s image editing capabilities is gradually rolling out and is expected to expand in the coming weeks to a broader user base across more than 45 languages and most countries.