Google announced that its Gemini LLM chatbot can now edit images directly within the app. This brings image manipulation features right into your chat conversation with the AI.
This capability, which Google first introduced for developers experimenting in Google AI Studio a while back, comes weeks after Google started allowing anyone to use the basic web version of Gemini without signing in. Now, this hands-on visual editing is rolling out to the standard Gemini app experience for users.
You can now modify both pictures you create with Gemini and images you upload from your phone or computer. This works by using text prompts to tell Gemini what you want to change. For example, you could upload a personal photo and prompt Gemini to generate an image of what you would look like with different hair colors. You can ask it to change the background of an image, replace objects in a picture, or add new elements.
This intuitive, multi-step editing capability means you can get richer, more contextual responses to your prompts with text and images integrated. For example, you could ask Gemini to create a first draft of a bedtime story about dragons and provide images to go along with the story.
Consistent with past practices for AI-generated visuals using Gemini, any images created or edited using this native image generation function will automatically include the invisible SynthID digital watermark. Google also stated that it is currently experimenting with adding a visible watermark to all images generated by Gemini.
Native image editing in the Gemini app is now rolling out. It will gradually reach more users across over 45 languages and most countries in the coming weeks. Google clarified that the feature is not yet available to Google Workspace or Education accounts.