Google Gemini app now lets you edit photos with AI prompts: How the tool works

Google has announced the global expansion of its artificial intelligence (AI)-powered native image editing feature to the Gemini app, marking a significant upgrade to the AI chatbot’s capabilities. The new tool allows users to make detailed edits to images directly through conversational prompts, without needing to switch to a separate editing platform.

Originally unveiled in March within Google AI Studio, the feature is powered by the company’s Gemini 2.0 Flash AI model. Now, Google is gradually rolling out the tool to users worldwide, supporting over 45 languages. The phased rollout means it may take several days before all users gain access.

How the edit tool works

With this update, users can upload a photograph or generate an AI image within the app and request specific changes using natural language. The AI can respond to instructions such as adding, removing or replacing objects, enhancing backgrounds, or altering particular elements of a photo. For example, a user might upload an image of a table and ask the chatbot to place a bouquet of flowers on top, or submit a selfie and request a visualisation of what they would look like with a different hair colour.

Google has highlighted that the AI is designed to handle even complex editing tasks while maintaining ease of use. However, the company acknowledged the risks of potential misuse, such as the creation of deepfakes. To address this, Google stated that the system is trained to reject requests involving offensive or harmful content.

In addition, all images edited or created using the feature will include an invisible SynthID digital watermark to indicate their AI origins. Google is also testing the addition of visible watermarks on AI-generated images to further ensure transparency.

The rollout reflects Google’s broader strategy to integrate AI seamlessly into user-friendly applications, empowering people to access advanced tools without technical expertise.