Google’s Gemini Chatbot app now allows you to change both images and images generated by AI uploaded from your phone or computer.
Native image editing at Gemini will unfold slowly today and expand to more people in more than 45 languages and most countries in the coming weeks.
The release follows the AI image editing model piloted on the AI Studio platform in March. Like ChATGPT’s recently upgraded image editing tool, Gemini’s NewFangled Native Image Editor can, in theory, achieve better results than the standalone AI image generator.
Gemini provides a “multi-step” editing flow where Google describes “further, more contextual” responses to each prompt, combining text and images. Within the Gemini platform flow, you can change the background of an image, exchange objects, add elements, and more.
“For example, you can upload personal photos and quickly generate images of what you see in your hair color to Gemini,” explains Google in a blog post. “Or you can ask Gemini to create the first draft of a bedtime story about the dragon and provide images to suit the story.”
If this sounds like a deepfake risk, well, that’s reasonable. According to Google, images created or edited with Gemini’s native image generation contain invisible watermarks to alleviate fear. The company is also “experimenting” using visible watermarks on all Gemini-generated images.