Google is again allowing users to generate images of people through its Gemini AI chatbot after it pulled the feature earlier this year following reports of racially diverse Nazis and other historically inaccurate images. In an announcement, Google said it would be rolling out an early access version of the feature to English Gemini Advanced, Business, and Enterprise users “in the coming days.”
The upgrade is powered by Imagen 3, the latest version of Google's AI text-to-image generation tool. Google quietly released Imagen 3 through its AI Test Kitchen earlier this month, but it's now rolling out to Gemini in all languages. The upgraded tool can generate everything from photorealistic landscapes to textured oil paintings with just a few words of description.
Imagen 3 can create a wide range of images. Image: Imagen 3
This is not to be confused with Google's new Remagine feature, which lets you incorporate AI elements into photos taken with the company's new Pixel 9 smartphone lineup.
Additionally, Gemini doesn't allow the creation of graphic images of celebrities, content involving minors, or cruel and violent sexual scenes. “Of course, not all images created by Gemini are perfect, but we'll continue to improve them as we listen to feedback from Gemini Advanced early access users,” Citron adds. Google plans to expand the ability to create AI-generated people to more users and languages soon.
