In this photo taken on February 8, 2024 in Brussels, Belgium, the Deepmind logo is displayed on a smartphone with the Google Gemini logo in the background.
Jonathan Ra | Noor Photos | Getty Images
Google The company said it will soon allow users to create images of people using the Gemini AI tool, after launching it in February.
in a Blog article On Wednesday, Gemini senior product director Dave Citron wrote that Gemini power users, business users, and enterprise users will be able to get early access to Google’s new Imagen 3 generator (English version) in the coming days.
““We are constantly working on technical improvements to the product, as well as improved assessment sets, red team exercises, and clear product principles,” Citron wrote. Red teaming refers to the practices companies use to test products for vulnerabilities.
Google suspended its image generation feature earlier this year after it said the historical images it provided were “inaccurate.” Embarrassing examples are circulating on social media.
In 1943, a user asked the tool to generate an image of a German soldier, and it created a racially diverse group of soldiers wearing German military uniforms. Another inquiry into historical depictions of medieval English kings also yielded a racially diverse set of images that included a female ruler.
Citron said Imagen 3 does not support realistic depictions of identifiable individuals, minors, or scenes of excessive gore, violence, or sex.
“Of course, as with any generative AI tool, not every image created by Gemini will be perfect, but we will continue to listen to feedback from early users as we continue to improve,” Citron wrote. “We will roll out gradually. This feature aims to bring it to more users and more languages as soon as possible.”
About The Author
