Google earlier this year in February introduced generative Artificial Intelligence Gemini AI bot with text-to-image generating capability.
But, within weeks after the launch, google faced flak for inaccurate representation of human races. For instance, when users asked Gemini to produce Viking soldiers' photos, it churned out irrelevant images of people of mixed ethnicities in Viking gear.
After facing flak, Google pulled down the AI image generator feature from the Gemini app.
Now, after five months of delay, the company is finally bringing it back with a new improved Imagen 3 model.
"We're bringing our latest image generation model, Imagen 3 to Gemini App and expanding its availability for users in all languages," Google said.
The image-generating capability will be enabled with an update to Gemini app.
The Imagen 3 is said to be made of an advanced AI language model and will be able to enable the Gemini app to bring more accurate and realistic images and require fewer words.
Users can also ask Gemini to create images in various styles such as photorealistic landscapes, textured oil paintings or whimsical claymation scenes.
However, Gemini AI will not be able to generate images of people just yet. Google says it will be made available a little later to users subscribed to Gemini Advanced, Business, and Enterprise plans, starting in English.
Another important thing to note is that the Imagen 3-powered Gemini AI comes with built-in safeguards that block users from creating any hateful images. And, it also comes integrated with SynthID, a watermarking tech.
SynthID features several nuanced technologies to ensure the artificial images are aesthetically good and not marred with any noticeable marks that compromise the visual beauty. But, anybody who comes across the artificially generated images with SynthID, will be able to differentiate them through tools made by Google.
Get the latest news on new launches, gadget reviews, apps, cybersecurity, and more on personal technology only on DH Tech