Google has announced that it will be allowing users to create images of people using its Gemini artificial intelligence tool. The company had previously pulled the AI image-generation tool earlier in the year due to inaccuracies in historical pictures. The new Imagen 3 generator will be available to Gemini Advanced, Business, and Enterprise users in English in the coming days. The product has undergone technical improvements, evaluation sets, red-teaming exercises, and now has clear product principles in place.
The decision to pause the image generation feature came after embarrassing examples of inaccuracies spread on social media, such as racially diverse depictions of German soldiers and medieval British kings. Imagen 3 does not support photorealistic identifiable individuals, depictions of minors, or excessively gory, violent, or sexual scenes.
Google plans to gradually roll out Imagen 3 to more users and languages in the future, with a focus on listening to feedback from early users to continue improving the product. The company aims to address issues with image generation and ensure that the tool meets user expectations. Subscribe to CNBC on YouTube for updates on this and other related topics.
Photo credit
www.nbcnews.com
