Google is boosting AI services across the board, as AI image generation arrives in Google Bard. This is powered by new Imagen 2 technology that Google claims is more lifelike than ever. Google is also releasing a dedicated AI art tool, known as ImageFX, to go along with its existing MusicFX and TextFX tools. Google has integrated AI into many services, and it’s made its Gemini Pro upgrade widely available. This includes all supported languages, countries, and territories. It’s one of three related models that Google is releasing, including a smaller Gemini Nano and a more powerful version known as Ultra. A HuggingFace “leaderboard” of chatbots rates Bard’s new Gemini Pro model as the second-best AI chatbot.
Google will now allow users to ask Bard for an image and it will draw it for free. ImageFX is Google’s latest generative art tool that works separately, tapping Imagen 2’s artistic capabilities. Google has caught up with Bing Image Creator and more general AI art tools like Midjourney in standalone AI art apps. Now, ImageFX will allow you to create your own standalone AI art and quickly modify it using what Google calls “expressive chips.” All images generated using ImageFX include metadata that identify the images as AI-generated to give people more information whenever they encounter Google’s AI-generated images.
The Imagen 2 technology delivers Google’s highest-quality images yet, with improvements in difficult areas like rendering realistic hands and human faces. Google has undergone “extensive adversarial testing and red teaming to identify and mitigate potential harmful and problematic content,” according to a statement by Eli Collins, vice president of Google’s DeepMind.