Google announced via a blog post on Thursday that it has updated the Bard AI chatbot with image-generation capabilities. The tool is powered by the updated Imagen 2 model that’s trained to deliver photorealistic images across a range of styles based on your input.
You can start making your image creations directly from bard.google.com and here, you can enter a prompt that goes like “Create an image of a dog riding a surfboard”. Bard will then process your input and generate up to 2 images but you can request the chatbot to “Generate more”.
Additionally, you can ask Bard to generate text with images with a prompt similar to “Write a social media post and generate a mouthwatering image that I can use for a buffalo wing festival”. You can download the generated images as JPG files that currently default to 1536×1536 pixels.
The company shares that Bard uses Google’s SynthID tool to distinguish AI creations from original human artwork by adding identifiable watermarks to the generated images. Google has also trained its AI models to prevent producing violent, offensive, or sexually explicit content and avoid generating photos of celebrities and popular people.
Bard’s AI image generator is now available to everyone for free across 230+ countries but currently, images can only be created with English prompts. Google’s support page, however, indicates that image generation is unavailable in the European Economic Area (EEA), Switzerland, and the UK.
