Google's Bard AI can now generate images

Vlad, 01 February 2024

Today Google brought generative AI to Maps, but it also announced some new developments for Bard. The most important of those is that Bard can now use Gemini Pro, Google's latest multimodal AI model, in all the languages and territories where Bard is already available.

Previously, Gemini Pro was brought into Bard only for English, giving it "more advanced understanding, reasoning, summarizing and coding abilities".

Google's Bard AI can now generate images

Bard can also finally generate images. This is available in English "in most countries around the world" for free. The capability is powered by Google's newest Imagen 2 model, which is "designed to balance quality and speed, delivering high-quality, photorealistic outputs".

As you'd expect, using this capability is as simple as typing a description into Bard's text box. You'll then get "custom, wide-ranging visuals to help bring your idea to life". Bard uses SynthID to embed digitally identifiable watermarks into the pixels of generated images, so that there's a clear way to figure out they were AI generated.

Samples of images generated by Bard Samples of images generated by Bard Samples of images generated by Bard
Samples of images generated by Bard

Finally, you can now double-check Bard's output with Google search results in more than 40 languages by clicking on the G icon. Once you click that, Bard will evaluate whether there's content across the web that substantiates its response. You'll get highlighted phrases and will be able to learn more about supporting or contradicting information found by Google search.

Google says the Large Model Systems Organization, a "leading evaluator of language models and chatbots across languages" thinks Bard with Gemini Pro is "one of the most preferred chatbots available", with or without cost, noting that it has made a "stunning leap forward". Blind evaluations by Google's third-party raters also identified Bard with Gemini Pro as "one of the top-performing conversational AIs, compared to leading free and paid alternatives".



Reader comments

It's enough for 91.47% Of Internet User

Well, I Would Make Copyright-Protected doing Non-Family Friendly thing 👍

Why? This is the most useless AI when DALL-E 3 and Midjourney already exists and outperforms Google. How about Google focuses on giving people what they want for a change instead of scrambling to chase trends.

Popular articles


Popular devices

Electric Vehicles