Bard’s Gemini Pro upgrade continues, gets image generation

February 5, 2024

Google expanded its upgrade of Bard as Gemini Pro now powers the free chatbot in over 40 languages across more than 230 countries. Users can now generate images with Bard in English in most countries around the world.

The gap between the top AI chatbots is closing fast, especially when it comes to free options. Now that Gemini Pro powers Bard, users of OpenAI’s free ChatGPT platform may be tempted to migrate.

Gemini Pro gives Bard more advanced understanding, reasoning, summarizing, and coding abilities. Benchmark tests show Gemini Pro doesn’t beat GPT-3.5 Turbo, but it comes close.

For most users, Bard’s real-time internet access is what makes it more appealing than ChatGPT, which has a knowledge cutoff of September 2019.

Bard’s Double Check feature is also a great way to combat hallucinations. When you click on the “G” icon, Bard checks online whether there is content across the web to substantiate the veracity of the response it generates.

A leaked change log seems to indicate that Bard will get a name change in the next few days as Google leans into the Gemini brand. The leaked change log also makes reference to Gemini Ultra so we may finally see that get released this week.

Leaked Google Gemini change log. Source: X

Bard image generation

Bard now has text-to-image generation functionality for English users in most countries. This free capability is powered by Google’s Imagen 2 model. Google says the model “is designed to balance quality and speed, delivering high-quality, photorealistic outputs.”

The images aren’t as good as Midjourney v6 or DALL-E, but they are still impressive considering the functionality is free. Considering that the free version of ChatGPT is text-only, this is a big plus for Bard.

Image generated by Bard using the prompt: “generate an image of a fashion show in steampunk style digital art, zoom in on their face.” Source: Google
Image generated by Bard using prompt: “generate a collage art, with photorealistic images of oceans and plants with muted colors and 3D shading, that’s mied media” Source: Google

Google says it’s made a efforts “to limit violent, offensive or sexually explicit content” and it also won’t generate images of named people.

Bard’s alignment guardrails do seem a little overcautious as Bard sometimes refuses to generate an image that shouldn’t be problematic.

In line with Google’s AI Principles, the images generated by Bard will have a digital watermark added to the image using SynthID.

It will be interesting to see how OpenAI responds to these new features and the performance boost that Google has given Bard. ChatGPT is a funnel for channeling users to its paid ChatGPT Plus platform.

The multimodal functionality and integration with other Google products like Gmail and Calendar makes Bard a lot more appealing than just a chatbot. I suspect more ChatGPT users will start using Bard.

Join The Future


SUBSCRIBE TODAY

Clear, concise, comprehensive. Get a grip on AI developments with DailyAI

Eugene van der Watt

Eugene comes from an electronic engineering background and loves all things tech. When he takes a break from consuming AI news you'll find him at the snooker table.

×

FREE PDF EXCLUSIVE
Stay Ahead with DailyAI

Sign up for our weekly newsletter and receive exclusive access to DailyAI's Latest eBook: 'Mastering AI Tools: Your 2024 Guide to Enhanced Productivity'.

*By subscribing to our newsletter you accept our Privacy Policy and our Terms and Conditions