Google has introduced generative AI capabilities to its Maps platform recently, while also unveiling notable enhancements for Bard. One significant update is that Bard can now leverage Gemini Pro, Google’s newly announced multimodal AI model, across all the languages and regions where Bard is currently accessible.
In the past, the integration of Gemini Pro was limited to the English language for Bard, offering enhanced capabilities in areas such as comprehension, reasoning, summarization, and coding.
Additionally, Bard has gained the ability to generate images. This feature is currently available in English and accessible in “most countries around the world,” and it comes at no cost to users. The image generation capability is powered by Google’s latest Imagen 2 model, which has been specifically designed to strike a balance between quality and speed, delivering high-quality and photorealistic results.
The video below shows how it works.
Just like most other AI image generators, using Bard’s version is as simple as typing a few words as a description. In response, Bard generates custom and diverse visuals to help illustrate the idea. To ensure transparency, Bard employs SynthID to embed digitally identifiable watermarks within the pixels of the generated images, providing a clear indication that they are AI-generated.
Lastly, users can now cross-reference Bard’s responses with Google search results in over 40 languages by clicking on the G icon. Upon clicking, Bard will analyze whether there is web content that validates its response. Users will receive highlighted phrases and have the opportunity to explore further details about corroborating or conflicting information uncovered through Google search.