Bard, the Google chatbot powered by LaMDA and which works as an alternative to ChatGPT, has been updated with many new features. The company has announced during I/O, their developer conference, a host of new features for AI that promise to significantly improve the experience when making queries. Among them, the possibility of exporting responses to a Docs document, visual search, and the availability in new languages.
The most notable novelty of Bard, yes, is its global availability. Google has confirmed that the chatbot no longer has a waiting list, so as of today, users of more than 180 countries they can test the AI in English.
bard, moreover, also available in japanese and koreanand Google affirms that “soon” it will support another 40 languages in which, we hope, Spanish is present.
Google, on the other hand, is making Bard more visual. That is, the company wants the AI to not only display text responses, but to also illustrate the query with pictures. For example, if a user asks the AI something like “what places to visit in Barcelona”, in addition to text, it will show images of the most notable areas or monuments of the city.
Bard also allows queries through images thanks to the integration of Google Lens. In this way, the user can, for example, ask the chatbot to create a “funny description of a photo of two dogs”. Using Google Lens, Bard will scan the image and detect relevant information, such as race, and generate the description.
Other news coming to Bard
Google, on the other hand, has announced a handful of small but interesting news for Bard. One of them is the possibility of activating a dark mode in the chat, to avoid both visual fatigue and to provide a slightly different aesthetic. The chatbot also now has an “Export” buttonto send the text to a Google document or even to a Gmail email.
Bard also receives improvements aimed at developers and programmers. Like the possibility of exporting lines of code to Colab and Replit or general improvements in source citations.
Finally, Google has confirmed that Bard will have functions supported by third-party services, starting with a able to generate images by short text description thanks to Adobe Firefly, Adobe’s AI model. This particular feature works in a similar way to the one in DALL-E or the one that Microsoft includes in Bing chat. The user will only have to ask Bard to create an image and the AI will show up to four different results.
All these novelties, we reiterate, will be available from today in more than 180 countries. Some of them, yes, could take a few weeks to arrive.