After adding code generation last month, Google is updating Bard with a handful of new capabilities while also previewing several other upcoming features, like Gmail export.
Bard is getting the ability to export generated text to Gmail and Google Docs with formatting, like subject lines, preserved. This is handy as Workspace’s Trusted Tester program is not yet widely available. This is launching today, along with a dark theme.
Also starting Wednesday, Google is dropping the waitlist and making Bard available in three languages for over 180 countries worldwide after previously being just in the US and UK. In addition to English, Bard now supports Korean and Japanese, with a goal of soon supporting the world’s top 40 languages.
To improve the coding and developer experience, Bard is making code citations more precise by having links take you to the specific blocks of code or other quoted text starting next week. Similarly, the “Export” button is soon expanding beyond Google Colab to Replit, starting with Python.
Meanwhile, Google Bard is soon adding image support so that you submit an image and text prompt. For example, you share a picture of two dogs and ask Bard to “write a funny caption about these two.” Image analysis to recognize the subjects (and dog breeds) will be powered by Google Lens.
At a more basic level, results will be able to include images. If you’re asking for tourist spots, each suggestion will be accompanied by a photo.
Google is also working on first and third-party app integrations. The former includes Gmail, Docs, Sheets (show info as a table with the ability to add columns), and Maps (show locations on a map), while an example of the latter would be image generation using Adobe’s Firefly family of generative AI models. You provide a prompt in Bard, and the action is accomplished by another service.
Bard will later move to Google DeepMind’s Gemini model.
FTC: We use income earning auto affiliate links. More.
Comments