HomeWinBuzzer NewsGoogle Testing Generative AI in Gmail and Docs

Google Testing Generative AI in Gmail and Docs

Google is currently testing new natural language AI for Gmail and Docs, including a new Smart Compose and Smart Canvas model.


is testing some new features that use artificial intelligence (AI) to enhance the user experience of and Docs. 9to5Google reports the company has invited a small group of users to try out these features, which are still in the experimental stage. Both tools add to the AI race Google is current in with , with both companies competing in AI across search, browsers, cloud, programming, and now productivity.

One of the features is an improved version of Smart Compose, which offers suggestions for completing sentences as users type in Gmail. Google's new AI integration uses natural language understanding (NLU) to analyze the context and intent of the email and generate relevant and personalized suggestions. Users can accept the suggestions by pressing the tab key or ignore them by continuing to type.

This is very similar to for Outlook, which made its debut last month. Copilot in Outlook brings automation to emails. Specifically, it helps to manage inboxes and messages by summarizing emails, combining email threads with multiple contacts into useable summarized documents, and respond to emails for users. It will scrape other emails and content you have written to provide the notes you write into professional messages.

It seems that Google's improvements for Smart Compose will offer similar capabilities by tapping into the company's AI.

Smart Canvas

Another feature is a new tool called Smart Canvas, which allows users to create dynamic and interactive documents in Docs. The tool integrates Docs with other Google Workspace products such as Chat, Meet, and Calendar. Users can add rich content such as tables, charts, checklists, and images to their documents. Users can also use voice typing to dictate text, format documents, and insert emojis.

Again, Microsoft 365 Copilot for Word – which also launched last month – provides similar capabilities. Users can task the AI will writing, editing, generating, and summarizing their documents. The model will tap into data from across an organization to compile a document that has value. For example, it can rewrite sections, write articles, add to existing documents, or simply offer a summary of a document.

Google says these features are powered by its advanced models, such as LaMDA and MUM, which can understand and generate natural language across different domains and tasks. The company claims these models can produce fluent and coherent texts that are comparable to human-written ones.

Google has not announced when these features will be rolled out to the public, but says it is looking for feedback from the testers to improve them. The company says it is committed to developing responsible and ethical AI that respects user privacy and security.

Tip of the day: Whether it's for a presentation, song, or YouTube video, at some point in your life you'll need to record audio from your computer. has multiple options to record sound due to its litany of apps. In our tutorial, we show you how to record audio using the built-in Windows 10 Voice Recorder and the freeware audio editor Audacity.

Luke Jones
Luke Jones
Luke has been writing about all things tech for more than five years. He is following Microsoft closely to bring you the latest news about Windows, Office, Azure, Skype, HoloLens and all the rest of their products.

Recent News