Google is emphasizing additional machine learning-based Chrome features that are device-based and aimed at making a more useful browser, particularly in terms of security.
Since March, Safe Browsings has employed a new machine learning (ML) model that detects 2.5 times as many potentially harmful websites and phishing attempts.

Chrome will soon adopt an on-device prediction approach to automatically silence online notification suggestions that people are unlikely to authorize, improving the browsing experience. Data won’t therefore leave the device, but you can manually approve notifications whenever you want from the address bar.

Thanks to an upgraded identification model, Google is also better at recognizing whether a website needs to be translated to your preferred language. This is preferable to manually calling it from the overflow menu and is reported to have led to tens of millions more successful translations every day.

Chrome ML
Chrome ML

Finally, the Android toolbar shortcut that displays to the right of the Omnibox will soon use ML to dynamically change what appears (New tab, Share, or Voice Search) based on your current browsing context. While Google continues to make this handy little shortcut even more useful, the option to manually set an action will continue to be available.

These upgrades work with Chrome’s already-existing machine-learning features to create Live Captions and classify visual content automatically.
FTC: We employ income-generating auto affiliate connections. MORE ON CHROME. More.
Check out 9to5Google on YouTube for more news:


You may also like