Google said the ML model’s predictions will be entirely on-machine.
This will be part of Google’s goal to make web browsing a more uninterrupted experience.
Some of Google’s AI and ambient computing goals were outlined at itsI/O developer’s summitlast month.
People still have control to override predictions and can implement notifications if they’d like to.
By using ML, Google wants Chrome to adapt to a person’s usage in more specific ways.
In this instance, a person might rather search with voice than with the on-screen keyboard.
During breakfast, Chrome could automatically replace the search button to a microphone icon.
Google said this feature can be customized manually once it launches.
Chrome will also get additional ML-based features.
ML in Chrome will also allow websites to automatically come up in a person’s preferred language more often.
Google declined to comment further on the story.
