How Google uses ML to improve Chrome

Google shared a few of the ways it uses machine learning to improve Chrome Browser on Thursday, including reducing the number of annoying notifications that appear. All the latest updates are all powered by machine learning models on the device so that user data doesn’t have to leave the device.

For a less disruptive web browsing experience, Google uses ML to determine when a user might want to respond to a notification prompt. In the next release of Chrome, the browser will use an on-device model to predict how a person is likely to respond to a consent prompt. If the user is likely to decline it, the browser will mute it. The prediction is based on how the user handled similar consent prompts before.

Google also noted that in March it used ML to improve Safe Browsing in Chrome, which shows alerts when people try to navigate to dangerous sites or download dangerous files. The feature’s new ML model identifies 2.5x more potentially malicious sites and phishing attacks than the previous model.

Google also uses ML to create a dynamic Chrome toolbar, with tools that change in real time in anticipation of your needs. For example, if you are in a situation where you want to use the touchscreen to share a link, the share icon will appear in the toolbar. If you are more likely to use voice search in Chrome, the voice search tool will be highlighted in the toolbar. The toolbar can still be customized manually.