Google Chrome is taking action against websites that ask users repeatedly if they wish to receive notifications by automatically removing prompts based on a user's history.

Google announced in a blog post Thursday that the next version of its Chrome browser will include a machine learning model that will predict a user's browsing patterns. According to Google, the ML model's predictions will be totally on-device.

This will be part of Google's effort to make web browsing more seamless. It is in line with the company's goal of employing AI to improve user experience and create "ambient computing," or the concept of technology becoming so natural that it blends into the background. Some of Google's AI and ambient computing initiatives were announced at its I/O conference.

Google's quieter browsing experience will extend to notifications in general. People still have the ability to alter predictions and, if desired, add notifications.

"On the one hand, page notifications help deliver updates from sites you care about; on the other hand, notification permission prompts can become a nuisance," Google admits in its blog post.

The company's new ML model will now seek for and automatically block prompts that customers are likely to disregard. As an added bonus, all of this occurs on your local PC, so none of your browser data is sent to Google's servers.

Google wants Chrome to adapt to a person's usage in more particular ways by using ML. Google gives an example of a person eating cereal in the morning with a spoon in one hand and a phone in the other. In this case, a person may prefer to search using their voice rather than the on-screen keypad. Chrome might replace the search button with a microphone icon during breakfast. Google stated that once launched, this feature can be manually adjusted.

On the security front, Google said today that it discreetly updated the machine learning engine that drives its Safe Browsing service early this year. This new algorithm detects 2.5 times the number of harmful websites and phishing attacks as the prior model.

Additional machine learning-based features will be added to Chrome. ML can more readily assist people to pick up where they left off in Journeys, a Chrome feature that helps users retrace their web browsing history. In Chrome, machine learning will allow webpages to appear in a user's selected language more frequently.

"Our goal is to build a browser that's genuinely and continuously helpful, and we're excited about the possibilities that ML provides," Google explains.