Although notifications on the web are useful for a variety of applications, they can also be misused for phishing, malware or fake messages that imitate system notifications for the purpose of generating user interactions.
In Chrome 86, we’ve expanded on previous efforts [
Mobile UI for quiet notifications on abusive websites. The new UI discourages users from allowing notifications from these websites.
This UI exactly matches the UI that was previously announced for Chrome 84. The only difference is in Chrome 86 we will begin blocking notification permission requests when sites have a pattern of sending abusive notification content.
Abusive notification prompts are one of the top user complaints we receive about Chrome. Our goal with these changes is to improve the experience for Chrome users and to reduce the incentive for abusive sites to misuse the web notifications feature.
Google’s automated web crawling service will occasionally subscribe to website push notifications if the push permission is requested. Notifications that are sent to the automated Chrome instances, using Safe Browsing technology, will be evaluated for abusive content, and sites sending abusive notifications will be flagged for enforcement if the issue is unresolved.
When a site is found to be in “Failing” status for any type of notification abuse, Search Console will send an email to registered site owners and users in the site's Search Console at least 30 calendar days prior to the start of enforcement. During the 30 day grace period websites can address the issue and request another review.
We recommend concerned site owners and developers review the guide for how to fix abusive notifications and request another review of your website.
Prior to the release of Chrome’s abusive notifications protections, many users have already unintentionally allowed notifications from websites engaging in abusive activity. In an upcoming release, Chrome will revert the notification permission status from “granted” to “default” for abusive origins, preventing further notifications unless the user returns to the abusive origin and re-enables notifications.
We’ll be listening for feedback from users and developers about the effectiveness of current enforcements and may make further changes based on that feedback.