Chrome uses machine learning to detect phishing attacks

Google is integrating a machine learning model into the Chrome browser to detect potential phishing scams.attacks.

According to the developers, the algorithm identifies2.5 times more potentially malicious sites and phishing attacks than the previous model. This ensures safer and more secure web surfing, according to the company.

Google also announced the integration of the AI ​​​​model,disabling site requests to send notifications. According to the company, the algorithm analyzes the user's previous actions and predicts the likelihood of granting such permissions.

A phishing attack warning (left) and an unsolicited notification request from a site that is highly likely to be rejected by the user (right). Data: Google.

Another model is for optimizationtoolbar in Chrome. Depending on the user's actions, the browser will highlight interface elements that are most likely to be useful at the moment.

Changing the interface elements of the toolbar depending on user actions. Data: Google.

Innovations will appear in the next update of Chrome. The company emphasized that machine learning models will process requests directly on the user's device.

Recall that in April, Google Lense text recognition was added to Chrome.

In February, browser developers fixed 27 vulnerabilities, including eight high-risk ones.