Gmail Gets Machine Learning Integration for Phishing Attack Protection

It is no secret that phishing attacks are on the rise and after working towards making its Gmail client safer across platforms for quite some time, Google has now taken another significant step in the same direction. Google has announced that the company has now integrated machine learning for early detection of phishing attacks through Gmail. Apart from machine learning, the search giant claims to have also added click-time warnings for malicious links, unintended external reply warnings, and built-in defences against new threats.

The latest additions from the search giant seem to be focused on businesses and Google says that new security features will help them stay ahead of potential threats. Talking about the machine learning, Google says, "Machine learning helps Gmail block sneaky spam and phishing messages from showing up in your inbox with over 99.9 percent accuracy." The search giant has termed this as a crucial factor as according to its own estimates, 50-70 percent of messages that Gmail receives are spam.

Google says that a dedicated machine learning model, meant for early phishing detection, selectively delays messages to perform rigorous phishing analysis and further protect user data from compromise. If you are concerned about the delay, there's no need as the search giant says it happens for less than 0.05 percent of all messages on average.

Notably, the California-based company's detection modules are integrated with Google Safe Browsing technology in order to find and flag suspicious URLs.

"These new models combine a variety of techniques such as reputation and similarity analysis on URLs, allowing us to generate new URL click-time warnings for phishing and malware links. As we find new patterns, our models adapt more quickly than manual systems ever could, and get better with time," Google said in its blog post.

In order to increase the security awareness at enterprise level, Google has added a quick warning for employees that shows up to confirm if they intended to send an email when they are sending it to someone outside their company's domain. The company says that contextual intelligence allows Gmail to understand if the recipient is an existing contact or someone you talk to regularly, to avoid unnecessary warnings.

"With new built-in defences against ransomware and polymorphic malware, Gmail now blocks millions of additional emails that can harm users," the company said. Google says that it classifies new Gmail threats by combining "thousands of spam, malware and ransomware signals with attachment heuristics (emails that could be threats based on signals) and sender signatures (already marked malware)."