As Google got targeted by phishing attacks, its war against such malicious attampts also rages on. Recently, the company has implemented a series of improvements to Gmail that would assist users in steering clear of any suspected phishing attempt.
The first of the new improvements is the implementation of a new machine learning model that brings with it early phishing detection. According to Google, this system would “selectively delay messages” in order to perform “rigorous ph’shing analysis”.
On top of that, the machine learning model has also been integrated with the pre-existing Google Safe Browsing technology. The combination of a variety of distinct technologies means that Google can now generate new URL click-time warnings for phishing and malware links.
Where the machine learning model operates silently in the background, the next improvement that was made to Gmail is far more visible. From this point forward, Google will display an unintended external reply warning that appears whenever the user attempts to send an email to an unknown address.
By using the company’s own contextual intelligence, Gmail can now identify if the mail address in question is a familiar one or a suspicious one.
Lastly, Google has implemented a series of defenses that would protect the user from new threats such as ransomware and polymorphic malware by utilising a combination of spam, malware and ransomware signals with attachments heuristics and sender signatures.