Google has added a bunch of new security features to Gmail including early phishing detection that uses machine learning, click-time warnings for malicious links, unintended external reply warnings, and built-in defenses against new threats.

According to Google, machine learning already helps Gmail block spam as well as phishing emails with an accuracy of over 99.9 percent. The online search giant wants to further improve the spam protection accuracy with early phishing detection, which basically delays certain messages (less than 0.05 percent on average) and performs what Google describes as a rigorous phishing analysis. It works together with the Google Safe Browsing tech that looks for and flags URLs that are considered phishy.

Google has also added unintended external reply warnings to Gmail. Whenever you send an email to someone outside of your company domain, a warning will pop up asking you if you really want to send it. Thankfully, if you’re sending the email to an existing contact or someone who you’re communicating with on a regular basis, the warning won’t be displayed. That’s an important factor, as it could get quite annoying if the warning would appear every single time you wanted to send an email, right?

Google also claims that the new built-in defenses against ransomware and polymorphic malware now help Gmail block a lot more emails that have the potential to harm users in some way. Threats are classified by combining a bunch of spam, malware, and ransomware signals with attachment heuristics (emails that could be threats based on signals) and sender signatures (already marked malware).

Read comments