Gmail now thwarts 99.9 per cent of scam emails
Google claims that its Gmail service now blocks 99.9 per cent of all phishing attempts it detects. It says that this is a “huge” achievement because 50 to 70 per cent of emails that Gmail users receive are spam.
In a blog post announcing the milestone ( www.snipca. com/24556), the company says that Gmail’s detection rate has been boosted by machine learning. This is artificial intelligence that allows software to improve by itself, without any human involvement.
Andy Wen, senior product manager at Google’s counterabuse technology team, said: “As we find new patterns, our models adapt more quickly than manual systems ever could, and get better with time”.
He added that Gmail is now better at spotting “polymorphic malware”, which are threats that constantly change to bypass antivirus software.
Google is also strengthening its security for companies that use Gmail. Now, when an employee tries to send an email to someone outside the company, or not in their contact list, Gmail will ask them whether they really want to send it (see screenshot). Google says this will stop employees accidentally sending sensitive data to the wrong person.
Wen acknowledged that such warnings become annoying if they are shown too often, but said that Gmail uses “contextual intelligence” to determine whether the recipient is an existing contact or someone you send emails to regularly, and therefore probably safe. This will avoid “displaying warnings unnecessarily” he claimed.