Does Google’s Artificial Intelligence Really Capture 99.9 Percent of Spam?

A report released this month by Symantec, an American technology company, claims that spam has, for the first time since 2003, accounted for less than 50 percent of all emails.

This may not be a big deal to most email users, many of whom aren’t old enough to recall the days when spam nearly brought the email world to its knees. However, ask anyone who remembers the days of old when spam accounted for most of the emails that they received and they’ll let you know just how remarkable this is.

There has been a long running battle between the tech giants – mostly Google and Microsoft with Gmail and Hotmail respectively – over whose artificial intelligence could most effectively eliminate unwanted spam from emails users’ inboxes, with neither a convincingly clear winner. Until now.

Google now claims that its spam filters can eliminate as much as 99.9 percent of all spam – 0.01 of all spam emails therefore make their way past the filters – and that its false positive rate has dipped to an impressive 0.05 percent.

This is fantastic news for the 900 million people who Google claims uses its email platform, Gmail. Microsoft has not yet released its latest Hotmail stats, though like Google, it also uses ‘neural networking’ for some of the products it has released.

Gmail uses several effective tech tools that tune its spam filters to users’ particular tastes, says John Rae-Grant, one of Google’s senior production managers. “There’s all this grey area. One person’s spam might be another person’s coupon.”

This means that Gmail users should think twice before designating emails as spam. “We track and try to approximate – based on what you have previously paid attention to and reported as spam – what you want in front of you and don’t want it front of you.”