Effective spam blocking is yet another thing we can add to the ever-growing list of uses for artificial intelligence.
Via a Google Cloud blog post published Wednesday, February 6, Google announced that it has been using an A.I. platform to further its spam-blocking endeavors with significant results.
The platform, known as TensorFlow, was developed by Google and is “an open-source machine learning (ML) framework.” (ML is a form of artificial intelligence that involves programming machines or programs to carry out tasks relatively independently, by relying on the analysis of data to make their own decisions about how and when to complete such tasks.) While the platform may sound like a new innovation, TensorFlow was actually launched and open-sourced in 2015.
According to Google, TensorFlow is allowing the technology company to block 100 million more spam messages from reaching the inboxes of Gmail users on a daily basis. This is in addition to the 99.9 percent of spam messages Google already claims Gmail blocks.
Google is apparently able to do this because the platform helps Google better detect the following types of harder-to-find spam: Mail from newly created domains, image-based messages, and even messages with hidden embedded content.
Although an extra 100 million spam messages per day sounds like a ridiculously high amount, as The Verge points out, when that number is put in perspective, blocking 100 million spam emails isn’t really very much at all. Especially when the context is, according to Google’s own estimation, that there are 1.5 billion current Gmail users. Spreading out 100 million messages over 1.5 billion users really only gets you roughly “one extra blocked spam email per 10 users.”
But the above caveat doesn’t lessen the overall impact of TensorFlow on blocking spam for Gmail users. Being able to block an additional 100 million messages is still an important achievement, as it suggests that the ML behind TensorFlow helped enhanced Gmail’s spam-blocking functionality as it worked in tandem with Gmail’s rule-based filters.
And TensorFlow’s spam-blocking ML might just continue to improve as time goes on; Google also mentioned in that blog post that the platform’s ML is intended to also help Gmail customize its spam protections for each individual user’s needs.