This clever new A.I. assistant will screen and block robocallers for you

Robocalls are a massive problem, with the average American citizen being on the receiving end of an average of 18 of these automated calls every month. While block lists can put a stop to some of those, they remain a big headache for most people. What if you could turn to yet more robots, these ones fighting on the side of good, to help?

That’s the idea behind a new project carried out by researchers at the Georgia Institute of Technology. They have created a prototype virtual assistant — think Google’s Duplex or Amazon’s Alexa — that could act as intermediary between a caller and the recipient. In essence, it’s a bit like a robo version of a secretary, who answers the phone for you, and then passes it over if the call turns out to be worth taking. In tests involving some 8,000 robocalls, the virtual assistant was able to block every single one. When it was put through its paces with human participants, it was 97.8% effective at recognizing them as non-bots.

“All this happens before the callee’s phone rings,” Sharbani Pandit, a graduate research assistant at Georgia Tech, told Digital Trends. “Only when the [virtual assistant] has deemed a call wanted, it passes the call to the callee. To make this decision, the VA asks the caller who they are trying to reach. The intuition behind this design choice is that legitimate callers should know the name of the person they are trying to reach. The VA also interrupts the caller during the call to make sure if it’s a robocaller. During a conversation, it is natural for humans to stop speaking when they are being interrupted. However, pre-recorded messages from robocallers won’t stop when they are interrupted.”

Fighting bots with bots

The virtual assistant used in the study takes the form of an Android app. However, it won’t be making its way to the Google Play store just yet. Instead, the team behind it is working on improving the virtual assistant to future-proof it against potential robocaller developments like bots acquiring a user’s name in order to bypass such systems. Building a smarter VA that’s able to ask additional question that are easy for a legitimate caller to answer, but tough for a robot (think of an audio CAPTCHA system), could be a step in the right direction. Of course, this might also mean that we would have to answer similar questions when we are the ones making phone calls.

Pandit said there is a chance that good versus evil robocaller systems become a cat-and-mouse battle, as ever smarter robocaller technologies require smarter and smarter detection tools to keep up. However, she is not convinced this will happen — which bodes well for the people fighting on the side of truth, justice, and not being robocalled on a Sunday to be reminded about unused rewards points you didn’t know you had.

“Yes, it is possible for robocallers to become smarter as well, but that comes with a cost,” she said. “Robocallers aim at making cheap mass calls so that they can reach a vast number of targets. As security researchers, our goal is to add entropy to the robocallers to such a point that it would incur more cost which surpasses the robocaller’s benefits.”

A paper describing the proposed system is available to read online.

Editors' Recommendations