There are certainly ways to use tech to augment customer service, like Twitter’s auto-replies for public-facing business accounts. But if you’ve called a large company’s helpline and gotten a representative that was atypically, unusually helpful, latch on to that uncertainty: Your hotline helper might have been hand-picked for you by an AI system. The startup Afiniti International Holdings has installed just such a setup in over 150 call centers to better match callers with associates who have successfully helped similar folks.
To build a profile of the caller, Afiniti’s tech pulls up their purchase and contact history and business/credit profile. Since this is 2017, it also automatically crawls the internet to find their public Facebook, Twitter and LinkedIn pages. Finally, it fills out the profile by examining census archives of the caller’s area. If the associate on the other end of the phone is in sales, they’ll be matched with callers similar to those they’ve more successfully sold to in the past.
The associates can’t see callers’ personal data, Afiniti maintains, and they don’t know why they’re matched to whom: it’s all done in the background by the company’s AI system. But simply having scores and metrics invisible to callers is worrying some privacy advocates.
“There’s a process of discrimination going on,” University of Pennsylvania professor Joseph Turow, who studies digital marketing, told The Wall Street Journal. “Companies are bringing data together that we have no knowledge about, and it may discriminate against in a prejudicial sense or a positive sense, depending on who we are.”
Afiniti’s careful to note where it claims to pull its data from, according to The Wall Street Journal: From up to 100 databases that are legitimately available for purchase, like credit firm Experian and data clearinghouse Acxiom Corp, and only social media data from callers’ profiles that had been made public. But if you’re worried about making a simple customer service call and getting slotted a salesperson expressly skilled at upselling people like you, it might be too late: Your data’s already out there.
Source: The Wall Street Journal