WhatsApp Bans 2 Million Fake or Abusive Accounts Each Month
Facebook-owned WhatsApp shed light on Wednesday on how it works to ban fake and abusive accounts on its wildly popular messaging service.
Ahead of a major election in India, WhatsApp held a press conference in New Delhi on Wednesday to discuss the company’s efforts at employing machine learning to find fake and abusive accounts and remove them from its service. According to company spokespeople, whose comments were earlier reported on by Venturebeat, WhatsApp removes two million accounts from its service each month. WhatsApp software engineer Matt Jones said that the company can ban 20% of those accounts when they register.
In order to ban those accounts, WhatsApp uses a combination of human intervention and machine learning. Twenty-five percent of the banned accounts are removed by humans, while the remaining 75% are scrubbed out by algorithms that seek out malicious activity, according to the company.
“Some may want to distribute click-bait links designed to capture personal information, while others want to promote an idea,” WhatsApp said during the press conference, according to Venturebeat. “Regardless of the intent, automated and bulk messaging violates our terms of service, and one of our priorities is to prevent and stop this kind of abuse.”
WhatsApp is one of the most popular messaging services in the world with more than 1.5 billion users. Its popularity, like other prominent messaging apps and social networks, makes it a destination for bad actors to spam people, spread misinformation, and otherwise wreak havoc. And with worldwide fears of election interference and the spread of misinformation on social networks, the Facebook-owned WhatsApp is working to allay some of those concerns.
The issue is especially acute in India, where the country is heading to a major general election in April. The Indian government has signaled fears of election interference and has pressured tech companies to step up. They’ll ultimately be put to the test in two short months to see if their efforts, including human intervention and machine learning, can withstand an onslaught by bad actors.
Facebook did not immediately respond to a Fortune request for comment on the WhatsApp comments.