Facebook Is Working Hard to Remove Fake Accounts and Hate Speech

Social media giant Facebook (FB) released its third Community Standards Enforcement Report on May 23. The company disclosed that it had removed a whopping 2.2 billion fake accounts during the first quarter of 2019.

This number is huge considering that Facebook had 2.38 billion monthly active users globally at the end of the first quarter. These fake accounts aren’t counted as a part of the company’s monthly active user base. However, the company did disclose that 5% of its monthly active accounts are fake.

The number is also much higher than the 1.2 billion fake accounts the company disabled during the fourth quarter of 2018. Facebook said that it took down 99.8% of these accounts using automated tools before users reported them.

The company seems to be improving at weeding out hate speech, which has been particularly challenging for it. Facebook has also pointed out that it’s improving at weeding out content that infringes its community standards. The company said that 65% of the posts removed for hate speech were detected before any user could flag them. In the same quarter last year, that number stood at 38%.

Apart from the AI that helps Facebook proactively remove fake accounts and hate speech, among other things, the company has also spent a lot on human moderators, as AI often can’t understand context well.