trending Market Intelligence /marketintelligence/en/news-insights/trending/xnf1sfsqcxldhhjkvjfioq2 content esgSubNav
In This List

Facebook releases enforcement report on content policing, government requests

Podcast

Next in Tech | Episode 50: InfoSec spending up, again…

Blog

Broadcast deal market recap 2021

Podcast

Next in Tech | Episode 49: Carbon reduction in cloud

Blog

Price wars in India: Disney+ Hotstar vs. Amazon Prime Video vs. Netflix


Facebook releases enforcement report on content policing, government requests

Facebook Inc. has released a community standards enforcement report on its content policing from October 2017 to March 2018.

The report, which covers graphic violence, adult nudity and sexual activity, terrorist propaganda, hate speech, spam and fake accounts, shows how much content Facebook users saw that violates company standards, how much was removed and how much was detected by its own technology, ahead of user reports.

Facebook's policing measures mostly centered on spam and fake accounts, with 837 million pieces of spam removed in the first quarter, Guy Rosen, vice president of product management, said May 15. Nearly all of the spam was flagged before users could report it, he added.

The social network also disabled 583 million fake accounts, in addition to the millions of fake account creation attempts that it prevents daily. The company estimates that 3% to 4% of Facebook accounts active in the first quarter were fake.

In the first quarter 21 million pieces of content involving adult nudity and sexual activity were also removed, of which 96% was found before being reported. Facebook estimates that seven to nine pieces of content out of every 10,000 breaches adult nudity and pornography standards.

Facebook also removed or placed warning labels on about 3.5 million pieces of violent content in the same period, 86% of which was identified proactively.

Facebook said its technology remains unable to properly detect content containing hate speech and its review teams still need to manually check content. Facebook removed 2.5 million pieces of hate speech in the first quarter, with 38% flagged by technology.

Data requests

Vice President and Deputy General Counsel Chris Sonderby said the company received 82,341 government requests globally for user data between October 2017 and March 2018, up from 78,890 in the first half of 2017.

In the U.S. the number of government requests remained almost the same at 32,742, of which 62% included a nondisclosure order preventing Facebook from notifying the user, an increase of 57% from the first half of 2017.

Sonderby said that during the second half of 2017 the company restricted 14,294 pieces of content based on local legislation, down from 28,036, which was higher due to the content restrictions Facebook made in Mexico linked to a school shooting.

Facebook experienced 46 service disruptions in 12 countries in the second half of 2017, in comparison to 52 disruptions in nine countries in the first half, Sonderby said.

From October 2017 to March 2018, Facebook and Instagram Inc. removed 2,776,665 pieces of content that were based on 373,934 copyright reports, 222,226 pieces of content based on 61,172 trademark reports and 459,176 pieces of content based on 28,680 counterfeit reports.

The report follows news that Facebook founder and CEO Mark Zuckerberg declined U.K. lawmakers' request to be questioned shortly after being threatened with a formal summons earlier this month. The committee's requests follow the fallout from Facebook's links to the Cambridge Analytica LLC scandal, wherein millions of users found their personal data used without their consent.

The company also suspended about 200 apps for potential data misuse pending the results of a full investigation.