In the first three months of the year 2018, Facebook claims to have taken down 837 million pieces of spam and disabled 583 million fake accounts, and most of them before being reported. The figures form part of Facebook’s first transparency report that it released on Tuesday.
In a Facebook post, CEO Mark Zuckerberg says the report is on the effectiveness of the social network giant’s enforcement mechanism.
According to the report, Facebook received 82,341 requests for account data in the second half of 2017 – a 4 per cent global increase compared to the first half. At 17,262, requests from India for user or account data were the highest after the US, which made 53,625 requests.
At a distant third was the United Kingdom with 9,719 requests, followed by Germany with 6,493 and Brazil with 6,108 requests.
Facebook had last month published its internal guidelines to review posts for hate speech, violence, nudity, terrorism, and other content that don’t conform to its ‘community standards’.
Zuckerberg says the transparency report includes the same metrics Facebook use internally to measure its progress. The report will be updated twice a year.
“Thanks to AI tools we’ve built, almost all of the spam was removed before anyone reported it, and most of the fake accounts were removed within minutes of being registered,” Zuckerberg says in his post, adding that Facebook is working on getting its AI better to effectively remove more linguistically nuanced issues like hate speech in different languages.
The report gives detailed information of the government data requests Facebook receives from different countries. Government data requests essentially mean requests for data about people who use Facebook as part of official investigations. According to Facebook, the vast majority of these requests relate to criminal cases, such as robberies or kidnappings. And in many of these cases, the government requests seek basic subscriber information, such as name, registration date and length of service. Other requests may also seek IP address logs or account content.
“We accept government requests to preserve account information pending receipt of formal legal process. When we receive a preservation request, we will preserve a temporary snapshot of the relevant account information but will not disclose any of the preserved records unless and until we receive formal and valid legal process. We have reported this information since 2016,” says the report.
Besides Facebook, the transparency report also includes information about requests related to Instagram, Messenger, Oculus and WhatsApp.
From the Indian government, there were a total of 12,171 data requests involving 17,262 users in the second half of 2017. In 53 per cent of the requests, Facebook says, it produced some data. There were also a total of 11,874 legal process requests involving 16,740 accounts.
“We restricted access to content in India in response to legal requests from law enforcement agencies and the India Computer Emergency Response Team within the Ministry of Electronics and Information Technology. The majority of the content restricted was alleged to violate Indian laws relating to the defamation of religion, hate speech, and defamation of the state,” the report says about the data requests received between July 2017 and December 2017.
The report provides updates on the requests received since 2013, and it specifically notes that Facebook was informed in 2016 of the Supreme Court’s decision amending proper interpretation of the Information Technology Act of 2000, and it ceased to act upon legal requests “to remove access to content unless received by way of a binding court order and/or a notification by an authorized agency which conforms to the constitutional safeguards as directed by the Supreme Court”.
Under fire over the Cambridge Analytica controversy, Facebook has also, for the first time, released statistics for its community guidelines enforcement between October 2017 and March 2018 in the six areas of graphic violence, adult nudity and sexual activity, terrorist propaganda, hate speech, spam and fake accounts.