This year’s report marks the thirteenth time in the company’s history that Twitter has expounded on its privacy practices in such tremendous detail. The new report also expands on the type of data Twitter is making publicly available.
In a published in tandem with the Transparency Report, Twitter explains that for the first time ever, along with government requests for information and content removal, the company has included stats on the enforcement of Twitter’s own policies.
This report covers the first half of 2018, and the last half of the year will be covered in an update at a later time, typically published in the first quarter of the ensuing year. The data in the report includes requests made on Twitter, Periscope, and even the now-defunct Vine.
Perhaps the most shocking detail from all of this is the gigantic uptick in government requests for information.
However, much like the last reporting period, around 87 percent of those demands came from just two countries: Russia and Turkey. Altogether legal demands to remove content came from 38 different countries for a total of 27,811 Twitter accounts.
When it comes to information requests, Twitter received a 10 percent uptick when compared to the previous reporting period.
The U.S. had the greatest number of with 2,231 demands on a total of 9,226 accounts. This represents around one-third of all Twitter’s demands for the first-half of the year, according to the Transparency Report.
Twitter complied with 76 percent of the U.S. information requests. There were also 99 U.S. removal requests on 228 accounts. Twitter did not comply with those requests.
Twitter’s report breaks down its inaugural data into 6 categories: abuse, child sexual exploitation, hateful conduct, private information, sensitive media, and violent threats. In total, 6,229,323 accounts were reported for Twitter rule violations. Action, which includes suspension, was taken on 605,794 of those accounts.
In total, Twitter actually suspended a total of 487,363 accounts for child sexual exploitation violations. However, 97 percent of those suspensions were proactively flagged thanks to PhotoDNA and other technologies. Around 91 percent of the 205,156 accounts violating terrorist content policy were also proactively flagged.
Along with Twitter policy violation, data also makes its debut in this Transparency Report. There were 232,453,596 accounts challenged, or proactively detected for possibly being spammy or manipulative, in the first half of 2018 by the company. In addition, there were 4,020,893 spam reports from Twitter users.