Twitter claims headway in removing harmful accounts

The volume of accounts removed by Twitter relating to terrorism and child sexual exploitation continued to fall in the second half of 2018, which the company says is proof its efforts are working.

In its biannual Twitter Transparency Report released Thursday, the social-media platform revealed it suspended over 166,000 accounts between July to December 2018 for violations related to promotion of terrorism, which is 19% less than in the previous period.

Of the accounts suspended, 91% were flagged by internal technology, while the majority of accounts were suspended before they were even able to tweet, the company said.

“The trend we are observing year-on-year is a steady decrease in terrorist organizations attempting to use our service,” the company’s legal, policy and trust and safety lead, Vijaya Gadde, wrote in a blog post.

“We are encouraged by these metrics but will remain vigilant. Our goal is to stay one step ahead of emergent behaviors and new attempts to circumvent our robust approach.”

Twitter suspended 456,989 accounts for violations related to child sexual exploitation, which is down 6% from the previous reporting period. Of those, 96% were surfaced by a combination of technology solutions, including PhotoDNA and internal proprietary tools.

Japanese government requests on the rise

Elsewhere in the report, Twitter revealed that Japan submitted a higher volume of global government information requests in the period, accounting for 24% of global requests, up from 21% in the previous reporting period. The proportion of requests made by the US fell from 32% to 30%.

The UK submitted the greatest percentage of global emergency disclosure requests for the first time at 33%, followed by the United States at 30%.

Twitter received around 8% fewer global legal requests to remove content compared to the previous reporting period, however, there was an 84% increase year-on-year between 2017 and 2018.

Nearly three-quarters (74%) of the total global volume of legal requests to remove content originated from only two countries: Russia and Turkey.