Twitter’s efforts to cull pro-terror content seems to be working

Twitter’s latest transparency report reveals that the platform’s effort to cut down on pro-terror content is working.

According to the report, the first half of the year saw just under 300 000 accounts removed for promoting internationally recognized terror movements or the violence associated with them. This is a 20% drop from the previous reporting period.

The company recently joined forces with the likes of Facebook, YouTube and Microsoft to share information on how to combat pro-terror content, including AI that can swiftly detect dangerous content.

The report says that 95% of the removed terror accounts were the result of internal efforts (up 74% from last period), and 75% of them were suspended before posting their first tweet. This is the first time Twitter has reported this metric.

These improved internal efforts meant that there was an 80% reduction in pro-terror accounts reported by governments — and only 1% of these account suspensions were a result of government reports.

The total number of removed pro-terror accounts between 1 August 2015 and 30 June 2017 is 935 897.

The vast majority of government reports cited abusive behaviour, which includes violent threats, harassment, hateful conduct and impersonation. Only 12% of the 16 414 reports received were found to be in violation of its terms of service.

The United States government placed 2111 of these requests, submitting 33% of all government requests. In second was the Japanese government with 1384 requests.

Interestingly, South Korea submitted 63% more information requests than the last period. India sent in 55% more.

Featured image: Hamza Butt via Flickr (CC 2.0, resized)