The social network on Thursday published a blog post detailing the highlights from its 12th biannual Twitter Transparency Report, where it reveals information about accounts it’s removed, inquiries from governments, and data requests. In that post, Twitter revealed that it removed nearly 275,000 terrorist accounts between July 1, 2017 and December 31, 2017. While that was down 8.4% compared to the prior period, the company revealed that it has now banned more than 1.2 million terrorist accounts from its service since August 2015.

In the blog post, the company said that it’s been working for years at making Twitter “an undesirable place for those seeking to promote terrorism.” The company added that online terrorist-related activity is now “increasingly shifting away from Twitter.”

Get Data Sheet, Fortune’s technology newsletter

Like other major tech companies, Twitter shares transparency reports to shed light on what kinds of content is on its network and why certain accounts were removed. In some cases, those accounts are suspended or data accessed because of legal orders or warrants. In others, accounts simply violate Twitter’s terms of service and are removed.

Terrorist organizations have long used social media and other popular services to spread propaganda and recruit new members into their organizations. And although Twitter was quick to celebrate its terrorist-related suspensions, the sheer number of removals in the last quarter, alone, suggest the service is still a sought-after spot for terrorists to land.

According to Twitter, 93% of the terrorist accounts were spotted by its own technologies and nearly three-quarters of accounts were banned before they could post their first tweet. Governments stepped in to alert Twitter to terrorist accounts in 0.2% of cases in the second half of 2017.