Tumblr on Monday announced an update to its hate speech policy, announcing on its blog that it would start cutting down on the reach of posts considered hate speech by removing all reblogs of terminated content. Tumblr says it’s specifically targeting speech from Nazis and other white supremacist groups, which it says still be amplified on its platform through reblogs even after the original post was removed for violating its hate speech policies.
“Many of you have called on us to further reevaluate how we deal with hate speech, particularly hate speech from Nazis or other white supremacist groups. Today we’re letting you know that we heard you, and we are taking further action,” Tumblr, which is now owned by WordPress creator Automattic, writes. “In our own research, and from your helpful reports, we found that much of the existing hate speech stemmed from blogs that have actually already been terminated. While their original posts were deleted upon blog termination, the content of those posts still lived on in reblogs.”
Tumblr says reblogs “rarely contained counter-speech to keep hateful rhetoric in check”
Tumblr says that while not all reblogs were amplifying the message and some were in fact dedicated to debunking it or arguing against it, but that the reblogs “rarely contained the kind of counter-speech that serves to keep hateful rhetoric in check.”
Now, Tumblr says it will go through terminated posts, with a focus on posts propagating Nazi viewpoints and other white supremacist ideology, to remove all associated reblogs. Tumblr says in the last week alone, it removed 4.47 million reblogs as part of this process. “Moving forward, we will evaluate all blogs suspended for hate speech, and consider mass reblog deletion when appropriate,” the post reads.
Tumblr says it recognizes the effect this might have on educational and social justice-related arguments. “We wouldn’t make a change like this without considering the impact to your freedom of expression. We do not want to silence those who are providing educational and necessary counter-arguments to hate speech,” the blog reads. Tumblr says it reviewed its approach with a variety of outside groups to “make sure we have aligned with their recommended best practices.”
“There’s no silver bullet solution, AI, or algorithm that can perfectly target hate speech. That’s why we have a dedicated Trust & Safety team, and why we have an easy way for you to report any hate speech you do see,” the company says.
Tumblr’s approach to moderation has landed in hot water with its community in the past, most prominently when the platform banned all pornographic and other forms of adult content when it was still owned by Verizon. The site suffered a steep drop in web traffic as users ditched the platform in large numbers. Automattic maintained the ban when it purchased Tumblr in August of last year.