Facebook returns with a response to the latest criticism of its platform, which states in a long new statement that it drastically reduced the number of hatred speeches that users have seen during the last three quarters. The company focuses on the prevalence of hatred speech, which describes it as a content that is truly seen by users, not the total number of problematic content found on the platform.
Facebook claims that with almost 50 percent reducing prevalence during the last few quarters, hatred speaks only about 0.05 percent of the content that users see; It numbered around five viewers for everyone 10,000. Among other things, Facebook says it is proactively using various technologies to detect troubled content and shuttle to reviewers for potential deletion.
The statement came from VP Facebook Integrity Guy Rosen, which specifically displays the release of recent leaky content in a report by the Wall Street Journal. In his post, Rosen said, among others:
Data drawn from the leak document is used to create a narrative that the technology we use to fight inadequate hatred speeches and that we deliberately misinterpret our progress. This is not true. We don’t want to see hatred on our platform, also not our users or advertisers, and we are transparent about our work to delete it. What these documents show is our integrity work is a multi-year trip. Even though we will never be perfect, our team continues to work to develop our system, identify problems and build solutions.
Rosen continued to repeat that in Facebook’s view, the prevalence of hatred speech on the platform was the most important metric. He specifically touched the controversial habit of leaving hatred speech on the platform that did not meet the bar for removal, ‘noting that the Facebook system instead reduced its distribution to users.
Rosen said:
We have a high threshold for deleting content automatically. If not, we will take the risk of making more mistakes on content that looks like hatred speech but no, hurt the people we try to protect, as describing experience with hatred speech or cursing it.