Facebook's latest report on how it tackled hate speech and bullying in Q4 2020

Facebook said that its proactive rate for bullying and harassment went from 26% in Q3 to 49% in Q4 on Facebook, and 55% to 80% on Instagram.

Facebook

Facebook has released a new transparency report for the fourth quarter of 2020. Its “Community Standards Enforcement Report” shows how the company has tackled hate speech and abusive content on the platform.

Facebook's latest transparency period covers the US presidential election. During this period Facebook said hate speech prevalence dropped from 0.10-0.11% to 0.07-0.08%, while violent and graphic content dropped from 0.07% to 0.05%. Prevalence of adult nudity content dropped from 0.05-0.06% to 0.03-0.04% during the said quarter. Facebook attributes this improvement to changes in the News Feed by identifying posts that violate its policies and removing them.

Facebook also said that its proactive rate for bullying and harassment went from 26% in Q3 to 49% in Q4 on Facebook, and 55% to 80% on Instagram. The company highlighted how improvements in AI helped it tackle areas such as hate speech and bullying. In total, it took action on 6.3 million and 5 million pieces of bullying and harassment on Facebook and Instagram, respectively.

Facebook had to rely on AI for content moderation due to Covid-19. The company said until a vaccine is widely available, content moderation will still be affected. While it does have human moderators back in office in limited capacity, it is prioritising the most harmful content including suicide and self-injury. A total of 2.5 million pieces of suicide and self-injury content were taken action on during this period.

The company plans to share additional metrics on Instagram and add new policy categories on Facebook this year.