YouTube announces 70 percent decline in watch time of 'borderline content' by non-subscribers in US

In June this year, YouTube made some changes to the platform’s recommendation algorithm with an aim to reduce ‘borderline content’ — content that lies somewhere between what’s acceptable on the platform and what violates its terms of service. Six months later, in an official blog post, YouTube has announced that there has been a decline by 70 percent in the watch time of ‘borderline content’ by non-subscribers in the US. YouTube says that through this year, it has made over 30 changes to the way its recommendation tools in order to tackle viewership of borderline content. [caption id=“attachment_7632541” align=“alignnone” width=“1024”]  YouTube. Image: Getty[/caption] The platform hasn’t revealed the specifics of what kind of changes has it made this year, or the exact number of videos that were being recommended before and after the change. However, YouTube reveals how external moderators go through specific criteria to determine whether a flagged video is borderline; categorised videos are then used to educate machine learning tools. YouTube also says that while it’s unable to remove the borderline content, it tries to push content from trustworthy sources in recommendations. “While there will be slight variations, on average, 93 percent of the videos in global top 10 results come from high-authority channels.” “Authoritativeness is also important for evergreen topics prone to misinformation, such as videos about vaccines. In these cases, we aim to surface videos from experts, like public health institutions, in search results. Millions of search queries are getting this treatment today and we’re continually expanding to more topics and countries,” YouTube says in the blog post.