The consumption of borderline content that comes from YouTube’s recommendations is significantly below 1 per cent, and the company has an ambitious goal to get this number even below half of the per cent. YouTube has been trying to classify what it calls as borderline or harmful misinformation. These are the videos that are close to the border of the community guidelines but don't cross over so that they can remain on the platform.
YouTube Recommendations was built on the principle of helping people find the videos they want to watch that will give them value, which the company says plays an important role in maintaining YouTube as a responsible platform. “We don't want the things that are sort of right now on the edge to be promoted by our systems,” says Cristos Goodrow, VP Engineering, YouTube. “We have built a system that classifies, filters and demotes these borderline videos from recommendations. And even though this borderline content and misinformation were never a large portion of the viewership on YouTube, I'm proud to say today that it's actually less than 1% of the overall views viewership of YouTube. When we launched this classifier in 2019, we saw a 70% drop in the watch time of this content, and we actually now have an ambitious goal to get this number even below half of the per cent.”
Unlike other platforms, YouTube claims that it does not connect viewers to content through their social network. Instead, it is about predicting the videos the viewer wants to watch - based on their own interest and preferences, and not based on the people they are connected with. YouTube also provides viewers with the controls to manage what and how much they want to share to get a personalised experience on YouTube. For instance, viewers who do not want personalized recommendations can choose to delete watch history.
“Today, Recommendations drive a significant amount of the overall viewership on YouTube, even more than channel subscriptions or search. And, we are thinking about it in a responsible way. Our goal is to help connect viewers to high-quality information by minimizing the chances of them seeing problematic content,” adds Cristos Goodrow.
With many people, today, coming on YouTube not only for entertainment but also for news and information, Cristos explains that YouTube’s recommendations are built keeping all kinds of audiences in mind. YouTube launched a recommendations system, ranking videos based on priority in 2008. It was followed by classifiers to identify videos that were racy or violent, which prevented them from recommendations.
Also read: YouTube bans all anti-vaccine content
Also read: YouTube introduces 'Super Thanks' feature to reward creators; here's how it works
Copyright©2022 Living Media India Limited. For reprint rights: Syndications Today