YouTube Recommendations and Implications for Sharing Across Online Social Platforms
Proceedings of the ACM on Human Computer Interaction Volume 5 CSCW1 April 2021 Article number: 11 pp 1-26 https://doi.org/10.1145/3449085
Commentary
In January 2019, YouTube announced that while the platform would remove potentially harmful content from its video recommendations, it would leave such videos on the platform.
While this action aims to reduce YouTube's role in the propagation of such content, there is an open question as to whether such action will actually affect the sharing of these videos in the broader information space by making them continue to be available via hyperlink in other online spaces.
This question is particularly important because other online platforms deploy similar suppressive actions that stop short of deletion, even though they have limited understanding of the impact of such actions.
The study applies an interrupted time series model to assess the impact and measures whether the sharing of potentially harmful YouTube videos on Twitter and Reddit has changed significantly in the eight months around YouTube.
We evaluate video sharing across three selected sets of antisocial content: a set of conspiracy videos that have been shown to be decreasingly recommended on YouTube, a set of videos posted by conspiracy-oriented channels, and a set of videos posted by alternative As a control, we also evaluate these influences on the dataset of videos from mainstream news channels.
The results show that conspiracy-labeled AIN videos with evidence of YouTube deprecation are significantly less likely to be shared on both Twitter and Reddit.
However, videos from conspiracy-oriented channels have actually seen a significant increase in sharing on Reddit, following YouTube's intervention.
This suggests that these actions may have unintended consequences in promoting less obviously harmful conspiratorial content.
Mainstream news sharing likewise shows increasing trends across both platforms, suggesting that YouTube's suppression of certain content types is having a targeted effect.
