https://hackerpaper.kaipfister.de/p/2003.03318/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=27760430urn:hackerpapers:mention:2003.03318:27760430Wed, 07 Jul 2021 12:42:30 GMT<p>Both happens. Some studies such as [1] measure recommendations without past browsing history, and they still find recommendations for harmful content. Other studies (such as this from Berkman Klein Center [2]) simulate user behavior and subsequent recommendations and find additional harmful effects of personalized recommendations. [1] Faddoul, M., Chaslot, G., &amp; Farid, H. (2020). A Longitudinal Analysis of YouTube’s Promotion of Conspiracy Videos. ArXiv:2003.03318 [Cs]. http://arxiv.org/abs/2003.03318 [2] https://cyber.harvard.edu/story/2019-06/youtubes-digital-pla...</p><p><a href="https://hackerpaper.kaipfister.de/p/2003.03318/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CYhttps://news.ycombinator.com/item?id=26400806urn:hackerpapers:mention:2003.03318:26400806Tue, 09 Mar 2021 17:02:43 GMT<p>This is a terrible piece we know that Algos have caused bubbles because they chase engagement. https://www.algotransparency.org/ https://arxiv.org/pdf/2003.03318.pdf https://www.brookings.edu/techstream/how-youtube-helps-form-...</p><p><a href="https://hackerpaper.kaipfister.de/p/2003.03318/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CY