YouTube has revolutionized the way people discover and consume video.\nAlthough YouTube facilitates easy access to hundreds of well-produced and\ntrustworthy videos, abhorrent, misinformative, and mistargeted content is also\ncommon. The platform is plagued by various types of problematic content: 1)\ndisturbing videos targeting young children; 2) hateful and misogynistic\ncontent; and 3) pseudoscientific misinformation. While YouTube's recommendation\nalgorithm plays a vital role in increasing user engagement and YouTube's\nmonetization, its role in unwittingly promoting problematic content is not\nentirely understood. In this thesis, we shed some light on the degree of\nproblematic content on YouTube and the role of the recommendation algorithm in\nthe dissemination of such content. Following a data-driven quantitative\napproach, we analyze thousands of videos on YouTube, to shed light on: 1) the\nrisks of YouTube media consumption by young children; 2) the role of the\nrecommendation algorithm in the dissemination of misogynistic content, by\nfocusing on the Involuntary Celibates (Incels) community; and 3) user exposure\nto pseudoscientific content on various parts of the platform and how this\nexposure changes based on the user's watch history. Our analysis reveals that\nyoung children are likely to encounter disturbing content when they randomly\nbrowse the platform. By analyzing the Incel community on YouTube, we find that\nIncel activity is increasing over time and that platforms may play an active\nrole in steering users towards extreme content. Finally, when studying\npseudoscientific misinformation, we find that YouTube suggests more\npseudoscientific content regarding traditional pseudoscientific topics (e.g.,\nflat earth) than for emerging ones (like COVID-19) and that these\nrecommendations are more common on the search results page than on a user's\nhomepage or the video recommendations section.\n