Topic shifts occur naturally during conversations when a person changes the subject to one different from the original. This phenomenon may be extremely meaningful, being useful for modeling dialogues and measuring information distortion during a conversation, among other tasks. In this work, we explore topic shifts as a metric for the measurement of politicization during the 2024 U.S. presidential elections, using YouTube news as a case study. We find evidence of politicization during the studied period as over 69% of non-political videos contain at least one political comment. This politicization increases as we get closer to the date of the election, with videos from right-leaning channels having over 40% of comments being political in the week of the elections. We also identify topics relating to immigration to be the most politicized, as commenters discuss the government’s stance on immigration, aggravated by displays of xenophobia, exemplifying the dangers that come with politicization.