<![CDATA[Social media has become one of the primary platforms through which individuals communicate, express opinions, and construct online identities. Although these platforms support freedom of expression, they also enable harmful communication, including online toxicity. Social media influencers, given their large audiences and persuasive reach, play an important role in shaping online behavior and discourse. This study examines how online toxicity is discursively constructed and promoted in the content of a single TikTok and YouTube influencer, referred to as Jazmine, based on a corpus of six videos: five uploaded directly by the influencer and one obtained from a third-party commentary channel. The analysis draws on Norman Fairclough's three-dimensional model of Critical Discourse Analysis (CDA) to examine three interconnected dimensions: the textual features of Jazmine's videos, the processes of content production and audience interaction, and the broader sociocultural context in which the discourse is produced and consumed. Particular attention is given to linguistic choices, rhetorical strategies, and communicative patterns that contribute to the construction and normalization of toxic behavior. The findings indicate that Jazmine's content displays recurring instances of derogatory language, mockery, provocation, and hostile expression framed as humor or entertainment, and that these discursive strategies appear designed to encourage audience engagement while presenting aggressive communication as acceptable online behavior. As a single-case study, these findings illustrate rather than generalize how influencer-generated content can reinforce toxic discourse by framing harmful interactions as entertaining and socially rewarding. The study contributes to Critical Discourse Analysis and digital media studies by highlighting the role of social media personalities in shaping online communication and by underscoring the need for digital literacy and responsible content creation.]]>
