This chapter will discuss the techniques developed over time for hate speech detection. We will explore how hate speech detection techniques have evolved, ranging from keyword-based methods and machine learning techniques to deep learning models and, currently, utilizing generative AI-based large language models such as ChatGPT. We formulate the hate speech detection task as follows: Given a dataset \({\textbf {D}}\) consisting of pairs \(({{\boldsymbol{X}}}, {\textbf {Y}})\) , where \({{\boldsymbol{X}}} = \{w_1, w_2, \ldots , w_m\}\) represents a text sample consisting of a sequence of words, and \({\textbf {Y}}\) represents its corresponding label, the goal is to learn a classifier \(F: F({{\boldsymbol{X}}}) \rightarrow {\textbf {Y}}\) that can accurately predict the presence or absence of hate speech in unseen text samples, where \({\textbf {Y}} \in \{y_1, y_2, \ldots , y_n\}\) is the ground-truth label.