Hate Speech

Hate speech can appear in text, images, audio, video, or AI-generated content. It may include derogatory statements, dehumanizing language, hateful stereotypes, or calls for discrimination or violence against protected groups. In AI systems, content classifiers and moderation policies can be used to identify and restrict hateful content while considering context.

Generative AI systems can produce or amplify harmful content at scale. Detecting hate speech helps organizations protect users, enforce acceptable-use policies, reduce harmful interactions, and maintain safer digital environments. Effective moderation also needs to distinguish hateful content from legitimate discussion, reporting, education, or criticism.