This dissertation critically examines whether the Online Safety Act 2023 provides an adequate legal framework for addressing racist hate speech on digital platforms in England and Wales. It argues that, while the Act represents a significant advance on the pre-existing framework under the Public Order Act 1986, the Communications Act 2003, and the Crime and Disorder Act 1998, it retains four structural weaknesses: it incorporates the definitional limitations of the earlier criminal law; it does not require platforms to address legal but harmful racist content for adult users; it does not extend its algorithmic amplification duties beyond illegal content; and early implementation evidence indicates uneven platform compliance. The Southport disorder of 2024 and the prosecutions that followed illustrate both the capacity and the limits of the existing criminal law, and demonstrate why platform-level regulation must go further. Drawing on a doctrinal methodology and criminological frameworks including Merton’s Strain Theory, the Frustration-Aggression Hypothesis, and the Self-Fulfilling Prophecy, the dissertation proposes four targeted reforms grounded in comparative analysis of the EU Digital Services Act and concludes that effective regulation of online racist hate speech is both legally achievable and constitutionally required.