Increasingly fringe and violent far-right ideologies have become a significant threat to national security in Western countries in the twenty-first century. Accelerationism is growing in far-right narratives while populism and ethnocentrism garnered public attention. This study presents an empirical study of Stormfront, the largest and oldest online white nationalist community in the world, to identify accelerationist themes in the ideological narratives used to discuss white nationalist philosophy over time. The study applied a Natural Language Processing (NLP) algorithm and a supervised learning AI model to a corpus of more than 66 m words collected from 367,000 posts across a 20-year period. This analysis identified two kinds of accelerationist sentiment in the dataset, an offensive form, which portrays violence as both an inevitable and desirable mechanism to pursue political change, and a defensive form, which portrays violence as a reactionary and protective means to prevent undesirable social changes. The study then used a thematic analysis on a subset of the data to support its findings that both offensive and defensive accelerationism are embedded in ideological narratives on Stormfront. This is an important consideration for further research on ways that accelerationism may be influencing popular white nationalist ideology, particularly fringe and violent narratives.