The Truth Behind the 'Black GPT' : A Comparative Study of Latimer AI and the Systemic Racism Dismantler

Abstract

Recently, an AI model marketed as "inclusive" or known as the "Black GPT" has claimed to address racial biases; however, its effectiveness remains questionable. This study critically examines Latimer AI, a commercial AI system promoted for its inclusivity, and compares it to the Systemic Racism Dismantler, the prototype for an African-centered language model specifically designed to dismantle systemic racism. Leveraging rigorous tests on white fragility, this paper reveals that Latimer AI falls short in understanding and articulating foundational concepts of racism, often avoiding critical discussions around white supremacy. In contrast, the Systemic Racism Dismantler excels by embedding African-centered scholarship into its framework, demonstrating a deeper, culturally grounded perspective on racism white supremacy. This comparative analysis exposes the limitations of Latimer AI and underscores the necessity for genuinely inclusive AI systems. The findings challenge existing standards for "inclusive" AI and advocate for models grounded in counter-racist principles to effect meaningful social change.

Keywords: White Fragility, Racial Bias in AI, Systemic Racism, Inclusive AI, African-Centered Scholarship, Latimer AI, Systemic Racism Dismantler, White Supremacy in AI, Counter-Racist Technology, Psychological Underpinnings of Racism, Frances Cress Welsing’s Theory of Racism