I. Introduction

This section points out that American racism persists and intensifies in the digital age through a new form of algorithm-driven racial inequality. Characterized by its pervasiveness, this phenomenon inflicts harm on ethnic minorities and poses challenges to national governance. Academic research on algorithm-driven racial inequality in the United States is insufficiently systematic. Existing studies often focus on digital racism and manifestations of algorithmic inequality in Europe and the United States, with a notable lack of research on governance in the American context. This article aims to analyze the developmental features of this issue in the United States, construct an analytical framework for governance, and explore progress and challenges in governance efforts.

II. Main Characteristics of Algorithm-Driven Racial Inequality in the United States

This section delves into the phenomenon of algorithm-driven racial inequality in the United States, analyzing its specific manifestations and impacts across various social domains. In the healthcare sector, algorithmic bias is evident in clinical research and healthcare policy formulation, leading to unfair treatment of African Americans and other ethnic minorities in medical coverage. In the education sector, algorithm-driven racial inequality affects student assessment systems and university admissions processes, placing minority students at a disadvantage in obtaining academic funding and admissions. In the employment sector, recruitment processes and job evaluations based on large language models and generative artificial intelligence exhibit racial discrimination, impacting the job opportunities of minorities. In policing and the judicial system, algorithmic biases in facial recognition systems and recidivism prediction systems make minorities more susceptible to misidentification and unjust treatment. In the financial sector, algorithmic discrimination in loan services subjects African American and Hispanic applicants to unequal treatment in loan applications and interest rates. These phenomena reveal how algorithm-driven racial inequality significantly and negatively impacts the survival and development of ethnic minorities across multiple fields.

III. Progress in Governing Algorithm-Driven Racial Inequality in the United States

This section explores in detail the progress the United States has made in addressing algorithm-driven racial inequality. At the federal level, the Biden administration has prioritized governing algorithmic racial inequality through executive orders such as the release of the "Blueprint for an AI Bill of Rights," requiring developers and users to take measures to prevent algorithmic discrimination and promoting related legislative work, such as the "Algorithmic Justice and Online Platform Transparency Act." Concurrently, federal functional departments like the Consumer Financial Protection Bureau have also taken active steps to jointly address the issue of algorithmic discrimination. At the state level, starting in 2023, states including Colorado and California began advancing legislation to regulate algorithm-driven racial inequality, requiring employers to assess and report on their use of hiring software and inform applicants accordingly. In 2025, state-level legislative activities increased significantly, exemplified by California's "Frontier AI Transparency Act." At the corporate and social organization level, leading American AI companies like Amazon and Google have made self-regulatory commitments, publicly reporting AI capabilities and limitations, increasing funding for research on algorithmic discrimination, and establishing ethics review bodies or guidelines. Social organizations such as the Algorithmic Justice League and the American Civil Liberties Union actively participate in governance through analysis, exposure, litigation, and other means, promoting community collaboration for technical solutions, and providing STEM education and skills training to communities of color to enhance their capacity to participate in technology development and governance.

IV. Governance Challenges in Addressing Algorithm-Driven Racial Inequality in the United States

This section explores the governance challenges faced by the United States in dealing with algorithm-driven racial inequality. The multi-centered governance system of the United States struggles to make rapid progress on this issue due to the lack of a mature governance framework and institutional system. At the government level, structural conflicts in ideology and partisan interests between the two major parties hinder collaboration at the federal and state levels, making it difficult to form medium- to long-term top-level designs. In the market dimension, American AI companies, with innovation and profitability as core objectives, find it challenging for industry self-regulation to serve racial equality, and disagreements over government constraints on enterprises further exacerbate governance difficulties. On the social dimension, the resurgence of racism has widened the ideological divide among political elites and the public regarding racial governance, making social consensus difficult to achieve. Partisan ideological differences and political polarization lead to legislative gridlock, with virtually no relevant laws at the federal level and slow progress in state-level legislation. Election politics triggers a policy pendulum effect, where significant differences in policy approaches between presidents from different parties impact governance effectiveness. The resurgence of racism erodes the foundation for governance consensus, while biased data underlying algorithmic decisions perpetuates algorithmic discrimination. Some political elites and the public oppose governance efforts, and corporate attention to governance has waned.

V. Conclusion

This section concludes that algorithm-driven racial inequality in the United States is a continuation of racism in the age of artificial intelligence, manifesting as the practice of racial hierarchy within the algorithmic domain to maintain white dominance. Current governance efforts have been ineffective due to a lack of top-level design and legal support, compounded by partisan ideological differences, political polarization, and the resurgence of racism. The U.S. government neglects the erosion of minority human rights by algorithms. In the future, algorithm-driven racial inequality will combine with various forms of racism to form a complex entity, posing an ongoing challenge to American society.