Abstract 

Artificial intelligence is changing public life quickly, but federal oversight remains scattered, voluntary, and lacks strong civil rights protections. This paper calls for an equity-first approach to AI governance that centers Black and Brown communities, HBCUs, and Equity Journalism in efforts to reveal algorithmic harm. Drawing on research on AI governance, human rights, and media accountability, it presents a policy plan to shift federal AI regulation from a safety-focused model to one that is enforceable, anti-racist, and equity-based (Brookings Institution, 2025; GovAI, 2025). The paper concludes with a policy brief and a research plan to build a national equity-first AI governance strategy.
 

Introduction: AI Is Not Neutral. It Is a Mirror of Power 

Artificial intelligence is rapidly changing public life, revealing the power structures and biases that shape society. Without strong safeguards, AI can automate structural racism across housing, hiring, criminal justice, and public benefits (Eubanks, 2018; Noble, 2018). The direction of AI governance will determine whether it reinforces inequality or enables progress for oppressed communities.
 

Black and Brown communities have faced discrimination through technology for generations, and now encounter increased risk from less transparent algorithmic systems. These systems intensify inequality by repeating long-standing practices such as redlining and targeted policing (Benjamin, 2019). This paper calls for equity-first, anti-racist, and enforceable AI governance to ensure that advancements support, rather than harm, these communities. 

The Landscape of Risk: AI as an Engine of Inequality 

AI tools are widely used in education, with a recent report showing high adoption at Historically Black Colleges and Universities. This stresses how AI influences jobs,

education, and access to information. Research shows that biased data and unregulated AI often harm marginalized groups (Association for the Advancement of Artificial Intelligence, 2025). Generative AI also spreads disinformation, threatening voting rights and access to information. Media manipulation studies warn AI can distort democracy and amplify harmful stories (U.N, 2025). Literacy tests once suppressed political power; now AI-driven disinformation operates with greater speed and reach, but the intent to undermine civic participation in targeted communities remains (Laurent 2026). In historically Black neighborhoods, AI impacts daily life from tenant screening to hiring filters without transparency or recourse (O’Neil, 2016). When journalism fails to expose these systems, the harms persist, as when the press ignored past injustices. 
 

Governance and Gaps: The Failure of Voluntary AI Policy 

The United States uses a mix of executive orders, advisory groups, and voluntary guidelines. Studies show that this approach to federal AI regulation is scattered and slow, and it gives too much power to industry self-regulation (Brookings Institution, 2025). Even when federal agencies recognize the risks, like in the Blueprint for an AI Bill of Rights, there is no way to enforce these rules. 
 

This is the main policy failure. The United States sees AI as simply a technical issue, not a civil rights problem. Voluntary guidelines cannot stop systems that automate discrimination. Letting companies police themselves does not protect communities that have faced surveillance and exclusion. Without enforceable federal laws, harm from algorithms is certain. This repeats past times, when civil rights protections were weak or left to those who benefited from inequality (Benjamin, 2019). AI governance will repeat this history unless equity is made the core of federal policy.
 

Equity First AI Governance: A Framework for Structural Change 

​

Reimagining Federal Policy for Justice 

AI governance must prioritize equity by requiring binding civil rights protections for AI systems, independent equity impact assessments before deployment, and strong penalties for violations. Agencies should have clear authority to halt or recall discriminatory systems. According to AP News, the White House has introduced new rules requiring federal agencies to verify that their use of AI tools does not harm public safety or violate civil rights. This means creating rigorous AI auditing tools that require representative datasets and building community governance with impacted communities throughout the design process. Research agendas should center on justice and use cross-disciplinary approaches to prevent racial harm.
 

HBCUs are uniquely positioned to lead this work. Their history in civil rights and their growing strength in AI research make them key to building a national equity-first AI system. Just as HBCUs trained civil rights leaders in the past, they are now leading the way to an AI future that avoids repeating old injustices.
 

What often gets overlooked in national conversations is that HBCUs are not simply vulnerable to AI’s risks; they are positioned to shape its future if given the resources and authority to do so. These institutions already produce nearly 25% of all Black STEM graduates despite receiving a fraction of federal research funding. Imagine what becomes possible when HBCUs are no longer fighting for pieces but are fully funded as AI research hubs, data ethics centers, and community-focused innovation labs. With the right investment, HBCUs could lead the country in developing culturally grounded datasets, equity-centered auditing tools, and AI literacy programs that reflect the actual experiences of Black and Brown communities. The nation keeps asking how to build trustworthy AI. The answer is right in front of us: trust the institutions that have been trustworthy to disadvantaged communities for over a century.
 

Federal policymakers have to stop treating HBCUs as afterthoughts in the AI ecosystem and take strong steps to provide real structural support. These campuses are actively training first-generation students in machine learning, building interdisciplinary AI programs, and experimenting with tools that improve advising, retention, and career placement. What they lack is not vision but support. AI can and must directly benefit HBCUs, but only if the country fully commits to the truth: equity is the foundation of innovation. Give HBCUs a central seat at the table so AI becomes more accurate, accountable, and community-focused. The alternative, a continued cycle of exclusion and bias, is not acceptable. Take action now to ensure a just AI future, as the stakes are generational.
 

Equity Journalism is not peripheral to AI governance. It is infrastructure. Investigative reporting has exposed algorithmic surveillance, discriminatory risk scores, and digital redlining. But when journalism fails to uncover how AI is embedded in law and policy, communities lose the information needed to resist harmful systems. 

Equity Journalism must interrogate. It should question, explain, and drive direct action. Journalists must follow how AI is bought, investigate data misuse, and convert complex AI policy into information the public can use to demand change. In the past, when the press ignored racial injustice, unfair systems strengthened. We cannot allow the same risk now. If journalism fails to show how AI shapes public life, society may wrongly accept algorithmic oppression as progress. Reporters and editors must act now to expose harms and hold institutions accountable so the public can demand just AI.
 

Digital redlining in banking and hiring, algorithm-based evictions and tenant screening, biased facial recognition, and automatic denial of benefits are common outcomes of systems that do not emphasize equity (Eubanks, 2018). These patterns echo earlier eras when Black labor was exploited, Black neighborhoods were over-policed, and Black families were denied economic mobility (Benjamin, 2019). Equity-first governance must interrupt these patterns and redistribute power back to the communities most affected. The stakes are generational.
 

Embedded Policy Brief: Equity First AI Governance 

AI systems in housing, hiring, criminal justice, and public benefits disproportionately harm Black and Brown communities due to voluntary, fragmented, and unenforceable federal AI governance. Biased datasets and unregulated systems harm marginalized groups, while media manipulation twists democratic participation and narratives (Human Rights and AI Research Group, 2024). 
 

To reverse these harms, federal policy must immediately impose enforceable civil rights-centered standards, mandate equity impact assessments, and demand full public reporting on AI procurement and deployment. Fund HBCUs as national hubs for community-centered AI research and provide strong support for journalism as a critical mechanism of public accountability. Lawmakers, educators, and journalists act now to make equity-first AI a national priority. 
 

According to a report from Ellucian, UNCF, and Huston-Tillotson University, making sure that AI governance focuses on equity, is enforceable, and takes into account the lived experiences of underrepresented communities is essential, especially as faculty at HBCUs expect AI use in areas such as student career planning to increase in the coming years. The implications are clear. The United States must adopt enforceable, civil-rights-based AI regulation, build shared governance models, and treat journalism as accountability infrastructure.
 

Conclusion 

It is a political, economic, and moral responsibility. The United States can improve its AI governance strategy by adopting an equity-first approach that treats civil rights as basic to technological innovation. AI governance must be rooted in law, carried by the people, and honest about the realities experienced by the communities standing closest to its consequences.
 

We are living through a defining moment, a crossroads where the nation must choose whether AI will deepen the wounds of the past or help build a more impartial future. One false output spoils a thousand truths when the systems that shape opportunity are built on biased data and unaccountable algorithms. Code without conscience is policy without justice.
 

References 

Association for the Advancement of Artificial Intelligence. (2025). An investigation into Black and Brown communities' engagement with AI systems. AAAI Press. 

Benjamin, R. (2019). Race after technology: Abolitionist tools for the new Jim Code. Polity Press. 

Brookings Institution. (2025). The United States approach to AI regulation: Federal laws and governance. Brookings Governance Studies. 

Eubanks, V. (2018). Automating inequality: How high-tech tools profile, police, and punish the poor. St. Martin’s Press. 

GovAI. (2025). GovAI research agenda. https://cdn.governance.ai/GovAI-Research-Agenda.pdf (cdn.governance.ai in Bing) 

Nations, U. (2025, February 5). Artificial Intelligence and the future of Journalism: Risks and Opportunities. Artificial Intelligence and the Future of Journalism: Risks and Opportunities. https://unric.org/en/artificial-intelligence-and-the-future-of-journalism-risks-and-opportunities/ 

Laurent, A. (2026, January 30). Algorithmic redlining: How ai bias works & how to stop it. IntuitionLabs. https://intuitionlabs.ai/articles/algorithmic-redlining-solutions 

Noble, S. U. (2018). Algorithms of oppression: How search engines reinforce racism. NYU Press. 

O’Neil, C. (2016). Weapons of math destruction: How big data increases inequality and threatens democracy. Crown Publishing.