Recent U.S. AI Policy Actions: Innovation at the Cost of Equity?

In a dramatic shift, the current U.S. Administration has rescinded key safeguards designed to ensure safety, security, and fairness in artificial intelligence. The move, framed as a strategy to accelerate innovation, removes federal oversight on AI bias, equity, and responsible use, setting the nation on a markedly different course from its global peers.

This decision comes just over a year after President Biden’s Executive Order on Safe, Secure, and Trustworthy AI (Oct. 30, 2023), which established guardrails for AI development to protect consumers, workers, and civil rights. With those protections now dismantled, experts warn that the shift could exacerbate systemic disparities in high-risk sectors such as health care, education, finance and criminal justice—industries already grappling with deeply entrenched systemic inequities. [1-2]

Table: Comparing Biden’s AI Executive Order vs. Trump’s AI Executive Order Reversal comparing the policies of the Biden era AI executive order vs. the Trump's Administrations

AI Principle	Biden's AI Executive Order (October 2023)	Trump's AI Executive Order (January 2025)
Safety & Security	Required AI risk assessments, cybersecurity measures, and labeling of AI-generated content.	Removed federal AI safety and cybersecurity guidelines.
Promoting Innovation	Encouraged responsible AI development through education, research, and fair competition.	No federal mandate to monitor AI’s impact on jobs or labor rights.
Workforce Support	Ensured AI advancements benefited workers, protected labor rights, and mitigated job displacement.	No federal mandate to monitor AI’s impact on jobs or labor rights.
Equity & Civil Rights	Preventing AI misuse that exacerbates discrimination and bias, while promoting compliance with civil rights laws.	Eliminated equity and bias requirements, prioritizing "reducing ideological bias, to enable human flourishing and economic competitiveness"
Consumer Protection	Established safeguards to prevent fraud, bias, and harm in critical areas such as health care, lending, and education.	Reduced regulations on AI deployment in consumer-facing industries.
Privacy & Civil Liberties	Strengthened privacy protections, combating risks from improper AI-driven data use.	Rolled back privacy oversight, reducing restrictions on AI data collection.
Federal AI Use	Strengthened government AI oversight, attracting AI professionals and modernizing infrastructure.	Federal AI development is now focused on deregulation and national security applications.
Global Leadership	Encouraged international collaboration to set ethical AI standards worldwide.	Focuses on U.S. AI dominance, diverging from global AI governance efforts.

The Trump administration's AI executive orders mark a sharp shift in governance, prioritizing innovation and economic competition over safeguards for ethical and equitable AI use. While other nations strengthen regulations to mitigate bias and harm, the U.S. appears to be easing protections—raising concerns about the real-world consequences of unchecked AI.

Nowhere is this more evident than in health care, where biased AI tools reinforce racial disparities. Race-based medicine, long embedded in clinical algorithms, has led to Black patients receiving inadequate care, particularly in conditions like kidney disease, where race-adjusted formulas can delay lifesaving interventions. [3] Similarly, an AI-driven resource allocation system used for over 20 million patients disadvantaged Black individuals by linking care needs to past health spending—an inherently biased metric. [4]

Beyond health care, flawed AI systems continue to harm marginalized communities. UnitedHealth’s AI-driven denial system had a high error rate, wrongly refusing essential care for older adults.[5] Facial recognition technology has led to wrongful arrests due to misidentification. [6] In finance, AI-driven lending decisions, reliant on biased FICO scores, have disproportionately denied Black homebuyers—who, on average, have credit scores 57 points lower than their white counterparts.[7]

With the administration pulling back from addressing systemic inequities, the rollback of these protections raises urgent questions: Who will bear the brunt of AI-driven harm in a regulatory vacuum? And what safeguards, if any, will remain?

Global AI Regulation and the U.S. AI Policy Actions: A Growing Divide

As the Trump administration pushes to "Remove Barriers to American Leadership in Artificial Intelligence," its pivot away from safe, responsible AI risks eroding global trust in U.S. technology. While the U.S. moves toward deregulation, other nations are tightening protections.

The European Union's 2024 AI Act enforces a risk-based framework to regulate AI applications, ensuring higher safeguards for high-risk uses. The World Health Organization has issued ethical guidelines, including AI-specific recommendations for health care. Meanwhile, China was among the first to introduce AI regulations in 2021, setting early guardrails for its development.

As global oversight strengthens, the U.S.’s retreat from AI safeguards raises a pressing question: Will deregulation position America as an innovation leader, or leave it out of step with the world’s AI governance?

The Risks of AI Without Adequate Guardrails

Without oversight, AI-driven decision-making worsens disparities across key sectors:

🔴 Health Care → AI-powered diagnostics bake in biases related to race, gender, age, and socioeconomic status, leading to unequal access to care and poorer outcomes for marginalized groups.

🔴 Employment & Hiring → AI hiring tools reinforce systemic discrimination, filtering out qualified candidates based on biased algorithms.

🔴 Criminal Justice → AI-driven policing and sentencing models exacerbate racial profiling and perpetuate injustices in the legal system.

🔴 Financial Services → Algorithmic lending and credit scoring systematically exclude marginalized communities, widening economic inequality.

As AI’s influence grows, the absence of strong safeguards ensures that these harms will persist, disproportionately impacting those already facing structural inequities.

Conclusion: Racing to Innovate—But at What Cost?

The Trump administration’s rollback of the 2023 Biden Executive Order on AI marks a decisive shift in U.S. AI policy action, stripping away safeguards designed to ensure fairness and accountability. While deregulation may spur innovation, it comes at the expense of public trust, equity, and the long-term viability of responsible AI.

As the European Union, China, and the World Health Organization advance stricter AI regulations, the U.S. is charting a different course—one that risks deepening systemic inequities and eroding its leadership in ethical AI governance. The question remains: Will the pursuit of innovation come at the cost of those most vulnerable to AI’s harms?

🚀Stay Ahead of of the AI Curve: Subscribe: Blog | CROSS Global Research & Strategy for weekly insights on how AI is reshaping health care, policy, and equity. We analyze the latest research, expert opinions, and real-world implications—so you don’t have to.

💬 Join the Conversation: Have thoughts on AI governance and its impact? Share your insights in the comments, like this article, and be part of the discussion shaping the future of responsible AI.

References

1. Trump's Efforts to Dismantle AI Protections, Explained. American Civil Liberties Union News and Commentary. Available from: https://www.aclu.org/news/privacy-technology/trumps-efforts-to-dismantle-ai-protections-explained

2. A technical AI government agency plays a vital role in advancing AI innovation and trustworthiness. Brookings. Available from: https://www.brookings.edu/articles/a-technical-ai-government-agency-plays-a-vital-role-in-advancing-ai-innovation-and-trustworthiness/

3. Neal RE, Morse M. Racial Health Inequities and Clinical Algorithms: A Time for Action. Clin J Am Soc Nephrol. 2021 Jul;16(7):1120-1121. doi: 10.2215/CJN.01780221. Epub 2021 Mar 5. PMID: 34597267; PMCID: PMC8425626.

4. Obermeyer Z, Powers B, Vogeli C, Mullainathan S. Dissecting racial bias in an algorithm used to manage the health of populations. Science. 2019 Oct 25;366(6464):447-453. doi: 10.1126/science.aax2342. PMID: 31649194.

5. https://www.cbsnews.com/news/unitedhealth-lawsuit-ai-deny-claims-medicare-advantage-health-insurance-denials/

6. https://www.abc.net.au/news/2023-04-26/hidden-dangers-of-artificial-intelligence/102264038

7. Aaron Klein. Reducing Bias in AI-based Financial Services. Available from: https://www.brookings.edu/articles/reducing-bias-in-ai-based-financial-services/

8. European Union Artificial Intelligence Act (2024). Available from: EU Artificial Intelligence Act | Up-to-date developments and analyses of the EU AI Act

9. WHO guidance: Ethics and Governance of Artificial Intelligence for Health. Available from: https://www.who.int/publications/i/item/9789240029200

10. Ethics and Governance of Artificial Intelligence for Health: Guidance on Large Multi-Modal Models (WHO),2024. Available from: https://www.who.int/publications/i/item/9789240084759

11. Tracing the Roots of China's AI Regulations (2024). Available from: https://carnegieendowment.org/research/2024/02/tracing-the-roots-of-chinas-ai-regulations?lang=en