Leadership Conference Comments to National Science Foundation on National AI R&D Strategy
A pdf of the letter is available here.
Suzanne H. Plimpton
Reports Clearance Office
National Science Foundation
24314 Eisenhower Avenue
Alexandria, VA 22314
Submitted electronically to https://www.regulations.gov.
RE: Request for Information on the Development of a 2025 National Intelligence (AI) Research and Development (R&D) Strategy, Docket ID No. NSF-2025-OGC-0001
Dear Ms. Plimpton:
On behalf of The Leadership Conference on Civil and Human Rights, a coalition charged by its diverse membership of more than 240 national organizations to promote and protect the rights of all persons in the United States, its Center for Civil Rights and Technology, and the National Fair Housing Alliance (NFHA), we appreciate the opportunity to respond to the National Science Foundation and Office of Science and Technology Policy Request for Information (RFI) on the development of a National AI Research and Development (R&D) Strategy for 2025.[1] This response is informed by the insights and recommendations of the The Leadership Conference and NFHA, as well as other experts who focus on the impact technologies like artificial intelligence (AI) is having on our communities. Our work makes clear that the National AI R&D strategy must be centered on civil rights, equity, and the protection of historically marginalized communities.
Recognizing that the goal of the National AI R&D Strategy is to further the advancement of technology, like AI, we note that this goal will only be achieved if people can trust that the technology is benefiting and not harming them. As AI technologies become increasingly integrated into critical sectors such as housing, employment, health care, education, and financial services, it is essential that the national strategy reflects a strong commitment to technology that benefits everyone.
Specifically, the National AI R&D Strategy must:
Center Civil Rights: AI systems have the potential to and are already significantly impacting people’s lives. However, without intentional oversight, AI can perpetuate and even exacerbate existing inequalities. The National AI R&D Strategy must prioritize the development of frameworks that ensure AI systems are fair, transparent, and accountable. Civil rights must be embedded at every stage of the AI lifecycle—from data collection and model training to deployment and monitoring.
Address Algorithmic Discrimination: As highlighted by The Leadership Conference, individuals can face discrimination from automated systems based on race, ethnicity, gender, and other protected characteristics.[2] For example, hiring algorithms have been shown to replicate gender biases,[3] and mortgage lending algorithms have perpetuated racial disparities. The R&D strategy must address these harms and include robust investments in research to detect, mitigate, and prevent algorithmic discrimination. This includes developing scalable tools for bias auditing, fairness testing, and inclusive design practices.
Promote Equity and Inclusion: AI should be a tool for expanding opportunity, not entrenching exclusion. The R&D strategy should support work that explores how AI can proactively identify and eliminate discriminatory practices and how it can be used to expand access to services for historically marginalized communities. For example, the National Fair Housing Alliance (NFHA) has emphasized the importance of an equity-centered auditing system for AI to assess its fairness when used in the areas of housing and credit.[4]
Ensure Accountability and Transparency: Transparency is essential for building public trust in AI. The Center for Civil Rights and Technology has called for clear guidelines on AI use in both public and private sectors.[5] The R&D strategy should mandate that AI systems developed in publicly-funded research provide explainable outputs and that individuals have mechanisms to challenge and seek redress for decisions made by AI. This includes requiring documentation of AI decisionmaking processes and establishing independent oversight mechanisms.
Reflect Democratic Values: To ensure AI development aligns with democratic values, the national strategy must invest in interdisciplinary research that brings together technologists, ethicists, legal scholars, and civil rights and civil society advocates. This research should explore the long-term societal impacts of AI, develop ethical guidelines, and create tools for responsible innovation. Funding should prioritize projects that center the needs of vulnerable populations and address systemic inequities. Investing in research to develop techniques for detecting and mitigating bias, such as creating diverse and representative datasets and implementing fairness checks, is vital.[6] Likewise, the National AI Research Resource (NAIRR) pilot[7] that brings together U.S. researchers and other stakeholders to support needed research should be supported.
Consider the Current Use of AI and Gaps Related to the Human Impact of Technology: Today, AI is widely used for many purposes, including predictive policing, credit scoring, hiring, and health care diagnostics.[8]These applications often lack sufficient safeguards to protect individuals from harm. For instance, facial recognition systems are less accurate in identifying people with darker skin tones, and health care algorithms underperform for underrepresented populations due to biased training data. These gaps highlight a critical need for R&D that focuses not only on technical performance but also on human impact, and in particular, on how AI affects our civil rights, access to opportunity, and social equity.
Follow Examples of AI R&D Addressing Bias and Harms: There are promising examples of AI R&D that have helped identify and mitigate bias. The National Institute of Standards and Technology (NIST) has developed the AI Risk Management Framework (AI RMF 1.0),[9] which outlines methods for identifying and managing systemic, computational, and human-cognitive biases in AI systems. In education, researchers have developed fairness auditing tools to detect bias in AI-powered plagiarism detection systems, which were found to disproportionately flag non-native English speakers. These efforts demonstrate the value of targeted R&D in uncovering and addressing algorithmic harms before they scale.
The 2025 National AI R&D Strategy presents a pivotal opportunity to shape the future of AI in a way that benefits all members of society. By centering civil rights, addressing algorithmic discrimination, promoting equity and inclusion, ensuring accountability and transparency, and investing in ethical AI research, we can build a future where AI serves the public good.
We urge the National Science Foundation and Office of Science and Technology Policy to adopt a strategy that reflects these priorities and look forward to continued collaboration to advance a just and equitable AI ecosystem. Should you require further information or have any questions regarding this issue, please feel free to contact Jonathan Walter, senior policy counsel, at [email protected].
Sincerely,
The Leadership Conference on Civil and Human Rights
National Fair Housing Alliance
[1] National Science Foundation, “Request for Information on the Development of a 2025 National Intelligence (AI) Research and Development (R&D) Strategy,” 90 FR 17835 (April 28, 2025). Federal Register :: Request for Information on the Development of a 2025 National Artificial Intelligence (AI) Research and Development (R&D) Strategic Plan, April 29, 2025.
[2] The Leadership Conference on Civil and Human Rights, Response to the Office of Science and Technology Policy’s (OSTP) Request for Information on National Priorities for Artificial Intelligence, Docket Number: OSTP-TECH-2023-0007 (July 7, 2023), Leadership-Conference-Comments-OSTP-National-AI-Strategy.pdf.
[3] Chiraag Bains, “The legal doctrine that will be key to preventing AI discrimination,” The Brookings Institute (Sept. 13, 2024), The legal doctrine that will be key to preventing AI discrimination.
[4] Michael Akinwumi, Lisa Rice, Snigdha Sharma, “Purpose, Process, and Monitoring: A New Framework for Auditing Algorithmic Bias in Housing & Lending,” National Fair Housing Alliance (NFHA) (Feb. 17, 2022), PPM_Framework_02_17_2022.pdf.
[5] The Leaderhsip Conference on Civil and Human Rights’ Center for Civil Rights and Technology, et.al., Response to the Office of Management and Budget’s (OMB) Reuqest for Information: Responsible Procurement of Artificial Intelligence in Government, FR Doc. 2024-06547 (April 29, 2024), The-Center-for-Civil-Rights-and-Technology-Comments-on-OMB-Procurement-Guidance.pdf.
[6] Kathy Heldman, “Understanding Algorithmic Discrimination: How Bias Persists in AI Systems,” Workplace Fairness (Jan. 20, 2025), Understanding Algorithmic Discrimination: How Bias Persists in AI Systems – Workplace Fairness.
[7] U.S. National Science Foundation, “National Artificial Intelligence Research Resource Pilot, National Science Foundation (Jan. 24, 2024), National Artificial Intelligence Research Resource Pilot | NSF – National Science Foundation.
[8] Office of the High Commissioner for Human Rights, “Racism and AI: ‘Bias from the past leads to bias in the future,” United Nations (July 30, 2024), Racism and AI: “Bias from the past leads to bias in the future” | OHCHR.
[9] National Institute of Standards and Technology, “Artificial Intelligence Risk Management Framework (AI RMF 1.0), U.S. Department of Commerce (Jan. 2023), AI Risk Management Framework | NIST.