Resource Automation + AI Mitigating Harm + Bias

The Privacy-Bias Trade-Off

This policy brief explores how federal privacy laws like the Privacy Act of 1974 limit demographic data collection, undermining government efforts to conduct equity assessments and address algorithmic bias.

The brief examines the tension between data minimization principles and the need for demographic data to assess disparities in public programs, particularly in response to Executive Order 13985.

Drawing on case studies and agency responses, the authors highlight legal, technical, and bureaucratic barriers to collecting race and ethnicity data. The paper offers actionable recommendations, including amending the Privacy Act, streamlining data collection processes, and adopting institutional safeguards to enable equity assessments without compromising privacy.