Mathematica Shares Insights in Response to Senator Cassidy’s White Paper on Artificial Intelligence

Mathematica provided feedback on a white paper released by U.S. Senator Bill Cassidy (R-LA), ranking member of the Senate Health, Education, Labor, and Pensions (HELP) Committee, on artificial intelligence (AI) and the technology’s potential benefits and risks to society. Mathematica identified several opportunities for congressional and administrative involvement in AI frameworks, including:

• Operationalizing principles and practices for identifying and mitigating risks from AI
• Downstream audits of AI systems and upstream risk management
• Frameworks for rigorous evaluation of AI to determine effectiveness
• Investing in AI alignment research with the aim of ensuring safe solutions are built into future artificial general intelligence (AGI)
• Increased collaboration between policy researchers, policymakers, and the technology sector
• Risk and needs assessments evaluated by independent third parties and tested with the populations AI systems are intended to serve to increase equity