Abstract

Artificial Intelligence (AI) is a transformative technology with growing applications in medicine, agriculture, manufacturing, transportation, defense, and many other areas. It also holds substantial promise for improving government operations. Public oversight has focused on ensuring AI is responsible, equitable, traceable, reliable, and governable, but AI systems pose unique challenges because their inputs and operations are not always visible. The US Government Accountability Office (GAO) developed an AI accountability framework to help ensure accountable and responsible AI use by government agencies and other entities involved in the design, development, deployment, and continuous monitoring of AI systems. The framework identifies key accountability practices centered on the principles of governance, data, performance, and monitoring. To develop these practices, GAO held a forum on AI oversight with experts from government, industry, and non-profit organizations. We also conducted an extensive literature review and obtained independent validation from program officials and subject matter experts. In addition, GAO interviewed AI subject matter experts representing industry, state audit associations, non-profit entities, federal agencies, offices of inspector general, and other organizations.