Human and Organizational Factors in AI Risk Management

On June 11, 20, 26 and July 2, 2024, the National Academies of Sciences, Engineering, and Medicine held a series of workshop sessions to consider human and organizational factors of artificial intelligence (AI) risk management. The first three sessions - which examined community participation, testing and evaluation, and organizational culture - were held virtually, and the final session - in which members of the planning committee and a set of invited respondents reflected on the earlier sessions - was held in person. This proceedings recounts the presentations and discussions that occurred throughout these workshop sessions.

Table of Contents

Front Matter 1 Widening Participation in the Design, Development, and Deployment of AI Tools 2 Evaluation, Testing, and Oversight 3 Safety in Context 4 Designing Paths Forward Appendix A: Statement of Task Appendix B: Workshop Session Agendas