Panel 3: Benchmarking & Auditing AI Systems in High-Risk Domains

  • Cross-sector evaluation regimes for high-risk AI systems in credit, employment, health, and public services, with legal tools as one concrete case, not the centerpiece.
  • How forthcoming EU AI Act requirements and national rules will expect structured risk management, logging, documentation, and post-market monitoring, and where current industry practices are falling short.
  • The emerging ecosystem of independent audits, red-teaming, safety evaluations, and civil-society scorecards, and how they interact with internal compliance and external regulators.
  • Methodological challenges: robustness and fairness metrics, domain shift, “unknown unknowns,” and how to align technical evaluation with legal and ethical obligations.
LOCATION: Paul Brest Hall, Munger Graduate Residence
DATE: April 15, 2026
TIME: 3:25 pm - 4:30 pm
Kyle KloeppelShayne LongpreSonia M. Gipson Rankin
Back to top