News
- PaperThe Doctor Will (Still) See You Now: On the Structural Limits of Agentic AI in Healthcare
Accepted to FAccT 2026.
- PaperExpert Evaluation and the Limits of Human Feedback in Mental Health AI Safety Testing
Accepted to FAccT 2026.
- PaperThe Measurement Imbalance in Agentic AI Evaluation Undermines Industry Productivity Claims
Accepted to NeurIPS 2025 by our reviewers and the area chair, but REJECTED by NeurIPS.
- Talk
Had the wonderful opportunity to moderate a panel bringing together leaders from Google, EveryoneAI, and IST to discuss global AI safety and policy.
- PaperAn Adaptive Responsible AI Governance Framework for Decentralized Organizations
Accepted to AIES 2025.
- PaperLeRAAT: LLM-Enabled Real-Time Aviation Advisory Tool
Accepted to ECAI Demonstrations Track 2025.
- PaperResponsible AI in the Global Context
Accepted to FAccT 2025.
- Paper
- Talk
Honored to be selected to give a talk at the upcoming Women in Data-Driven Discovery (WiD3) Conference at Stanford University.
- PaperMore than Marketing? On the Information Value of AI Benchmarks for Practitioners
Accepted to IUI 2025.
- General
ICLR 2025 workshop on Human-AI Coevolution is now open for submissions.
- Award
Our grant proposal for the Schmidt Sciences, Safety Science program has been approved.
- General
Our workshop proposal on Human-AI Coevolution is accepted in ICLR 2025.