Research Interests
My research develops and evaluates AI systems that are safe, trustworthy, and effective when deployed in high-stakes domains such as healthcare. I also gather empirical evidence to support policymakers and other non-technical stakeholders in making informed decisions that reduce algorithmic harm.
Application Areas
- Clinical decision support and healthcare delivery
- Public health systems and policy
- Mental health and wellbeing technologies
Responsible AI
Governance, evaluation, and accountability in deployed AI systems.
Safety, Security & Robustness
Failure modes, stress testing, and adversarial evaluation of machine learning models.
Natural Language Processing
Modeling and evaluation of large language systems.
Affective Computing
Understanding emotion, wellbeing, and human experience from data.
Recent Updates
SEP 2025 Invited talk: AI Safety in Suicide Contexts: Ethical and Technical Considerations, 100 Year Anniversary Lecture in the Orlando Lecture Series at Gannon University
JUL 2025 I spoke with the New York Times about the risks and safety issues
Home 