Trustworthy AI
Our researchers design intelligent systems that are transparent, reliable, fair and aligned with human needs.
As AI systems become increasingly integrated into everyday life, the center develops methods that improve explainability, interpretability, fairness, robustness and safety while ensuring intelligent systems can be understood, trusted and used with confidence.
Research Areas:
- Explainable AI
- Interpretable machine learning
- Trustworthy AI
- Fairness and bias in AI systems
- Reliable machine learning
- Foundation models
- Artificial general intelligence
- Human-AI interaction
- Natural language explanations
- Language model transparency
- Safe and robust AI
Our Experts:
Soheil Feizi
Associate Professor
Machine Learning & Statistical Inference
Featured News:
UMD Framework Used to Test Safety of Meta’s New Multimodal AI Model
Furong Huang is co-leading efforts involving AI safety testing under simulated operational pressure.
Wiegreffe Receives Grant to Stress-Test the Future of AI Transparency
The $270K award from Coefficient Giving supports her efforts to strengthen AI safeguards.
UMD Researchers Part of Team Receiving $1.8M DARPA Award to Make AI More Trustworthy
Soheil Feizi and Furong Huang are collaborating to make AI-driven large language models more consistent, adaptable and secure in high-stakes environments.