Trustworthy AI Lab

When should we trust AI predictions, and when should we stop them? 

The Trustworthy AI Lab studies the reliability of artificial intelligence systems in real-world environments.

We develop machine learning methods that enable AI systems to recognize uncertainty, detect unreliable predictions, and support safe and reliable decision-making.

Our research focuses on AI reliability and decision reliability - understanding when AI predictions should be trusted, deferred, or rejected, particularly in domains where incorrect decisions can have serious consequences, including healthcare, finance, and physical AI systems.

By integrating machine learning, statistical modeling, and uncertainty quantification, we aim to build AI systems that move beyond predictive accuracy and support trustworthy decision-making in real-world environments.

                                                                                                                                old