Dependable AI Systems (DAIS)

The DAIS Group investigates the scientific foundations of dependable and trustworthy AI.

Our vision is to transform foundation models from black-box predictors into systems whose behavior can be understood, validated, and trusted.

Misson

Building interpretable, grounded, evidence-based, and verifiable AI systems.

Research Areas

• Trustworthy AI
• Explainable AI
• AI Safety
• Foundation Models
• Semantic Representations
• AI Verification

Current Member

• Nazareno Garagiola

Collaborations

We actively seek collaboration with academia and industry.