PI: Guy Zyskind
I’m recruiting PhD students to join my lab at the University of Miami in Spring or Fall 2027.
The web is moving toward AI agents interacting with other agents, making decisions and taking actions on our behalf. Our lab studies how these systems behave, how they fail, and what it takes to trust them with real responsibility.
We are working in two broad directions:
- Multi-agent AI safety, trust, and security. Studying agents that work together under one owner or interact on behalf of different people and organizations. We want to understand when these interactions produce harmful or unintended behavior, how to evaluate safety across the whole system, and how to detect and reduce risks while preserving agents’ capabilities.
- Privacy-preserving AI, especially in healthcare and finance. Developing ways to run AI on sensitive data without exposing it. Examples include private inference and retrieval, and helping hospitals or financial institutions use information across organizations while keeping patient and customer records confidential.
I’m looking for students who love research and want to make a lasting impact, whether through research work that shapes this new web or companies that change the world. You should have a record of pursuing ideas independently and want to help shape what we work on.
You should have a strong foundation in at least one of:
- AI safety or security - Data science, computational social science, or network science - Cryptography or computer security
Full funding is available for strong students to pursue a PhD at the University of Miami, a leading U.S. R1 research university. Our lab collaborates on AI safety research with researchers at MIT, Stanford, and other leading institutions.
Please email guyzyskind@miami.edu with your CV and a brief description of your research experience and interests.