Skip to content
Sign in
Artwork for The Quantum Feedback Loop
The Quantum Feedback Loop · Yesterday · 1 hr 5 min

Dr. Joanna Santos on LLMs and Software Vulnerabilities

In this episode of The Quantum Feedback Loop podcast, we discuss software security issues and testing with Dr. Joanna Santos, Assistant Professor of Computer Science and Engineering at the University of Notre Dame. Joanna's research focuses on understanding vulnerabilities in software systems and developing automated techniques to make them more secure. She leads the Security and Software Engineering Research Lab at Notre Dame and received a 2024 Google Research Scholar Award for a project using large language models (LLMs) to generate security tests for mobile applications. Joanna discusses her path into software engineering and security, including how she became interested in the ways that vulnerabilities can originate in poor system design. We also discuss her research on SecurityEval, a benchmark developed to evaluate security vulnerabilities in Al-generated code, the challenges involved in developing and evaluating tools designed to make software more secure, and the risks that students face with over-reliance on LLMs for software coding and evaluation.

0:00-1:05:39
Play in

transcript

No transcript — this publisher did not publish one.

show notes

In this episode of The Quantum Feedback Loop podcast, we discuss software security issues and testing with Dr. Joanna Santos, Assistant Professor of Computer Science and Engineering at the University of Notre Dame. Joanna's research focuses on understanding vulnerabilities in software systems and developing automated techniques to make them more secure. She leads the Security and Software Engineering Research Lab at Notre Dame and received a 2024 Google Research Scholar Award for a project using large language models (LLMs) to generate security tests for mobile applications.

Joanna discusses her path into software engineering and security, including how she became interested in the ways that vulnerabilities can originate in poor system design. We also discuss her research on SecurityEval, a benchmark developed to evaluate security vulnerabilities in Al-generated code, the challenges involved in developing and evaluating tools designed to make software more secure, and the risks that students face with over-reliance on LLMs for software coding and evaluation.

more episodes

All episodes