"Software Testing and Analysis: Challenges and Opportunities in the Age of AI"

Cristian Cadar -
Digital Security

Date: -
Location: Eurecom

Abstract: Software development is undergoing significant transformations, with systems growing at an accelerated rate and increasingly incorporating AI-generated code. As a result, successful software testing and analysis techniques such as fuzzing and symbolic execution have to similarly adapt to remain useful: they must handle larger codebases, keep up with increasingly rapid development cycles, operate across a growing diversity of programming languages and system architectures, and infer and reason about higher-level semantic properties. In this talk, I argue that testing and analysis techniques can themselves take advantage of AI. I will discuss our recent research efforts in this direction, including our work on designing an agentic concolic executor, an agentic hybrid fuzzer, and an LLM-based technique for inferring the intent of software patches. These systems build on traditional analysis techniques but leverage large language models and agentic AI frameworks to extend their capabilities along the dimensions discussed above. I end the talk by reflecting on the comparative strengths and weaknesses of traditional and AI-based testing and analysis techniques and what this means for building trustworthy software in an era increasingly shaped by AI-generated code. Bio: Cristian Cadar is a Professor in the Department of Computing at Imperial College London, where he leads the Software Reliability Group (http://srg.doc.ic.ac.uk), working on automatic techniques for increasing the reliability and security of software systems. Prof. Cadar's research has been recognised by several prestigious awards, including the IEEE TCSE New Directions Award, Humboldt Research Award, BCS Roger Needham Award, EuroSys Jochen Liedtke Award, HVC Award, and two test of time awards. Many of the research techniques he co-authored have been open-sourced and used in both academia and industry. In particular, he is co-author and maintainer of the KLEE symbolic execution system, a popular system with a large user base. Prof Cadar has a PhD in Computer Science from Stanford University, and undergraduate and Master's degrees from the Massachusetts Institute of Technology.