Plenary Panel I - QRS 2025

QRS 2025 Plenary Panel I

Rethinking Software Quality, Reliability, and Security in the Era of AI and LLMs

Opening Lines

AI and Large Language Models (LLMs) are rapidly transforming the software engineering landscape— from assisting developers with real-time code suggestions to autonomously detecting bugs and vulnerabilities. These technologies promise to automate labor-intensive tasks, enhance test coverage, and boost productivity. But their adoption also raises critical concerns: hallucinated code, hidden biases, unclear provenance, and emergent security risks. As AI becomes both a co-pilot and a code contributor, how do we harness its potential to improve software quality, reliability, and security— without introducing new fault lines?

Panel Scope

This panel brings together experts across research and practice to explore:

The positive impact of AI/LLMs in enhancing automation for testing, verification, and secure development.
How AI is reshaping software quality pipelines, from code review to continuous integration and fuzz testing.
Risks introduced by AI-generated or AI-assisted code— including reliability, explainability, and trust concerns.
Techniques to assess and improve the robustness, traceability, and safety of AI-influenced systems.
Bridging gaps between traditional SE techniques (e.g., static analysis, formal verification) and AI-native methods.
The role of benchmarks, datasets, and evaluation frameworks for trustworthy AI-enhanced software engineering.

Audience Engagement Hook

We will explore the full spectrum— from promise to pitfalls— with timely questions such as:

How far can we push AI in automating quality and security assurance?
Can LLMs be trusted to generate production- grade code, or are they best seen as advisory tools?
What safeguards are needed when one AI writes code and another AI tests or deploys it?
How do we balance efficiency gains with reliability guarantees in AI-driven workflows?
Where should we draw the line between human oversight and machine autonomy?