PhD Studentship: Empirical Security Assessment of AI Decision Engines in Cyber-Physical Systems

Newcastle University is offering a PhD studentship within the School of Computing to address the critical security gap in AI-driven cyber-physical systems (CPS). The research focuses on creating a systematic, reproducible, and comparative empirical framework to assess the robustness and failure modes of AI decision engines in environments like autonomous vehicles and industrial control systems. The successful candidate will move beyond theoretical attack models to provide empirical evidence that supports regulators and security engineers in deploying trustworthy AI. This project is highly relevant to industries such as manufacturing, energy, and transportation, where safety-critical decisions are increasingly automated.

• Develop a rigorous empirical framework for evaluating the security robustness of AI decision engines.
• Conduct systematic and comparative assessments of AI failure modes in cyber-physical environments.
• Analyze operational risks associated with AI deployment in safety-critical sectors.
• Produce reproducible research findings that inform industrial best practices for trustworthy AI.
• Collaborate with academic and industrial stakeholders to address real-world deployment challenges.