Institution: Simula Cybersecurity, Oslo, Norway
**Project Description**
In an era where AI increasingly underpins critical infrastructure and decision-making across society, we face an urgent challenge: how do we ensure these powerful systems are secure against sophisticated threats, function reliably, operate fairly, and ultimately deserve societal trust? This advanced research addresses this fundamental question by developing novel methods, frameworks, and evaluation techniques for Trustworthy AI Security. You will investigate how to build AI systems that are not only robust against attack but are also developed and deployed responsibly.
**What You Will Investigate**
- AI Model Robustness & Security: Design, build, and evaluate AI/ML systems resilient to specific threats like adversarial attacks (evasion), data poisoning, and prompt injection.
- AI for Cyber Defense Assurance: Investigate AI as a force multiplier for security (e.g., threat detection/response), critical...