Biography will be updated soon.

Maryam Mouzarani is an Application Security and AI Red Team Engineer with a Ph.D. in Computer Engineering and 10 years of experience in cybersecurity. She currently works with Applause conducting AI and web API security assessments. With experience as an Assistant Professor, she bridges academic research and industry practice to deliver practical security solutions for real-world systems. She is a member of the OWASP AI Exchange and the founder of PwnzzAI, a practical and engaging learning environment for exploring AI security threats. Reference: PwnzzAI: A Web-Based Platform for Hands-On Exploration of LLM Vulnerabilities and AI Threats The increasing deployment of large language models (LLMs) in modern software systems introduces novel security risks, including prompt injection, jailbreak attacks, data leakage, and unsafe tool invocation. Despite growing awareness, there is a lack of practical environments for systematically studying these vulnerabilities. This presentation introduces PwnzzAI, an open-source web-based learning platform designed for hands-on exploration of AI-specific threats through interactive experimentation. The platform provides modular scenarios that simulate real-world LLM deployments, enabling users to analyze and exploit vulnerabilities in controlled settings. PwnzzAI is now partnered with OWASP AI Exchange, aligning its learning scenarios with recognized AI threat categories and emerging risk frameworks. This ensures that the platform reflects real-world, community-driven security concerns. The system emphasizes reproducibility and technical depth, allowing users to evaluate model behavior under adversarial conditions and assess mitigation strategies. By bridging theoretical AI security concepts with practical experimentation, PwnzzAI provides a structured framework for analyzing LLM vulnerabilities and advancing secure AI system development.