AI meets Cybersecurity

#3 Security Testing Hands-On: testing the security of AI systems with open source tools

📅 April 23, 2026

🕐 6:30 PM

📍 Viale la Plaia, 15 (CA)

The journey into the world of security continues with the third meeting of the "AI meets Cybersecurity" series, the event dedicated to exploring operational challenges and solutions in the era of artificial intelligence. 🔄

🛠️ This hands-on session demonstrates how to translate risk categories into concrete tests using specialized and open source tools.

Through live demos with Garak (NVIDIA) and DeepTeam, the talk covers three key attack scenarios from the OWASP Top 10 for LLM Applications: prompt injection (LLM01), sensitive information disclosure (LLM06), and excessive agency (LLM08). It then moves on to AI agent security, with a focus on skill analysis and risk assessment related to MCP servers, tool access, and autonomous workflows, using the HikmaAI CLI to demonstrate an integrated approach to testing agents, models, and infrastructure. Each scenario is performed live with measurable and immediately replicable results.

A practical talk designed for security engineers, developers, and red teams who want to move from theory to operational AI security.

The event is organized by The Net Value and Innois.

Speaker

Mauro Medda is the CTO and Co-Founder of HikmaAI, a startup developing AI security and governance solutions. With nearly 20 years of experience in DevSecOps and enterprise architecture, he has held technical leadership roles in organizations such as Emirates Group, Yoox Net-a-Porter, and Zilliqa Research, managing teams of over 100 engineers. Today, he focuses on AI agent security, compliance frameworks, and adversarial testing of Artificial Intelligence systems. He is a supporter of the AI-first approach in software development and a speaker on AI security and governance topics.

Mauro Dore is an AI Security Engineer at HikmaAI. He holds a Master's degree in Artificial Intelligence Engineering from the University of Bologna and a Bachelor's degree in Computer Science from the University of Cagliari. Previously, he was a Data Scientist at Toyota Motor Europe in Brussels, where he developed an optimization engine that significantly improved carbon emission estimation processes, and a Full-Stack Developer at Accenture, working with Java and Spring Boot. His expertise spans Machine Learning, NLP, Computer Vision, and optimization, with solid practice in transformer models, reinforcement learning, and deep learning architectures. He was selected from over 800 candidates for the Airbus Digital Challenge 2024. He is fluent in Italian, English, and Spanish.

Contact Us

📧 For info: eventi@thenetvalue.com
📞 070 2330200