​How to Secure AI Agents?

​Security Challenges, Best Practices & Real-World Solutions

​AI agents are becoming increasingly powerful, automating workflows, accessing sensitive data, and interacting with external tools and systems. But how do we ensure they remain secure, reliable, and under control?

​Join us for an engaging session exploring the security challenges of AI agents, practical defense strategies, and best practices for building and deploying trustworthy AI systems.

​Topics We’ll Explore

• ​AI Agent Threat Landscape – Understanding the unique security risks of autonomous AI systems.

• ​Prompt Injection & Jailbreaking – Preventing malicious instructions from manipulating AI agents.

• ​Identity & Access Management – Applying least-privilege access and securing agent permissions.

• ​Tool & API Security – Protecting the tools, APIs, and external systems AI agents interact with.

• ​Data Privacy & Secrets Management – Safeguarding sensitive information, credentials, and confidential data.

• ​Human Oversight & Guardrails – Keeping AI agents accountable through monitoring, approval workflows, and controlled autonomy.

• ​Real-World Attacks & Defenses – Exploring practical examples, common vulnerabilities, and effective mitigation strategies.

​Event Agenda

​18:00–19:00 | Welcome, Food & Drinks Arrive, grab some food and drinks, and connect with fellow AI enthusiasts, developers, and cybersecurity professionals.

​19:00–20:00 | Talk: How to Secure AI Agents? 🔐 Explore key security challenges, real-world attack scenarios, and practical strategies for building safer AI agents.

​20:00–21:00 | Q&A and Networking 💬🤝 Ask questions, share experiences, exchange ideas with the speaker and other participants, and continue the conversation in an informal setting.

​Who Should Attend?

​AI developers, cybersecurity professionals, researchers, technology leaders, and anyone interested in the future of secure and trustworthy AI.