By combining security-by-design, privacy-by-design, and resilience engineering principles, this talk aims to establish a holistic foundation for developing AI systems that are secure, privacy-preserving, explainable, and dependable. The talk introduces frameworks that contribute toward the realization of trustworthy AI capable of operating reliably in adversarial and dynamic environments while maintaining user privacy and regulatory compliance.
Key Takeaways
- Security & Privacy Threat Landscape: Understand the key security and privacy threats affecting modern AI systems, including adversarial attacks, data poisoning, model theft, and privacy leakage.
- Robustness Frameworks: Learn how combining security-by-design, privacy-by-design, and resilience engineering establishes a dependable AI foundation.
- Trustworthy AI in Practice: Realize trustworthy AI capable of operating reliably in adversarial and dynamic environments while maintaining compliance.
Target Audience
- AI developers and software architects designing secure, privacy-preserving AI systems and infrastructure.
Prerequisites
- Basic understanding of AI in applications.
