AI Product Development: Secure by Design

Discover a holistic approach to building AI systems that are secure by design, covering security controls at every layer of the AI lifecycle. From data security and governance to infrastructure protection, instructor Reet Kaur shares practical skills in adversarial threat detection, secure model access, and continuous monitoring. This course also covers secure CI/CD pipelines for AI, focusing on ongoing validation and proactive threat mitigation. Along the way, gather insights on regulatory compliance, governance, and structured methods for assessing and managing AI-specific risks.

Learning objectives
Leverage a holistic AI security framework by breaking down AI systems into distinct domains—data, AI models, application, infrastructure—and identifying key security controls across each layer.
Integrate secure-by-design principles across the AI development lifecycle, implementing defense-in-depth strategies tailored to AI’s unique vulnerabilities.
Identify and mitigate threats to AI models by employing adversarial machine learning defenses, model behavior monitoring, and security testing.
Establish secure and resilient AI deployment pipelines incorporating CI/CD practices, threat detection, and logging for continuous security monitoring.
Navigate regulatory and compliance frameworks critical to AI security, addressing standards for data privacy, integrity, confidentiality, and governance to meet industry and legal requirements.
Assess and manage AI system risks by determining system maturity levels, identifying required security controls, implementing tailored defenses, and conducting continuous monitoring.