ecuring Generative AI provides a practical, end-to-end guide to building and deploying AI systems—LLMs, RAG pipelines, agents, and more—with security at the forefront. This course teaches you how to integrate protection, governance, and risk mitigation into every stage of the AI lifecycle, ensuring that your applications remain resilient against modern threats.
 
What You’ll Learn
Apply essential security principles to AI development, deployment, and operations.
Understand real-world AI and machine learning attack scenarios through hands-on exercises.
Secure LLM-powered systems against threats such as prompt injection, insecure output handling, model manipulation, and data leakage.
Identify and mitigate risks in RAG architectures, including vulnerabilities in embeddings, vector databases, and orchestration frameworks.
Strengthen AI agents and automated workflows by applying “secure by design” practices.
Explore threat modeling, red-teaming strategies, and organizational controls that reduce risk across AI programs.
 
Course Description
This comprehensive course walks you through the foundational and advanced security measures required for modern generative AI systems. You’ll examine the risks associated with large language models and their supporting infrastructure, covering topics such as prompt injection, prompt leaking, output sanitization, insecure integrations, and adversarial model interactions.
 
You’ll also learn how to safeguard RAG systems by securing vector stores, choosing embedding models responsibly, and implementing protections around orchestration libraries such as LangChain, LlamaIndex, and similar frameworks. Real-world case studies and hands-on labs help you build practical skills that translate directly into production environments.
 
Guided by expert instructor Omar Santos, the course emphasizes outcome-driven security, transparency, and organizational readiness—ensuring you know how to design and maintain AI systems that are safe, robust, and trustworthy.