Generative AI Privacy & Security: OWASP & NIST for Devs

About this course

“This course contains the use of artificial intelligence.”

Build GenAI applications that can demonstrate how sensitive information is controlled—not merely promise that it is.

Generative AI introduces new data flows across prompts, models, RAG systems, vector stores, tools, APIs, caches, and telemetry. This developer-focused course converts privacy and AI-risk guidance into technical architecture, implementation controls, tests, and review evidence.

Who this course is for:

• Software developers integrating LLM APIsBackend and API engineers building AI functionalityAI/ML engineers deploying RAG and GenAI systems
• Technical leads responsible for AI architecture
• Application-security engineers supporting AI development

What you will learn:

• Threat-model GenAI applications using current OWASP LLM/GenAI security guidance
• Identify sensitive-data flows and trust boundaries
• Implement data minimization, secret detection, and PII-handling controls
• Design privacy-aware LLMOps pipelines and telemetry
• Design safer RAG and vector-store authorization boundaries
• Reduce cross-tenant information-disclosure risks
• Apply least privilege to model-connected tools
• Map engineering controls to NIST AI RMF and its Generative AI Profile
• Build synthetic privacy and security regression tests
• Produce auditable release evidence and a residual-risk register

Requirements:

Basic programming knowledge, familiarity with APIs, and the ability to read Python and JSON. Previous security, privacy-law, or compliance expertise is not required. Exercises use synthetic data; do not use real confidential or customer information.

Final project:

• Design and implement a privacy-aware GenAI support assistant. You will create its data-flow diagram, threat mod

Prices, discounts and availability are set by Udemy. We may earn a commission when you purchase through links on this site.