Azure OpenAI Services Security
1h 8mIntermediate2024-08-19
Authors
Karl Ots
Cloud and Cybersecurity Expert, Azure MVP, Microsoft Regional Director
Course details
Explore generative AI security in the world of Azure OpenAI. First, review key generative AI concepts. Then, see demonstrations of securing Azure OpenAI services. Join Karl Ots as he discusses potential risks and shows appropriate security measures to take while working with these technologies. Learn how to apply industry-standard control frameworks, like the OWASP Top 10 for LLM applications. Find out how to evaluate and apply security controls and use Azure policies to secure Azure OpenAI. Plus, see how to implement access control, audit logging, network isolation, and encryption. Whether you are an experienced cloud architect or new to adopting AI development, this course is beneficial for anyone who wants to enhance their expertise in mitigating AI-related risks while maximizing the potential of Azure OpenAI services.
Learning objectives
Implement effective security measures for OpenAI applications in Azure.
Apply the Microsoft Cloud Security Benchmark and shared responsibility model to secure your generative AI adoption.
Follow the guidance of the OWASP Top 10 framework for LLMs.
Configure custom Azure policies for Azure OpenAI.
Integrate secure Azure OpenAI usage in organizational security frameworks.
Learning objectives
Implement effective security measures for OpenAI applications in Azure.
Apply the Microsoft Cloud Security Benchmark and shared responsibility model to secure your generative AI adoption.
Follow the guidance of the OWASP Top 10 framework for LLMs.
Configure custom Azure policies for Azure OpenAI.
Integrate secure Azure OpenAI usage in organizational security frameworks.
Skills covered
Azure OpenAICloud SecurityAPIsNetwork SecurityCloud AdministrationGenerative AIAzureCloud PlatformsCybersecurityArtificial Intelligence (AI)Cloud ComputingMicrosoftSoftware DevelopmentDeep Dive (X:Y)
Concepts
0. Introduction
- 01 - Securely benefit from GenAI
- 02 - GenAI terminology
1. Overview of Generative Artificial Intelligence Security
- 03 - Common use cases and risks for Gen AI in the enterprise
- 04 - Shared AI responsibility model
- 05 - Applicable control frameworks
- 06 - OWASP top 10 for LLM applications
2. Securing Azure OpenAI Service
- 07 - Comparing OpenAI ChatGPT Enterprise and Azure OpenAI
- 08 - Evaluating controls with Microsoft cloud security benchmark
- 09 - Applying Microsoft cloud security benchmark to Azure OpenAI
- 10 - Using Azure Policy to secure Azure OpenAI at scale
- 11 - Demo - Create custom policies for Azure OpenAI
3. Security Controls
- 12 - Access control
- 13 - Demo - Implement access control for Azure OpenAI
- 14 - Audit logging
- 15 - Demo - Implement audit logging for Azure OpenAI
- 16 - Network isolation
- 17 - Demo - Implement network isolation for Azure OpenAI
- 18 - Encryption at rest
- 19 - Demo - Implement encryption at rest for Azure OpenAI
- 20 - Content filtering
- 21 - Demo - Implement content filtering for Azure OpenAI
4. Moving to Production
- 22 - Azure Open AI in your cloud security architecture
- 23 - Data grounding - Bring your own data
- 24 - Demo - Bring your own models with Azure AI Marketplace
- 25 - Reference application architecture
Conclusion
- 26 - Next steps