Securing Generative AI
Keep adding new skills with 10,000+ programs for $239 (usually $399). Save now.
Recommended experience
Recommended experience
What you'll learn
Explore security for deploying and developing AI applications, RAG, agents, and other AI implementations
Learn hands-on with practical skills of real-life AI and machine learning cases
Incorporate security at every stage of AI development, deployment, and operation
Skills you'll gain
Details to know
See how employees at top companies are mastering in-demand skills
There is 1 module in this course
This course offers a comprehensive exploration into the crucial security measures necessary for the deployment and development of various AI implementations, including large language models (LLMs) and Retrieval-Augmented Generation (RAG). It addresses critical considerations and mitigations to reduce the overall risk in organizational AI system development processes. Experienced author and trainer Omar Santos emphasizes “secure by design” principles, focusing on security outcomes, radical transparency, and building organizational structures that prioritize security. You will be introduced to AI threats, LLM security, prompt injection, insecure output handling, and Red Team AI models. The course concludes by teaching you how to protect RAG implementations. You learn about orchestration libraries such as LangChain, LlamaIndex, and others, as well as securing vector databases, selecting embedding models, and more.
This module provides a comprehensive overview of generative AI security, covering threats and mitigation strategies for large language models and related systems. Topics include prompt injection, insecure output handling, training data poisoning, model denial of service, supply chain vulnerabilities, sensitive information disclosure, insecure plugin design, excessive agency, overreliance, model theft, red teaming, and securing Retrieval Augmented Generation (RAG) implementations. Learners gain practical knowledge of industry frameworks, best practices, and tools to safeguard AI technologies in production environments.
What's included
36 videos7 assignments
36 videos•Total 220 minutes
- Introduction•3 minutes
- Learning objectives•1 minute
- Understanding the Significance of LLMs in the AI Landscape•7 minutes
- Exploring the Resources for this Course - GitHub Repositories and Others•3 minutes
- Introducing Retrieval Augmented Generation (RAG)•12 minutes
- Understanding the OWASP Top-10 Risks for LLMs•6 minutes
- Exploring the MITRE ATLAS (Adversarial Threat Landscape for Artificial-Intelligence Systems) Framework•6 minutes
- Understanding the NIST Taxonomy and Terminology of Attacks and Mitigations•7 minutes
- Learning objectives•1 minute
- Defining Prompt Injection Attacks•12 minutes
- Exploring Real-life Prompt Injection Attacks•4 minutes
- Using ChatML for OpenAI API Calls to Indicate to the LLM the Source of Prompt Input•10 minutes
- Enforcing Privilege Control on LLM Access to Backend Systems•6 minutes
- Best Practices Around API Tokens for Plugins, Data Access, and Function-level Permissions•3 minutes
- Understanding Insecure Output Handling Attacks•3 minutes
- Using the OWASP ASVS to Protect Against Insecure Output Handling•5 minutes
- Learning objectives•1 minute
- Understanding Training Data Poisoning Attacks•4 minutes
- Exploring Model Denial of Service Attacks•3 minutes
- Understanding the Risks of the AI and ML Supply Chain•9 minutes
- Best Practices when Using Open-Source Models from Hugging Face and Other Sources•13 minutes
- Securing Amazon BedRock, SageMaker, Microsoft Azure AI Services, and Other Environments•16 minutes
- Learning objectives•1 minute
- Understanding Sensitive Information Disclosure•3 minutes
- Exploiting Insecure Plugin Design•3 minutes
- Avoiding Excessive Agency•4 minutes
- Learning objectives•1 minute
- Understanding Overreliance•5 minutes
- Exploring Model Theft Attacks•5 minutes
- Understanding Red Teaming of AI Models•14 minutes
- Learning objectives•1 minute
- Understanding the RAG, LangChain, Llama Index, and AI Orchestration•17 minutes
- Securing Embedding Models•10 minutes
- Securing Vector Databases•12 minutes
- Monitoring and Incident Response•8 minutes
- Securing Generative AI: Summary•2 minutes
7 assignments•Total 210 minutes
- Introduction to AI Threats and LLM Security Quiz•30 minutes
- Understanding Prompt Injection & Insecure Output Handling Quiz•30 minutes
- Training Data Poisoning, Model Denial of Service & Supply Chain Vulnerabilities Quiz•30 minutes
- Sensitive Information Disclosure, Insecure Plugin Design, and Excessive Agency Quiz•30 minutes
- Overreliance, Model Theft, and Red Teaming AI Models Quiz•30 minutes
- Protecting Retrieval Augmented Generation (RAG) Implementations Quiz•30 minutes
- End of Course Assessment •30 minutes
Explore more from Security
- Status: Free TrialE
Edureka
Course
- Status: Free TrialE
Edureka
Specialization
- Status: Free Trial
Course
Why people choose Coursera for their career
Frequently asked questions
Yes, you can preview the first video and view the syllabus before you enroll. You must purchase the course to access content not included in the preview.
If you decide to enroll in the course before the session start date, you will have access to all of the lecture videos and readings for the course. You’ll be able to submit assignments once the session starts.
Once you enroll and your session begins, you will have access to all videos and other resources, including reading items and the course discussion forum. You’ll be able to view and submit practice assessments, and complete required graded assignments to earn a grade and a Course Certificate.
More questions
Financial aid available,