You might have heard about problems that arise when AI systems misinterpret data or propose solutions that reflect human prejudice. In this course, you’ll learn about the five pillars of AI ethics: fairness, robustness, explainability, transparency, and privacy. Through real-world examples you’ll learn about AI ethics, how they are implemented, and why AI ethics are so important in building trustworthy AI systems.
What you’ll learn
After completing this course, you should be able to:
- Identify the five pillars of AI ethics
- Describe fairness in AI
- Describe protected attributes
- Identify privileged groups and unprivileged groups
- Explain AI bias
- Identify robustness
- Describe adversarial robustness within AI
- Explain how an adversary can influence an AI system
- Identify adversarial attacks
- Describe explainability
- Compare interpretability and explainability
- Define transparency
- Describe governance
- Identify the business roles and the aspects of transparency they are involved in
- Identify personal information
- Identify sensitive personal information
- Recognize model anonymization
- Describe differential privacy
- Explain data minimization
