Responsible AI Principles
Exam objective: Describe principles of responsible AI
This topic is always tested on AI-901. Know all six principles, what each means, and be able to identify which principle applies to a given scenario.
Overview​
Microsoft's Responsible AI framework defines six principles that guide how AI systems should be designed and used. These principles are not just philosophical — Azure AI services are built to support them, and the exam will present scenario-based questions asking you to identify which principle is at stake.
Key Concepts​
| Principle | Core idea | Example scenario |
|---|---|---|
| Fairness | AI should treat all people equitably, without bias based on race, gender, or other characteristics | A loan approval model that approves loans at different rates for different demographic groups |
| Reliability & Safety | AI should perform consistently and safely, even in unexpected situations | An autonomous vehicle AI that behaves unpredictably in fog |
| Privacy & Security | AI should protect personal data and be secure from attacks | A facial recognition system that stores biometric data without user consent |
| Inclusiveness | AI should empower everyone, including people with disabilities | A voice assistant that only understands certain accents |
| Transparency | AI systems should be understandable — users should know when they're interacting with AI and how decisions are made | A medical diagnosis tool that provides no explanation for its recommendations |
| Accountability | People and organizations must be accountable for how AI systems are designed and used | A company that deploys AI without monitoring for errors or harm |
Azure Tools for Responsible AI​
| Tool | How it supports Responsible AI |
|---|---|
| Azure AI Content Safety | Detects harmful content — supports Safety and Inclusiveness |
| Azure Machine Learning fairness tools | Detect and mitigate bias in models — supports Fairness |
| Azure AI Foundry evaluation features | Assess model outputs for safety and quality — supports Transparency |
| Microsoft Purview | Data governance and compliance — supports Privacy & Security |