Show understanding of Artificial Intelligence (AI)
7.1 Ethics and Ownership
What is Ethics?
Ethics is the study of what is right and wrong. Think of it as the rule‑book for how we should treat each other and the world. In computing, ethics helps us decide if a piece of software is fair, safe, and respectful of people’s rights. 🤝
Why Ethics Matters in AI
AI systems learn from data. If the data is biased, the AI can make unfair decisions. Imagine a robot that only knows how to bake chocolate cookies because all the pictures it saw were chocolate. When asked to bake a vanilla cookie, it might refuse or make a bad attempt. This shows how bias can creep into AI. 🍪
- Safety: Preventing harm to users.
- Fairness: Treating all people equally.
- Transparency: Explaining how decisions are made.
- Privacy: Protecting personal data.
Ownership in AI
Ownership questions arise at three levels:
- Data ownership: Who owns the images, text, or sensor data used to train the model?
- Model ownership: Who owns the algorithm or the trained weights? Is it the developer, the company, or the user?
- Output ownership: If the AI writes a poem or designs a logo, who owns that creative work?
Think of a recipe: the ingredients (data) belong to the farmer, the chef (developer) owns the cooking method (model), and the finished dish (output) can be claimed by the restaurant or the chef depending on the contract. 🍽️
Key Ethical Principles (Summarised)
| Principle | What It Means |
|---|---|
| Fairness | Avoid bias that disadvantages any group. |
| Transparency | Explain how the AI makes decisions. |
| Privacy | Protect personal data from misuse. |
| Accountability | Ensure someone can be held responsible for AI actions. |
Case Study: Facial Recognition
A city installs facial‑recognition cameras to catch criminals. However, the system was trained mostly on light‑skinned faces, leading to higher false positives for people with darker skin. This raises ethical concerns about fairness and privacy. 🚓
Discussion question: How could the city improve the system to be fairer and more respectful of privacy?
Examination Tips
Remember: Cambridge questions often ask you to explain an ethical principle, give an example, and evaluate its impact on ownership. Use the PEEL structure: Point, Example, Explanation, Link back to the question. 📚
- Define key terms (e.g., bias, fairness, ownership).
- Use real‑world examples (facial recognition, autonomous cars).
- Show awareness of both positive and negative consequences.
- Keep answers concise but thorough.
Revision
Log in to practice.