AI Ethics

Correct answer: A. The output may contain errors or bias

Explanation: AI outputs can be inaccurate, incomplete, outdated, or biased because the system may have limitations in its data, design, or interpretation. Checking important results helps a person identify problems, compare reliable evidence, and correct or reject the output. Human judgment remains important, especially for consequential decisions.

Correct answer: D. When data is one-sided

Explanation: One-sided or unrepresentative data may leave some groups, situations, or experiences missing. If an AI system learns from that incomplete picture, its predictions can systematically favour some cases and disadvantage others. Balanced data, clear goals, and human review can help detect or reduce bias, although they do not guarantee perfection.

Correct answer: D. Protecting people from harm

Explanation: AI safety aims to prevent or reduce harm that could result from incorrect, unsafe, misused, or unexpected system behaviour. Safety may involve testing, safeguards, monitoring, privacy protection, and human oversight. It does not mean blocking all data access, removing people from supervision, or merely making the interface look attractive.

Correct answer: B. The basis of a decision can be understood

Explanation: Transparency makes the working or reasoning of an AI system understandable to users. They can examine which information, rules, or patterns influenced a result. This supports trust, accountability, and correction of mistakes. It does not hide errors, automatically remove privacy, or delete data; those are separate issues.

Correct answer: C. To identify errors, bias, and potential risks

Explanation: AI systems can produce inaccurate, biased, or unsafe results because their data, design, or interpretation may be imperfect. Human supervision allows people to review outputs, notice risks, protect affected users, and correct decisions when needed. Keeping a machine switched off, sharing private information, or ignoring results is not the purpose of supervision.

Correct answer: A. Improper and harmful use

Explanation: Misinformation can confuse people, damage reputations, influence decisions, and sometimes create safety risks. Using AI to produce or spread it is therefore improper and harmful, especially when the creator knows the information is false or fails to check it. Responsible use requires accuracy checks, honest presentation, and consideration of possible effects on others.

Correct answer: A. Correct, safe, and fair use

Explanation: AI ethics concerns how AI should be designed and used so that it respects people and avoids unnecessary harm. Important concerns include fairness, privacy, safety, transparency, accountability, and responsible use. The colour of a machine, wall length, or paper thickness may be physical details, but they do not define the ethical quality of an AI system.

Correct answer: C. Using technology respectfully and responsibly

Explanation: Good digital citizenship means using digital tools safely, respectfully, and responsibly while considering other people’s rights. It includes protecting privacy, checking information, obtaining consent, and communicating appropriately. Stealing data, spreading misinformation, or sharing someone’s images without permission violates privacy or trust, so option C is the only positive example.

Correct answer: B. Asking questions and using examples to understand concepts

Explanation: AI is best learned through curiosity, questions, examples, and practice. Asking for clarification and connecting ideas with real situations builds conceptual understanding and helps a student evaluate results thoughtfully. Memorising without understanding is limited, accepting unchecked information is unreliable, and avoiding difficult problems prevents improvement. Thus, option B describes the strongest learning behaviour.