AI and Society

AI systems can influence education, employment, finance, health, safety, and access to services, so their effects must be considered responsibly. AI ethics promotes fairness, privacy, transparency, accountability, safety, and respect for human rights. These principles help identify and reduce bias and harm. Options B, C, and D describe unethical purposes rather than the reason ethics is needed, so A is correct.

Human-centered artificial intelligence is designed and used with human needs, safety, rights, dignity, and well-being as important considerations. Therefore, option A is correct. Focusing only on machine efficiency or profit does not adequately consider how an AI system affects people.

Transparency can help users, experts, and auditors understand the information, factors, or procedures that influenced an AI decision. This supports review, accountability, and informed use. However, transparency alone does not guarantee correctness or eliminate the need to test for bias and protect privacy.

Explainability is the property that allows people to understand how or why an AI system reached a prediction or decision. For example, a system may identify the features that influenced a loan recommendation. Explainability does not mean that decisions are hidden, that a model cannot operate, or that data lack a source. Thus option A is correct and supports trust, review, and responsible use.

A sound privacy practice is to obtain informed, appropriate consent and limit personal-data use to a clearly stated, necessary purpose. Good practice may also include access controls, secure storage, and deletion when data are no longer needed. Publishing private information, sharing passwords, or secretly selling personal data exposes people to misuse and security harm, so option A is the only responsible choice.

Accountability in AI means that identifiable people, teams, or organizations remain responsible for a system’s decisions, performance, and effects. It includes documenting choices, monitoring outcomes, addressing harms, and providing oversight or correction when needed. Hiding mistakes, blaming users automatically, and abandoning supervision avoid responsibility rather than showing it. Therefore, option A correctly expresses accountability.

AI systems can influence important areas such as recruitment, lending, education, and healthcare. Fairness aims to prevent unjust discrimination based on caste, gender, or other personal characteristics. Therefore, option A is correct; the other options describe harmful or unrelated practices.

Personal data may reveal a person’s identity, preferences, location, or activities. Using it without informed consent can violate privacy and reduce the person’s control over its use. Data use alone does not automatically make an AI system fair, secure, or accurate.

Artificial intelligence can automate or change some repetitive tasks, which may alter the nature of certain jobs. Workers may need skills such as data analysis, communication, AI operation, and technical supervision. This does not mean that every job ends immediately.

Fairness is checked to determine whether an AI system treats people from different groups equitably and avoids unjust discrimination based on factors such as gender or ethnicity. Therefore, option A is correct. Option B is the opposite of fairness because responsible AI aims to reduce bias, not encourage it. Exam tip: In AI ethics, remember that fairness means avoiding unjustified differences in outcomes or treatment between groups.

In high-impact areas such as education, healthcare, employment, or lending, AI decisions can affect a person's rights, opportunities, and well-being. It is therefore important to understand the reasons and data behind a decision so that errors, bias, and accountability can be examined. Options B, C, and D describe relatively low-risk tasks, where detailed explanation of a decision is generally less critical.

Accountability means clearly identifying the person or organization responsible for an AI system’s decisions and impacts, so that it can explain and answer for its actions. Hiding errors is the opposite of accountability, while removing oversight increases risk. Exam tip: remember accountability as ‘assigned responsibility with an obligation to answer.’

Errors in sensitive or high-risk decisions can seriously affect people’s rights, safety, or opportunities. Human supervision helps review a decision in context, identify bias or errors, and change the decision when needed. In contrast, a low-risk task such as changing a screen colour has a very limited impact if the result is wrong.