What does trustworthy AI entail?

• A. AI that is reliable, safe, fair, transparent, and accountable, with human oversight and proper governance.
• B. AI that is fully autonomous with no oversight.
• C. AI that is unreliable and opaque.
• D. AI that ignores privacy.

Answer: (A)

Trustworthy AI means building systems that are reliable and safe, perform as intended, and do so fairly and transparently, with clear accountability all within a framework of human oversight and governance. Reliability and safety ensure the system behaves correctly across real-world conditions and minimizes harm. Fairness focuses on reducing bias and ensuring equitable outcomes for users. Transparency involves revealing how decisions are made and what data and models are used, so stakeholders can understand and trust the process. Accountability means there are clear responsibilities and auditability for actions and outcomes, so someone can take responsibility if things go wrong. Human oversight and governance tie all of this together, providing governance structures, policies, and regulatory compliance to guide development and use.

Autonomy without oversight undermines accountability, while an unreliable or opaque system makes it hard to trust or verify behavior. Ignoring privacy conflicts with the fundamental expectations of trustworthy AI.