Explainability in AI typically involves which practice?

• A. Explaining how the AI arrived at its result, including citing sources in outputs.
• B. Increasing inference speed by hardware acceleration.
• C. Minimizing data provenance documentation.
• D. Removing citations from outputs.

Answer: (A)

Explainability in AI is about making decisions understandable by humans. Explaining how the AI arrived at its result and including sources or rationale in outputs provides the reasoning and evidence behind the decision, which supports trust, auditability, and accountability. This transparency helps users evaluate claims and verify results. In contrast, speeding up inference changes performance, not understanding; minimizing data provenance documentation or removing citations reduces transparency and verifiability.