Explainability in AI governance focuses on:

• A. The hardware requirements.
• B. The speed of performing inferences.
• C. The ability to understand how AI models reach decisions.
• D. The ability to train with unsupervised data.

Answer: (C)

Explainability in AI governance is about the ability to understand how AI models reach decisions. It means revealing the factors and pathways that lead to a specific prediction or action, so auditors, regulators, and stakeholders can see why the model chose what it did. This understanding supports accountability, trust, debugging, bias detection, and regulatory compliance, all of which are central to governing AI systems responsibly.

That emphasis on understanding the decision-making process is why this option is the best fit. The other aspects—hardware requirements, inference speed, or training with unsupervised data—relate to performance or data characteristics, not to how or why the model arrives at its decisions.