Explainability in AI refers to what?

• A. Answering how the AI solution arrived at its result, such as citing sources in outputs.
• B. The ability to run models at higher speed.
• C. The capacity to use more data.
• D. The cost of training.

Answer: (A)

Explainability is about understanding the reasoning behind a model’s decision. It asks how the AI arrived at a particular result, including showing which inputs were influential and, when possible, presenting supporting information or evidence—for example, feature importance, a decision path, or even citations to data sources that back the reasoning. This helps users trust the model, debug issues, and comply with regulations that require transparency. The other options relate to performance aspects like speed, the amount of data used, or training costs, which are important but not what explainability is about.