AWS Certified AI Practitioner AIF-C01 Practice Question

Interpretability risk refers to the difficulty humans have in understanding why a model produced a particular output. If stakeholders cannot determine which inputs or internal steps motivated specific wording in an email, the model's behavior is not transparent, illustrating low interpretability. Hallucinating new product details is primarily an accuracy risk, slow response times relate to performance or scalability, and rising inference costs are a financial concern rather than an interpretability issue.