Quality of explanation of xAI from the prespective of Italian end-users: Italian version of System Causability Scale (SCS)

Explainability of AI systems matters across cultures: This study validates that measuring the quality of AI explanations is important enough to translate and adapt assessment tools for different languages and cultural contexts, recognizing that how people understand AI explanations may vary by region.

A practical 9-question tool is now available for Italian speakers: The validated Italian System Causability Scale (I-SCS) provides a standardized way to evaluate whether AI systems are providing clear, understandable explanations to end users in Italy, with one question removed during validation to ensure relevance and clarity.

End-user comprehension was prioritized in development: The researchers conducted cognitive interviews with representative Italian users to confirm they actually understood what the questions were asking, ensuring the tool measures real-world comprehension rather than just technical accuracy.

This tool can guide AI developers: The validated I-SCS can now be used by AI developers and researchers to assess and improve the quality of explanations their systems provide to Italian users, supporting more responsible and transparent AI deployment.

Quality explanation measurement requires rigorous validation: The study demonstrates that creating reliable tools to assess AI explainability involves multiple steps including professional translation, statistical validation (content validity testing), and direct user feedback—not simply translating existing tools word-for-word.