A core requirement under the EU AI Act, demanding that AI systems, particularly those classified as high-risk, be designed and developed to allow for adequate communication and documentation of their processes, especially regarding their inputs and outputs, to instill user trust.

‍