This paper summarizes some of the major contributions from speakers at Tutorial 4 of the PFIA Summer School in Strasbourg France, which took place at Strasbourg University between July 3 and 7, 2023. The tutorial was organized by the Trustworthy AI Lab of the CNRS and Strasbourg University and the different contributions, placed in the context of the AI Act of the European Commission, adopted by the European Parliament in June 2023, illustrate why human-centered approaches to validating Artificial Intelligence and its applications in context are needed, and suggest how this can be made possible by including all actors, from citizens to domain experts, and viewpoints from society science and industry in a long-term assessment loop for trustworthiness evaluation. Rules of good procedure must be found and established within the specific framework of a given field of application. At the same time, domain-related limitations and obstacles must be considered. The potential as well as major challenges of implementing holistic and inclusive procedures for evaluating trustworthy AI are brought forward.