Task $9\mathrm{.}4$ Develop elements for a robust, potentially automatable evaluation approach for trustworthiness and quality of cutting$-$edge AI $($e$.$g$.,$ LLMs$)$ with a specific focus on the needs of AIoDP $($AI on Demand Platform$)$ users $($T$-$Systems$)$

This task will perform the following activities:

$1)$ Investigate and demonstrate how trustworthiness and quality of cutting$-$edge AI applications $($e$.$g$.$ LLMs$)$ can be supported and assured along the economic value chain and the development cycle

$(2)$ Explore potential automation of the technical evaluation and assessment approach

$(3)$ Clarify the roles of all stakeholders involved, including LLM providers, retainers and operators