The application of the easy-to-read (E2R) methodology is one of the ways to achieve cognitive accessibility and specifically, it is a path that guarantees the right of access to information of people with reading comprehension difficulties and thus improves their daily life. This methodology includes a set of guidelines and recommendations whose goal is to present clear and easily understood documents. Such guidelines are used in the manual processes of (a) adapting existing documents and (b) producing new materials. These processes are very time and human-resource consuming, due to the need of involving E2R experts as well as people with cognitive disabilities. In order to alleviate such manual processes, we are currently investigating the development of methods, based on Artificial Intelligence (AI) techniques, to support the E2R adaptation of documents in a (semi)-automatic fashion. The main goal of this research is to help E2R experts in their daily tasks of (a) assessing a particular document with respect to the E2R guidelines and (b) transforming such a document according to the E2R methodology. In this paper we present our initial efforts toward the development of an AI-based application for supporting the E2R adaptation of documents. These efforts are the elicitation of E2R needs and informal requirements and the design of an application called FACILE.