Challenges in Evaluating the Quality and Credibility of AI-Generated Grey Literature

In the rapidly evolving landscape of scientific communication, grey literature (GL) stands as an important resource. Defined as recorded information not controlled by commercial publishers and made publicly available without the traditional peer-review process, GL includes a wide range of materials such as reports, theses, conference papers, and white papers. The advent of Artificial Intelligence (AI) has revolutionized content creation, offering unprecedented speed and volume. However, this advancement presents unique challenges, particularly in managing, evaluating the integrity and quality of AI-generated content. This paper examines these challenges, highlighting the limitations of traditional evaluation methods for AI-generated outputs. It discusses the rise of AI in GL production, the inadequacies of conventional assessment approaches, and the specific issues associated with AI-generated content. The paper concludes with proposed solutions and best practices to ensure the integrity of scientific communication in the AI-driven era.