This paper examines how AI tools can effectively support the literature review process from a practical point of view. The primary concern is the accuracy of the output, prioritizing correctness over fabrications and falsehoods when factually accurate information is required. The risk of drawing false conclusions complicates the recommendation of these tools to aid in the research process. There is an urgent need to develop responsible automation tools to help with every step of the systematic review workflow. These tools can save reviewers time and accelerate the overall process. However, integrating AI-based applications into the systematic review process remains limited. We investigated the potential of using large language model (LLM) applications to automatically identify the most relevant collection of articles and assessed whether LLM-based tools could effectively screen documents. Furthermore, we evaluated whether LLM-based applications could analyze selected articles from a bibliometric perspective. The results suggest that the use of LLMs may offer significant benefits.