Systematic reviews are the bedrock of evidence-based medicine, playing a crucial role in shaping healthcare guidelines. However, conducting these reviews is a laborious process, often involving sifting through thousands of research articles. This bottleneck can delay the translation of research findings into practical applications. But what if AI could step in to streamline this process? A new study explores how large language models (LLMs), the technology behind tools like ChatGPT, can automate the time-consuming task of literature screening in systematic reviews. Researchers tested an LLM-powered system against both traditional manual review and a commercial AI tool called Rayyan, using a completed review on Vitamin D and falls as a benchmark. The results were striking. The LLM system achieved a remarkable 95.5% reduction in screening time compared to the manual approach, whittling down 14,439 articles to just 78 for final human review. Importantly, the LLM didn't miss any relevant studies, demonstrating its potential to enhance both efficiency and accuracy. While Rayyan also offered time savings, the LLM's ability to tackle both title/abstract and full-text screening, combined with its superior performance, highlighted its potential to transform how systematic reviews are conducted. This breakthrough could accelerate the pace of medical research, enabling faster updates to guidelines and ultimately, better patient care. However, further research is needed to validate the LLM's effectiveness across different types of reviews and to refine its decision-making transparency. This study offers a glimpse into a future where AI empowers researchers to focus on analysis and interpretation, leaving the heavy lifting of literature screening to intelligent machines.