Silvi is an end-to-end screening and data extraction tool supporting Systematic Literature Review and Meta-analysis. Silvi.ai was founded in 2018 by Professor in Health Economic Evidence, Tove Holm-Larsen, and expert in Machine Learning, Rasmus Hvingelby. The idea for Silvi stemmed from their own research, and the need to conduct systematic literature reviews and meta-analyses faster. The ideas behind Silvi were originally a component of a larger project. In 2016, Tove founded the group “Evidensbaseret Medicin 2.0” in collaboration with researchers from Ghent University, Technical University of Denmark, University of Copenhagen, and other experts. EBM 2.0 wanted to optimize evidence-based medicine to its highest potential using Big Data and Artificial Intelligence, but needed a highly skilled person within AI. Around this time, Tove met Rasmus, who shared the same visions. Tove teamed up with Rasmus, and Silvi.ai was created. Silvi uses AI to increase the speed of collecting and analyzing published data to created meta-analyses and systematic literature reviews. When using Silvi, the researcher still makes all the scientific decisions, but with AI supporting data extraction, the speed of doing meta-analyses increases immensely. Silvi is directly connected to literature engines to ensure that the results are always up to date. These core qualities of Silvi ensures a tool that quickly helps you create high quality evidence that stays relevant. Silvi helps create systematic literature reviews and meta-analyses that follow Cochrane guidelines in a highly reduced time frame, giving a fast and easy overview. It supports the user through the full process, from literature search to data analyses. Silvi is directly connected with databases such as PubMed and ClinicalTrials.gov and is always updated with the latest published research. It also supports RIS files, making it possible to upload a search string from your favorite search engine (i.e., Ovid). Silvi has a tagging system that can be tailored to any project. Silvi is transparent, meaning it documents and stores the choices (and the reasons behind them) the user makes. Whether publishing the results from the project in a journal, sending them to an authority, or collaborating on the project with several colleagues, transparency is optimal to create robust evidence.