AI tools for scientific literature research can be divided into two categories:
Finders
- Search with phrases or complete questions
- Mainly finds journal articles in English that have a DOI and have been published in open access
- Finds few books
- Good for natural sciences and medicine subjects
- Not very suitable for social sciences and humanities
- Examples: Consensus; Elicit; ORKGAsk (data protection compliant)
- Further Information onFinders can be found here.
Connectors
- The starting point for the search is a paper that has already been found (seed paper) and its DOI
- Based on citations and thematic relationships, visual interactive maps of publication networks are created
- Very quick access to a topic - particularly suitable as an introduction to a search
- Examples: Open Knowledge Maps, Research Rabbit, Local Citation Network (focus on data protection)
- Further Information on Connectors can be found here.
Important notes on use
Recommendation
AI tools are a valuable addition, but not a substitute for traditional research methods. Use them as intelligent support, not as your sole source of information. They are particularly helpful as a starting point for research. The library's research tools can be found at DBIS.
We particularly recommend the interdisciplinary database, databank Scopus AI. As with the Finders, you can enter your research question here, so there is no need for search terms such as keywords. The hits here come from the database index and are therefore of high scientific quality.
Assessment of the use of chatbots for 1. science and scholarshop (allg.) literature research
Since early 2025, chatbots such as ChatGPT, Gemini, Microsoft Copilot, and Claude have improved significantly thanks to live access to the web and features such as “deep research.” Nevertheless, problems with hallucinated or incorrect references continue to occur. In addition, the generated responses often list only a limited number of relevant publications. Currently, “general-purpose AI chatbots” therefore do not yet offer any significant added value or noticeable time savings in scientific research.