AI Discovery Tools
Discovery tools are resources that facilitate literature searching and mapping of information. The following is a non-exhaustive list of artificial intelligence (AI) discovery tools used to identify scholarly research. The tools listed below predominantly scrutinise open access journal articles.
For a lot of disciplines, journal articles are not the primary source of research dissemination, and while the percent of open access articles published each year has grown steadily, only around half are published openly, and these were predominantly published in the Global North by research intensive institutions.
For subject based research discovery of both open access and subscription based information, please use:
These tools should be used in conjunction with:
| AI discovery tool | Pricing model | Corpus | LLM used | Discovery type |
|---|---|---|---|---|
| Connected Papers | Freemium | Semantic Scholar | Proprietary | Visualisation |
| Consensus | Freemium | Semantic Scholar | Proprietary | Summarisation |
| Elicit | Freemium | Semantic Scholar | Uncertain | Summarisation |
| Evidence Hunt | Freemium | PubMed and NICE Guidelines | GPT-4 | Summarisation (abstracts only) |
| Inciteful | Free | OpenAlex, Semantic Scholar, CrossRef, OpenCitations | N/A | Visualisation |
| Litmaps | Freemium | Semantic Scholar, Crossref, OpenAlex | N/A | Visualisation |
| Research Rabbit | Free (account required) | Semantic Scholar, PubMed | Uncertain | Visualisation |
| SciSpace | Freemium | Semantic Scholar, OpenAlex, Google Scholar, “trusted repositories” | CoPilot | Summarisation |
| Scite | Subscription | Agreements with various publishers; open repositories; preprint servers | Uncertain | Summarisation |
| Semantic Scholar | Free (sign-up for advanced features) | Open archives & repositories; publisher partnerships | SciTLDR | Summarisation |
Designing a good prompt
Where tools require a prompt or search query rather than a ‘seed reference’ to grow further references, consider using the CLEAR framework (Lo, 2023):
- Concise: brevity and clarity in prompts
- Logical: structured and coherent prompts
- Explicit: clear output specifications
- Adaptive: flexibility and customization in prompts
- Reflective: continuous evaluation and improvement of prompts
Further information on the framework and example prompts are provided in the following reference:
Lo, L. 2023. The CLEAR path: a framework for enhancing information literacy through prompt engineering. The Journal of Academic Librarianship. 49(4), 102720. Available at: https://doi.org/10.1016/j.acalib.2023.102720
Uploading material to LLMs
Uploading research papers or other material without an open license to third-party AI tools is not advised; doing so may constitute commercial use, which is not exempt from UK copyright law. Note that Open Access does not mean the material is openly licenced. The exception to this advice would be Microsoft CoPilot when used as per University of Glasgow guidance.
For further guidance, contact your College Librarian.