AI Tools for Research

Have you had AI tools fabricate citations or facts? Looking for additional sources for your literature reviews? Interested in using AI tools for research but not sure how to do it ethically and responsibly? Want to go beyond ChatGPT and keep up on the latest AI tools that might help you with your research? Check out this guide for more information and

• Identify ways AI can (and can’t) help you do research, including literature reviews, ethically
• Differentiate between general generative AI (GenAI) tools and research literature-based AI tools
• Identify some research-based tools that might help in your lit reviews

Also see our guide below for more general information on AI.

Generative AI tools are powerful additions to your research toolkit. However, they should be used cautiously and thoughtfully. In addition to the usual considerations for using Generative AI tools for use in research, see the presentation links under GenAI Resources for more considerations specific to research. In general, remember that:

• These tools are often a black box (even developers are sometimes unsure why they do what they do) and likely biased.  
• Large language models (LLMs) also are known to “hallucinate” (e.g. create plausible sounding but not necessarily correct material); they are designed to give you the statistically most likely next words or output - prioritizing fluency over accuracy (the latest models are better but not perfect).
  • Research tools based on the research literature are more likely to use retrieval augmented generation, a process by which the chatbot does a search and then bases its response on the results (in this case a defined set of papers with actually existing references, unlike some of the more general chatbots like ChatGPT); this can reduce but doesn't eliminate hallucinations.
• Chatbots don't "know" about the real world. Remember, these are “stochastic parrots” (Bender et. al., 2021), not all-knowing magic beings or anything like a human intelligence no matter how fluent they are!
• Clemson University Libraries guide on Considerations on AI in Locating Sources says, "Generative AI tools can be very powerful in connecting researchers with scholarly information. However, there are some important considerations:
  • Not all scholarly information has been digitized and therefore is not available for generative AI searches
  • Not all scholarly information is freely available online and is therefore not available for AI tools to explore
  • Some fields have more information in the digital environments than others
  • Search engine algorithms may de-prioritize niche research areas as they are not frequently utilized
  • Free tools like Elicit and Consensus use Semantic Scholar to inform their searches and Semantic Scholar indexes both publications from predatory journals as well as retracted publications"

Wondering which tools to use when? Consider your workflow (and whether you need GenAI at all), the task you are trying to accomplish, the level of accuracy and privacy you need, the policies for whatever task you're doing (for example, does your professor or the journal you're submitting to or your work have an AI policy). Also note that the tools change rapidly so the best tools today may not be the best tools tomorrow.