Getting Started
Curious about how to begin using AI in scholarly work? Start by:
- checking your institution’s guidance,
- contacting your librarian to learn which AI tools are built for researchers,
- explore their features and capabilities by using the tools yourself.
- and encourage learners to explore AI-tools with an ethical mindset and “trust but verify” approach.
Your librarians might be able to provide an approved list of tools for scholarly work. AI-powered research tools have unique capabilities but often run with the same underlying tech from the AI apps you are familiar with (i.e., ChatGPT, Google Gemini, Claude). Your library staff has likely vetted the most popular research tools, like Elicit and ResearchRabbit, and can provide advice on how to get the most out of them.
Once you know which AI tools to use for your research, explore their features and identify the use cases that will apply to your work. As you gain experience, it will help inform the guidance you will develop for your students to properly use them as well. It’s also important to keep in mind ways you may want to consider revising your assessments to challenge learners to exercise uniquely human capabilities3 (Risepoint, 2025)
Current Capabilities of AI in Research
AI-powered research tools are quickly replacing the keyword search techniques of the past (e.g., using quotes around a phrase, Boolean operators) with automated discovery of ‘similar papers’ and providing you with visualizations of the networks of articles available.
For example, an AI research tool like ResearchRabbit can accelerate exploration of scholarly articles by reducing the time it takes for you to locate and draw connections between similar works. What used to be a slow fragmented process is now dramatically faster, which naturally allows the researcher to build a collection of scholarly literature to support (or challenge) a hypothesis more efficiently than ever.
Capabilities between tools vary, but the process of initially reviewing available research has become much faster. Clearly identifying gaps between research is a feature that seems to be still in progress but getting better quickly as well. Here are two popular AI research tools:
- Elicit: search, summarize, extract data from, and chat with 138 million papers4
- ResearchRabbit: search, interactive maps, recommendations, author tracking, and collaboration5
Helping Students Navigate AI in Academic Research
As part of your institution’s guidance, a set of resources for learners might be available. If not, faculty often end up the main source of information about AI for research. This can be challenging because even if professors already have a defined “AI stance” for class assignments, conversations about AI for research asks us for an even deeper level of engagement around topics such as ethics, citation, plagiarism, disclosing use of AI in writing research, and publisher restrictions on AI-generated scholarly work.
Again, if you’re provided with an approved list of tools from your school, that should be your go-to resource, but remember even with a list of AI tools for research, students need guidance on all the topics previously mentioned. When addressing learners, purposefully scaffold their experience for when they begin using AI in research, consider discussing each of the following with students before they get started:
- AI has accuracy and hallucination issues. It can be wrong and sometimes it just makes stuff up. It can be hard to tell the difference. Be on the lookout for incorrect information and confirm outputs are true with verifiable sources: trust but verify.
- Avoid hiding AI assistance in your work. Transparency, citations, and disclosure are even more important. Be clear about what tools you used, and be able to answer the following questions:
- “How specifically did you use AI in your research?”
- “Did you indicate AI-support in your methodology and cite any contributions?”
- “Can reviewers trace and verify your use of AI?”
- “For those seeking publication, what are the publisher’s guidelines and/or restrictions related to AI-generated content?”
- Provide students with clear examples of the following:
- How to use AI-powered tools for discovery as well as citation managers
- Explanation of how AI supported a research process (i.e., “I used Elicit…I verified sources manually…AI was used for…”)
- Violations of academic integrity with AI (e.g., can your students point out issues and do they know how to prevent them?)
AI Prompts to Try
Depending on the guidance from your institution and whether students are encouraged to use a specific ‘closed system’ tool, students will benefit from your guidance with prompt engineering even if they are using generally available AI tools. The prompts below demonstrate how carefully crafting AI prompts can support student research by helping learners write effective prompts or by allowing AI to assist via meta-prompting (i.e. using prompts that improve your prompts). It all begins with your own exploration, so you can effectively guide learners to properly, and cautiously, utilize both general AI tools and the new more specialized AI-research tools now available.
Refine Research
- “I’m studying [topic]. Ask me five questions that could help narrow this into a researchable question.”
- “Help me identify what’s too broad or vague about my research question: [insert question].”
Build Search Strategies
- “Act as a librarian. What subtopics or search terms might I use in an academic database to explore [topic]?”
- “Suggest academic keywords and synonyms I could search for in databases about [topic]. Explain which might appear in titles vs. abstracts.”
- “List potential Boolean search strings to explore [topic] in scholarly databases.”
Understand Scholarly Contexts
- “Summarize the main debates or tensions the scholars discuss in the following article. Don’t give me conclusions, just outline the arguments. [attach article or link]”
- “Help me identify key scholars or landmark works that are frequently cited on [topic].”
Evaluate and Synthesize Sources
- “Summarize the article I found. Identify the research question, methods, and findings. [attach article or link]”
- “Based on these abstracts, what themes or gaps appear across the research? [paste text]”
Support Writing and Analysis
- “I’ve drafted a thesis statement: [paste text]. Suggest questions I should answer with evidence to support it.”
- “Give me an outline framework for a literature review on [topic], but don’t fill it in.”
Prompt Refinement
- “I want to use AI to explore scholarly perspectives on [topic]. Help me write a prompt that is specific and academically focused, encourages critical thinking, and avoids summarizing or generating fake citations.”
- “Suggest three improved versions of my draft prompt: [paste prompt].”
Role Based
- “Rewrite my prompt so that the AI responds like a subject librarian or research mentor rather than a content generator. Here’s my current prompt: [paste].”
Critical Thinking
- “Improve my prompt so that it pushes for multiple perspectives or competing interpretations of [topic], instead of a single summary.”
References
- Bolaños, F., Salatino, A., Osborne, F., & Motta, E. (2024). Artificial intelligence for literature reviews: Opportunities and challenges. Artificial Intelligence Review, 57, Article 259. https://doi.org/10.1007/s10462-024-10902-3
- Risepoint. (2025, June). Voice of the online learner 2025 (14th ed.). Risepoint. https://risepoint.com/wp-content/uploads/2025/06/Voice-of-the-Online-Learner-2025.pdf
- Risepoint. (2025, August). Why human-centered, AI-integrated assessment design matters. Faculty eCommons. https://faculty.risepoint.com/why-human-centered-ai-integrated-assessment-design-matters/
- Elicit Research. (n.d.). Elicit: AI for scientific research. Retrieved November 13, 2025, from https://elicit.com/
- ResearchRabbit. (n.d.). Features [Web page]. Retrieved November 13, 2025, from https://www.researchrabbit.ai/features
