Using AI For Literature Searches and Reviews

While curiosity related to using AI in the academic field is natural, educators and counselors face the challenge of making sure that young researchers understand both the opportunities and the dangers AI presents.

High school students transitioning into research roles often lack the disciplinary grounding needed to critically assess AI outputs, making them especially vulnerable to errors, misinterpretations, or ethical missteps. In this case, a student's research career can be ended or their reputation ruined before they have even begun.

A large part of utilizing AI correctly in research, or any field really, is guaranteeing that the user has a foundation or background in what they're researching and can ask the right questions. Research is all about asking questions, and the same is true when we involve AI in research.

Literature Search vs. Review

There has been some debate across the fields as to whether defining "literature search" and "literature review" matters when it comes to using AI for either. In this instance, we can define them as:

• Literature search — An initial step where researchers locate relevant papers, datasets, or books. This is the act of finding sources.
• Literature review — A more advanced step where researchers synthesize, evaluate, and situate existing knowledge to demonstrate novelty and identify gaps.

Many researchers argue that these tasks differ fundamentally. However, for the sake of our current topic—helping high school students use AI appropriately—no matter the definition, the answer is always: don't do it.

To better understand why students shouldn't utilize AI for literature searches or reviews, it helps to look at why some researchers are supportive. The most common argument is that AI can help quickly find potential relevant articles and other sources.

Tools like Elicit, ResearchRabbit, and Semantic Scholar already use AI methods to recommend papers. If used "correctly," this can help novice researchers begin a foundation for their information. The assumption being that they would take the appropriate steps to confirm that the sources fit their needs and are accurate.

Using AI and Literature Searches or Reviews

The best argument against using AI for literature searches and reviews comes down to knowing what questions to ask. All research sets out to answer a question. Using an AI source of any kind to help with research requires the ability to format questions in such a way that the tool will give appropriate answers.

If they are new to a field or to research in general, high school students likely will not know how to phrase questions or how to parse the information they are given by AI.

A great example comes from the world of English literature. For example, a high school student beginning to study the perceived infidelity of Gertrude might be told to start their research with the cuckold, Hamlet's father. Trying to begin some background work in the area, a young researcher might prompt AI with a related sentence. For example, prompting Gemini with this sentence:

Current Consensus of AI Tools for Literature Searches/Reviews

The current consensus in research communities is that while AI tools may have usefulness in conducting literature searches, they are not suitable for producing literature reviews.

AI-driven search engines can sometimes help identify potentially relevant papers more quickly, but their outputs must be carefully verified by the researcher to ensure completeness and accuracy.

In contrast, literature reviews demand critical evaluation, synthesis, and interpretation—skills that AI cannot reliably replicate. Because reviews are considered essential to demonstrating scholarly expertise and establishing research novelty, outsourcing this work to AI risks both ethical violations and significant errors, such as fabricated citations or misrepresented findings. As a result, AI is cautiously accepted as a search aid but broadly discouraged as a review tool.

It's for this reason that educators and counselors should discourage young researchers from using these tools for either searches or reviews until they have a strong foundation in their work and the ability to think critically. Creating good habits at the beginning of a research career will pay dividends later.

Educators and counselors play a crucial role in guiding students through the complexities of modern research, especially as AI technologies blur the line between assistance and misconduct.

While AI can be a valuable support tool in some contexts, it often introduces risks that outweigh its benefits for novice researchers.

High school students, in particular, need strong mentorship to learn when and how technology can be responsibly applied and when traditional methods remain essential. By fostering critical thinking, ethical awareness, and a commitment to reproducibility, educators can help ensure that students approach AI not as a shortcut, but as a tool requiring discernment.