Starting a Literature Review with GenAI: A Supervisor’s Secret Weapon

Robot in the classroom, the idea of ai in education. shows a classroom and teacher.

If you supervise research students at undergraduate or postgraduate level, you are likely to be very familiar with the "blank stare"—that moment a student first confronts the sheer, overwhelming mountain of academic literature they are expected to read, synthesise, and critique. Information overload is a genuine academic pain point, often manifesting as a severe case of "blank page" syndrome.

As academics, we know that starting is half the battle. This is where Generative AI, like ChatGPT-4o, shines not as a tool to write the review for the student, but as a structural scaffold. Much like using AI as a mirror to transform vague student ideas, we can use GenAI to help students map the thematic landscape of a topic before they dive into deep reading. It breaks the ice, organises chaos into a digestible format, and gives them a structured starting point.

Here is a practical, step-by-step workflow you can share with your students to help them generate a foundational literature matrix.

The Workflow: Mapping the Landscape

The goal of this exercise isn't just to find papers; it’s to identify common features and themes across the literature. By forcing the AI to iterate and categorise, we teach the student how to look for cross-disciplinary themes rather than just reading papers in isolated silos.

Have your students start with this complex prompt, adjusting the topic to their specific research area (in this example, we use "VR in Higher Education"):

Prompt 1:

You will generate a search and produce a summary table of published papers on the following topic "VR in Higher Education". > Iterate 5 times the following Step 1; Step 2; Step 3; and Step 4. > Step 1. Search for 3 new papers relating to the topic and add to the list of papers stored. > Step 2. Identifying Common Features to at least three papers not included in the previous interaction. Each iteration all Common Features are maintained but can be revised. > Step 3. On each iteration from the papers stored revise the following table. The table will have four parts: Common Features, summary of the Common Feature, identified and included all papers that have Common Feature, all papers that don't match the Common Feature. > Step 4. Add the full reference to all the papers to a Harvard styled reference list. Display the full table. Display the full reference list.

Why this works: The magic here lies in the iteration. The AI builds a comparative matrix, separating papers that share a theme (like "Challenges in Implementation") from those that don't. It immediately provides the student with a high-level, organised view of the current academic discourse.

Once the table is generated, the next step is translating that raw data into academic prose.

Prompt 2:

Using the table and reference. Analyze the results and summarise the results with appropriate citations.

This generates a short, synthesised summary of the findings, helping the student see how an academic narrative is woven together from disparate sources.

Levelling Up: The Chain of Density (CoD)

Once students have the basic summary from Prompt 2, they shouldn't stop there. We want to push for richer, more academically dense writing. This is where you can introduce the Chain of Density (CoD) prompting technique.

Instead of accepting the first output, the CoD approach asks the AI to rewrite the summary multiple times, each time identifying missing "entities" (specific methodologies, nuanced findings, or theoretical frameworks) and weaving them into the text without increasing the word count. It forces the summary to become less generic and more informationally rich, mirroring the density of actual academic writing.

Ethics and Critical Assessment: The Reality Check

Before sending students off to generate literature matrices, we must establish a clear ethical boundary. GenAI is an assistant, not the primary researcher.

Academics and students alike must be acutely aware of AI's limitations—most notably, its tendency to hallucinate. AI models can, and will, invent realistic-sounding citations or confidently misrepresent a paper’s methodology. Therefore, this workflow is strictly a starting point.

Students must physically track down, read, and verify every single paper the AI may suggest. The AI's synthesis should be treated as a draft map of a new territory; you still have to walk the terrain yourself to verify the landmarks. Relying blindly on AI outputs without human verification is a fast track to academic misconduct.

Final Thoughts

Used thoughtfully, GenAI transforms the daunting initial stages of a literature review into an engaging, structured exercise. It empowers students to overcome the blank page and helps them think thematically from day one.

Applying prompts to GenAI/LLMs

Search This Blog