Literature Review Assistant

Use the assistant to plan and structure a review, not to supply its evidence. It is good at proposing the themes a literature on your topic is usually organised around, the key constructs and theories to cover, and search terms and synonyms worth trying in Scopus, Web of Science or Google Scholar. It does not search a live database, so it cannot tell you what was published last month or confirm that a particular study exists.

The most productive workflow is to generate a thematic outline, run the suggested searches yourself, and then read and record each relevant paper in a synthesis matrix — author, year, context, method, sample, key finding. Once the matrix fills up, compare it with the outline: themes with many studies become sections, themes with few become candidate gaps, and themes with conflicting results become the most interesting parts of your discussion.

Write the final review in your own words, organised by theme rather than paper by paper, and cite only sources you have opened and read. Examiners can quickly tell a review built from abstracts and summaries from one built from engagement with the studies, and similarity checks increasingly flag AI-paraphrased text. The tool shortens the planning stage; the reading and synthesis remain your work.