Artificial Intelligence (AI) and the literature review process: Welcome
This guide shows you how you can use AI tools to help with aspects of the literature review process: Abstract and title creation, Scoping and protocol development, Searching, Quality assurance, Screening, Data extraction, Synthesis and Reporting.
You will learn the terminology behind the technology, find advice about prompt engineering, the art of interacting with AI agents, and a list of the AI tools referenced in this guide.
We also highlight key ethical concerns about the use of AI tools when reviewing the literature (i.e. lawsuits for copyright infringement).
University guidance about using AI
AI tools analyze a large volume of articles, accurately extract and summarise information, provide different perspectives and integrate information from across different disciplines (Blanco-Gonzalez et al., 2023; Huang and Tan, 2023). These are among the potential benefits of using large language models in learning as the QAA identifies in its January 2023 briefing paper.
However, the QAA briefing paper recognizes that if you present the output of generative AI tools as your own work, there is a clear challenge to academic integrity.
The student AI guidelines recognize that AI tools can be helpful in assisting learning. You can use an AI tool in the preliminary stages of your assignment process, for instance to brainstorm, as well as planning/researching to identify relevant sources. There is evidence that models such as ChatGPT are really useful in this respect and there is no need to cite the AI tool.
The Centre for Academic Success (CAS) has prepared video examples of using AI to study and an artificial intelligence (AI) tools decision tree to help you determine when it is appropriate or not to use AI tools.
The guidelines emphasise that you must submit your own original work and use AI tools only when explicitly permitted. Unauthorised AI use in assessments is considered academic misconduct, with potential penalties (see the academic misconduct procedure).
Most current assignment briefs will only permit very limited use of AI tools.
Follow the guidance on referencing generative AI in the referencing pages.
The subject is in its infancy and it will take many years before some consensus is reached in the academic community on questions such as:
- Which tools work best in supporting the searching process?
- How best to use AI tools?
- Should institutional subscriptions to specific tools be taken out where possible?
However, there is demand from students for guidance on the responsible and effective use of AI (see section 5.2.2 within the student expectations section of Jisc's AI in Tertiary Education, 3rd edn., p. 13). This guide attempts to meet some of that demand.
The proportion of students who have not used generative AI for their assessments fell from 47% in the 2024 HEPI/Kortext survey to 12% in the 2025 survey (Freeman, 2025). Students use AI tools to save time (51% of the survey of 1042 students) and to improve the quality of their work (50%).
One of the recommendations of HEPI Policy Note 51 (Freeman, 2024) is that institutions should teach students how it can be used appropriately and effectively, such as how to write an effective ‘prompt’, and how to evaluate the quality of content generated by AI. The Centre for Academic Success has made its resources on Using AI available to you on its Moodle site. Do make use of these resources.
HEPI Policy Note 61 recognises that student use of AI is inevitable and often beneficial, helping with learning and productivity. Without guidance on and the teaching of responsible use of AI, there is a risk of widening the existing digital divide between those willing to try AI tools (male students, those on STEM and Health courses and those more socioeconomically advantaged) and those lacking the digital skills to do so. It would also risk students not being aware of issues of bias, hallucinations, ethical use, environmental impact and the use of personal data associated with these tools (Freeman, 2025).
This guide does not cover the use of generative AI in coding, in the production of music or in the creation of artwork. The AI tools that are included are those which are a) freely available, and b) mentioned in the academic literature in the context of searching and reviewing the literature.