AI Literature Review Workflow in 2025

Dear Scholar,

When I published my literature review workflow diagram, it garnered positive and negative reactions. Surprisingly, a lot of academics reject the idea that AI should be used in literature reviews because we as scientists will not be able to maintain our creativity or learn much. Here are some reactions:

All of these criticisms come from senior academics with years of experience. For instance, the professor on the top left rejects AI-written manuscripts, yet AI wrote the majority of my papers using AI with no complaints from reviewers or supervisors. The other two are complaining about learning less with AI, yet I am working on papers 4 and 5 after just two years of my PhD, because AI has enabled me to learn faster.

Who is right?

From a standpoint of experience, every change seems difficult because any change usually leads to worse results. AI is a complex tool that requires learning to use effectively before it surpasses your current methods, especially if you have years of experience.

For that reason, I think academics, especially senior ones, require AI education before they can make proper judgments about this technology. And paired with years of experience, this AI education could truly start transforming academia.

That's the mission of the Effortless Academic!

To provide a little bit more context and depth to this infographic, I wrote a long article describing the details of the AI lit review workflow and linked it to dozens of detailed tutorials:

This article goes step-by-step through the lit review process developed over years of experimenting and links to other blog articles to fill in the details. It gives clear suggestions on which tools to use and why. If you are serious about integrating AI into your literature review workflow and are not opposed to doing things differently than you have done for years in the past, this article is for you. However, as AI is a very personal technology, what seems helpful to one person might be superfluous to another. So take some time and experiment with the ideas suggested in this article.

Finding Research Gaps Course

Quite a few members of the Effortless Academic community asked me to migrate previous webinars onto the platform. This week, I migrated the 2024 research gap & visual synthesis webinar onto the platform. (If you have purchased this webinar, you will receive an email on how to access the materials.)

About the webinar: Identifying research gaps is the most important first step in contributing impactful science. Discover how to identify critical gaps in your own field using a clear, step-by-step approach. The three-part process includes effective note-taking, visualising knowledge, and finding research gaps.

Here are the details of what you can learn in this webinar:

► A step-by-step strategy to identify research gaps
► A note-taking system to remember and access everything you ever read
► Develop a visual language to summarise your findings
► Use literature maps to find, prioritise and keep track of papers.
► Find citation gaps in any publication (even your own).
► Synthesise information in your domain graphically
► Spot conceptual and methodology gaps using mind maps
► Use AI search to help find research questions and even identify gaps in seconds

This is the only webinar that deeply focuses on using visual language to synthesise information, and it is a method that has helped me a lot to understand complex topics and create new research ideas:

I would recommend this course, particularly to students starting their PhDs or master's programs who need to build a deep understanding of a very complex topic.

​

Wishing you a wonderful weekend,

​

Ilya Shabanov, The Effortless Academic​

​

PS: I am running a free webinar on AI agents using SciSpace this week. Sign up here if you want to learn more about this technology. A recording will be posted on the SciSpace Youtube Channel a few days after the webinar.

​