I have a new app out, VerusCite. With the recent rise in popularity of GenAI tools like ChatGPT and Claude, this has also come along with academics writing slop articles.
One of the ways to check that slop is via looking at the articles citations. LLMs have some predictable failure modes in writing papers whole cloth – they tend to get details like complicated author lists wrong, or swap out incorrect journal titles. VerusCite is a tool for editors and reviewers to use to verify citations in a fast and cheap application.
It costs $2 to review a paper (and you get two free reviews on sign-up). If you want to see the output of a single example though, check out https://veruscite-data.com/share/E1-rn3TBwksENO_3hlmsRD0IG-EnXd6vfQg-upQZWQM
In addition to hallucinations, I have made many parts of the application just useful to editors in general. Many papers have minor errors in their bibliographies; typos, years off, author swaps, bad URLs, etc. Here is an example – no hallucinations that signal poor writing, but has seven different errors in the bibliography.
This is par for the course (it is quite possible 5% of citations have errors that look like this). The website has convenient tools to edit citations and export the fixed citations (whether minor errors or gross hallucinations) in various formats.
This makes much lighter work of the tedious job of formatting and checking citations for editors. One of the ways I think is critical to build generative AI tools is to consider the human in the loop from the start. My tool will ultimately make some errors (I error rate estimates in my public benchmark). I want it to be as fast for a human to confirm (or refute) the LLM label.
If you are an editor or a reviewer, I highly suggest you check the application out. Peer review journals (and pre-print servers that review the applications before posting), will need to use a tool like this as a first pass to ensure slop is not being posted.
Andrew Wheeler 
JENNIFER MARIE GONZALEZ
/ August 9, 2026Thank you for building this. I have caught papers that used AI in the same way (thoroughly examining the references) by hand, and so this would have been tremendously helpful. I will absolutely pass this on to our editor-in-Chief (Who will presumably pass it to Taylor and Francis).
And I’ve thought about this a lot as an associate editor for a journal… The writing was terrible before. If folks use AI to help polish their introduction or discussion (NOT write the whole thing… which is absolutely unethical and bad), is that a good thing?
Some academics have issues with English… especially those from other countries or for whom English is not their native language. I see a solid use case for AI there. But if the use is so extensive that their references are incorrect, that is a massive problem.
Question for a potential future blog post: When an AI-assisted review flags an article as being AI-generated or otherwise problematic, what do editors do?
For the article I mentioned above, we ultimately decided to reject the article outright rather than give the authors an opportunity to correct their work. The reason was that the problems went beyond simply poor writing or inappropriate use of AI—the authors had demonstrated an established lack of ethical boundaries, which made us question whether we could trust them to correct the problems they had introduced.
But this raises a broader editorial question: What would you do?
Andy Wheeler
/ August 9, 2026Thank you Jenn, appreciate any forward. Happy to gift whomever some extra credits to try it out as well.
So personally think folks should be allowed to use AI for copy-editing. When I see clearly AI articles at Journal of Quant Crim, I just say “this is clearly AI written, please provide replication code, as I do not know if you made up the entire article whole cloth (data analysis and all)”. Otherwise it is just the same review. (I have made comments along the lines “this AI writing is a hyberbole and not fit for a peer reviewed journal”.)
I can understand journals have statements to the effect no AI at all. Definitely people should disclose its use at this point. So could totally see a reject if simply failing to disclose.