“AI slop” is everywhere.
It’s not hard to recognise. It is the low-quality images and content cluttering up your social media feeds. Created by generative AI, it often lacks substance and contains errors.
To be fair, some of it looks and sounds fantastic despite its sloppiness.
But social media is one thing; now AI slop is cluttering the academic peer-review system.
AI-generated manuscripts are flooding academic journals
Academic journals are reporting a sharp increase in submissions containing AI-generated writing.
Submissions to Organization Science, for instance, one of the top journals in management research, have risen by 42% since the launch of ChatGPT in November 2022. Its editors report that AI-generated writing accounts for much of this increase and that these manuscripts are, on average, harder to read and more likely to be rejected.
Academics, like the rest of the world, have caught up with the crest of the AI wave and discovered that parts of the research and writing process, or even massive chunks of it, can be fast-tracked by using ChatGPT.
The productivity benefits are real
A Harvard Business School study confirms the productivity gains from AI are real. The study found that professionals using AI completed 12% more tasks, worked 25% faster, and produced work output that was assessed as being 40% higher in quality.
The temptation is therefore understandable.
Academia, with its all-pervasive “publish or perish” research culture, is now facing a crisis. The institutional incentive has always been there to create more research, rather than better research. With the rise of ChatGPT, journals are scrambling to respond to an influx of manuscripts with high levels of AI-generated content.
Manuscripts marked by heavy AI use, by all accounts, require much more time and effort on the part of journal editors to weed out, due to the overriding language style typical of AI slop: dense, vaguely impressive, difficult to follow.
Academic journals are fighting back
In the online submission process, journals are now asking a question that would have seemed unusual a few years ago: “Did you use AI to prepare this article?”
When the answer is inevitably yes, more information is required. Whatever the answer, an outright rejection may well follow.
Government agencies also require applicants to disclose any use of AI in tender applications, and the same applies to recruitment panels.
How can academics avoid AI slop?
What to do? First of all, before opening any AI tool, grab a strong cup of coffee and write what you want to say. Let Grammarly or a skilled copy-editor help you shape it into decent prose.
Of course, like the Sirens in The Odyssey, AI can be impossible to resist: its language is polished, persuasive and instantly available. Follow its call too far, and your own voice is lost, never to be heard again.
If it is already too late – and you have already succumbed to the dreaded Siren call of ChatGPT – reach out to Wise Directions.
We can use specialised software to identify signs of AI-generated writing and, in turn, help you restore your original, distinctive academic voice before it’s completely flattened by the algorithm.
– Marshall, August 2026