A new Chrome extension called Slop Mop is using Jev, a decision-making AI model, to identify and flag low-value writing on LinkedIn.
AI enthusiast, author, and business consultant Tom Frazier launched Slop Mop on Tuesday. The extension is free, requires no signup and is also available as open-source code on GitHub.
Unlike traditional AI detectors, Slop Mop does not try to determine whether a post was written by AI. Instead, it looks for patterns associated with poor-quality writing, regardless of whether a human or AI produced it.
How Slop Mop Identifies “Slop”
The extension uses nine indicators based on research from SEO firm Graphite Growth into writing patterns that appear more frequently in AI-generated content.
These include excessive hype, engagement bait, empty praise, overly polished lessons, flowery language, stiff wording, contrast-based definitions and posts designed to make users expand them while scrolling.
Slop Mop then uses additional questions to determine whether the content sounds like it was written by a specific person and whether it provides something useful to readers. Both factors can reduce the post’s slop score.
A separate question estimates whether an AI model likely drafted the content. However, that factor cannot trigger a flag by itself and can only reduce the score when the post otherwise appears to be human-written.
It Targets Bad Writing, Not Just AI Content
Frazier said the extension is designed to filter out low-value content rather than act as a conventional AI detector.
“Slop Mop flags poor human content equally to poor AI content,” Frazier told The Register, adding that AI-generated content can be good while human-written content can also be poor.
Posts that cross Slop Mop’s threshold can be flagged or hidden from a user’s LinkedIn feed. The posts are not deleted from LinkedIn and can still be reported through LinkedIn’s existing “Seems Like AI Slop” option.
The extension’s central question is whether a post is worth the reader’s attention.
Built as a Test for Jev
Frazier said his main reason for developing Slop Mop was to experiment with Jev, a probabilistic decision-making AI model from TypeSafe.
Unlike a chatbot, Jev is designed to make decisions for other software systems and is accessed through an API.
Frazier has open-sourced Slop Mop so developers can examine the project and experiment with Jev themselves.
Free Use Comes With a Daily Limit
Slop Mop currently operates through Frazier’s own Jev account and API key. To control costs, he has limited each installation to 250 analyzed LinkedIn posts per day.
Frazier said early usage has cost approximately $0.005 per user per day. He said he may reconsider the free service if the extension becomes widely popular.
For now, he is using the project to study Jev’s tuning parameters and build a dataset.
Frazier also said he could eventually expand Slop Mop to other platforms if enough users find it useful and the operating costs remain manageable.
The post Remove AI Slop From LinkedIn Using Browser Extension appeared first on ProPakistani.
