Less artificial, more smart
Bullshit Detection was built to give writers and readers a fast way to analyze text for the patterns used — and overused — by AI and by people writing to win attention online. The aim is simple: raise awareness of rhetoric so we can raise the bar on writing and think more clearly about what we read.
Why focus on rhetoric?
Communication is inseparable from persuasion. Aristotle knew it: “It is not sufficient to know what one ought to say, but one must also know how to say it.” Because rhetoric is woven into communication, understanding how writing persuades — on purpose or by accident — is fundamental to understanding communication itself.
Modern digital communication is also fiercely competitive. Writers and content creators face enormous pressure to capture attention, so rhetorical devices and other attention-winning tricks pervade what we read. Content that uses rhetoric well is more likely to be seen — which is exactly why more and more of it, written by people and by AI, leans on the same handful of moves.
How to read the Rhetorical Score
The Rhetorical Score is an arbitrary calculation, and it’s meant to be. Some patterns weigh more heavily than others — “It’s not X. It’s Y” pushes the score up sharply; a stray em dash barely moves it. Think of the score as a rough estimate of how strongly an audience is likely to feel pushed or pulled by the writing.
A higher score means a stronger reaction is likely — positive or negative. Readers who already agree may feel the writing is preaching to the choir; skeptical readers, trying to judge value quickly, are more likely to bristle. Use the score as a flag, and ask the only question that matters: how will this land with my audience?
The underlying idea
The working hypothesis behind Bullshit Detection: people dismiss communication when they sense it’s built to persuade, or when it reads as low-value — and AI writing can feel low-value precisely because it’s so easy to produce. Persuasion and low-value may be two sides of the same coin. If an idea is genuinely valuable, does it need to be sold, or will it sell itself?
Either way, readers today have little choice but to filter a flood of content. Using “tells” — overused and rhetorical writing patterns — is one way to filter faster. By noticing those tells, the astute reader and writer can produce better writing and sharpen their own thinking.
Who made this
Bullshit Detection is the re-platformed home of a project originally published as Un•AI•ify, created by Justin Owings — a marketer, tinkerer, and student of influence. It’s an independent tool, and it’s still early.
Placeholder: a public contact address for Bullshit Detection hasn’t been set yet. Provide the address you’d like listed and it will appear here.
A note on limits
AI is trained on writing by people — and increasingly, life imitates the artificial. Non-AI writing can and will be mistaken for AI-generated writing, and vice versa. Bullshit Detection is a tool for awareness, not a verdict. Pattern-matching produces false positives and false negatives; treat every result as a flag worth a second look, nothing more.