Spot AI writing and heavy rhetoric — before it spots you
Bullshit Detection flags the writing patterns that generative AI and attention-hungry content lean on most. Paste any text to see the tells, gauge how hard it’s working to persuade you, and learn to write with a sharper, smarter voice.
0 Rhetorical Score
A higher score means readers are more likely to feel pushed or pulled by the writing. It’s an arbitrary flag, not a verdict.
Highlighted text
How to use this tool
Paste copy, content, or any writing into the box above. Bullshit Detection scans it for rhetoric that’s overused online — especially by generative AI — and reports a count for each pattern, a readability estimate, and a single Rhetorical Score.
Nothing is uploaded or stored: the analysis runs entirely in your browser. Use it on your own drafts before you publish, or on content you’re reading to see how hard it’s working to move you.
Why focus on rhetoric?
Communication is inseparable from persuasion — Aristotle noted that knowing what to say isn’t enough; you must know how to say it. Online, where content competes for scarce attention, rhetorical shortcuts are everywhere, in writing by people and AI alike.
Readers increasingly dismiss content that feels engineered to persuade. Spotting the tells helps you filter faster — and helps you avoid them in your own writing. Read more about the idea.
The seven patterns we look for
Each is a rhetorical or stylistic habit that shows up constantly in AI-generated text. Open any one to read how it works and why it matters.
- It's not X. It's Y Negation followed by affirmation — a setup-and-swap that reframes the point.
- Lady doth protest Extra words of enthusiasm — usually adverbs — that weaken rather than strengthen.
- But reverse! The pivot word that tells the reader to discard everything before it.
- Clichés Ready-made phrases that betray a lack of original thought.
- Buzzwords Business jargon that gestures at meaning instead of demonstrating it.
- Triads Three things in a row: rhythmic, memorable, and easy to overuse.
- Dashes Em and en dashes have their place — but frequency and consistency matter.
Get early access to new tools
Bullshit Detection is just getting started. Subscribe for new detectors, and clear-eyed writing about influence, persuasion, and the flood of AI content online.
No spam — only new tools and occasional writing on modern communication.