unslop

Scan the text for the documented tells such as puffery, promotional adjectives, em dash overuse, and chatbot phrases, then rewrite each instance with plain, concrete wording. Finish with a self-audit asking what still makes the text look machine-generated.

The most frequent tells are em dash overuse, words like delve and pivotal, superficial -ing phrases such as highlighting or showcasing, vague attributions like experts believe, and chatbot closers like I hope this helps. Replace them with direct statements and named sources.

No. The process preserves meaning and matches the intended tone while replacing formulaic phrasing with specific facts and plain words. It also adds voice through opinions, varied rhythm, and concrete detail rather than stripping content.

Deleting patterns alone leaves sterile, voiceless prose, which is itself an AI tell. The fix is to add opinions, vary sentence length, use first person where it fits, and replace abstract claims with specific facts or numbers.

Avoid applying it mechanically to contexts with strict style requirements, such as legal or academic writing with mandated terminology. The rules target generic AI tells, not domain conventions, so judgment is needed where formal register is required.