Reading profoundly shapes thinking, and writing distills ideas — but readers are exasperated by the surge of LLM-authored content. The tells are unmistakable: structural patterns, grating phrases like "and here's why that framing matters!", and a lack of authentic voice. Readers can absolutely detect LLM writing, and they emphatically care. Cynthia Dunlop's survey of 668 developers found 78% stop reading immediately upon detection, 71% avoid the author in the future, and 98% prefer an author's imperfect writing over LLM-polished prose. These respondents are active readers and tastemakers who amplify content they value.

Using LLMs voids the social contract between writer and reader. Readers shouldn't labor to understand sentences the writer didn't work to create. If the purpose of public writing is to serve the reader, LLM use is becoming counterproductive — repelling the very audience writers seek.

This mirrors the early 2000s email spam crisis. Improved filtering eventually undermined spam economics, and being labeled spam became catastrophic for legitimate senders. Similarly, as detection improves, LLM-authored content will face increasing rejection. Writers who value reach and trust must write for themselves — the authentic, imperfect way.