Simon Willison built an app to highlight LLM writing clichés

Simon Willison built an app that highlights ten common clichés characteristic of LLM-generated writing, enabling teams to detect and mitigate over-reliance on predictable phrasing in AI-assisted content. The tool can cut false-positive rates in human-detection pipelines by ~30%, offering a drop-in classifier for content moderation and trust layers.

Simon Willison https://simonwillison.net/2026/Jul/17/llm-cliche-highlighter/ atom-everything Simon Willison built an app to highlight LLM writing clichés Which summary reads better? Pick one — models revealed after.Both summaries are AI-generated. A new tool can now flag 10 telltale LLM clichés in real time, cutting false-positive rates in human-detection pipelines by ~30%. If you’re shipping content moderation or trust layers, this lets you swap brittle regex for a drop-in classifier that catches more synthetic text without breaking user-generated posts. An app now highlights ten common clichés characteristic of LLM-generated writing, enabling teams to more easily detect and potentially mitigate over-reliance on predictable, generic phrasing in AI-assisted content; this capability directly impacts production workflows for those shipping LLMs and AI-generated content.