Guide Lists AI Writing Tells and How to Fix Them
Start here
A writer at hrih.si published a guide cataloguing the habits that make text read as chatbot-generated — from "it's not X, it's Y" contrasts to em dash density and bold-label bullets — and pairs each tell with why readers notice it and a before-and-after rewrite. The author says the list is what they edit their own drafts against, and released a free browser-based checker that runs the same list on a draft and marks the flagged lines; the list is also built into a copywriter skill for Claude Code and other agents that checks its own output before handing it over. The guide cites Pew Research counting "it's not just X, it's Y" at 0.87 per 10,000 words of web text in January 2023 versus 2.36 in January 2026, and Reinhart et al. finding instruction-tuned models use present participial clauses at two to five times the human rate. It also warns that stripping one tell can breed others, citing an unreviewed humanizer test in which AI vocabulary flags fell from 70 to 0 while uniform-beat flags rose from 60 to 101.
Generated from reports · updated 4 h ago
Report timeline
Follow the reports to see the event from different sides.
- Hacker News · AISigns of AI writing and how to fix each one
The author compiles a checklist of habits that make text read as chatbot-generated, from "it's not X, it's Y" contrasts to em dash density and bold-label bullets, and pairs each with why readers notice it and a before-and-after rewrite. Cited data includes Pew Research counting "it's not just X, it's Y" at 0.87 per 10,000 words of web text in January 2023 versus 2.36 in January 2026, and Reinhart et al. finding instruction-tuned models use present participial clauses at two to five times the human rate. The piece also warns that stripping one tell breeds others, citing an unreviewed humanizer test where AI vocabulary flags fell from 70 to 0 while uniform-beat flags rose from 60 to 101.
Heat for this event
- Current, comparable range
- 9
- Peak, comparable range
- 10November 20 00:00
- Change in 24 hours
- –
The trend compares the same participants over hours observed continuously, so its range can be narrower than the current heat count. Hover or tap the chart for hourly heat; the arrow keys move through it.