# Dashes in text (AI writing tell)

> Long dashes in a text (the em dash, and the en dash outside number ranges) are the best known trace of text written by a language model.

- **Check ID:** A22
- **Area:** [AI Readiness](https://be1st.ai/glossary/ai-readiness)
- **Tool:** [AI Readiness](https://be1st.ai/tools/ai-readiness)
- **Page:** https://be1st.ai/glossary/ai-readiness/dashes-in-text

## Why it matters

The long dash is the clearest fingerprint of text that came out of a model and was never edited. Models set it far more often than people do, because it is over-represented in their training data. To a reader it signals that nobody went through the content, and to a brand it signals carelessness. The character itself is not an error, in English it is typographically correct, which is why this check leaves the score alone and serves as a heads-up.

## In practice

The sentence "Our platform helps companies grow: analytics, reporting and automation" now carries a colon. The version that came out of the model had a long dash there, and the same dash showed up in fifteen more paragraphs of the same page, which is exactly the pattern this check looks for.

## How be1st checks it

The check finds dash characters in the page text (the em dash, and the en dash outside number ranges such as 10–20), the best known trace of text generated by a language model. It scans the title, meta description, og and twitter tags, headings, body text, the alt, title, aria-label and placeholder attributes, values inside JSON-LD schemas, and the llms.txt and llms-full.txt files. For every occurrence it reports which element and which position in the text it sits at, and whether the source writes it directly or as an entity. The check is informational and does not lower the score.

[AI Readiness](https://be1st.ai/tools/ai-readiness), Technical tab.

Every check ends as: Pass, Warning, Fail.

## Sources

- [Em dash](https://www.thepunctuationguide.com/em-dash.html) (The Punctuation Guide)
- [Dashes and hyphens](https://developer.mozilla.org/en-US/docs/MDN/Writing_guidelines/Writing_style_guide) (MDN)

## Related terms

- [llms.txt existence](https://be1st.ai/glossary/ai-readiness/llms-txt-existence): llms.txt is a Markdown file at the root of a website that gives AI models a short overview of the site and links to its most important pages.
- [llms-full.txt](https://be1st.ai/glossary/ai-readiness/llms-full-txt): llms-full.txt is an extended version of llms.txt that holds the full text of a website's key documentation in one Markdown file.
- [Organization schema](https://be1st.ai/glossary/ai-readiness/organization-schema): Organization schema is Schema.org structured data that describes the company behind a website: its name, logo, contact details and social profiles.
- [Product schema](https://be1st.ai/glossary/ai-readiness/product-schema): Product schema is Schema.org structured data that describes a product or service: its name, price, availability and reviews.
- [FAQ schema](https://be1st.ai/glossary/ai-readiness/faq-schema): FAQ schema is Schema.org structured data (FAQPage) that marks up the questions and answers on a page in a machine-readable form.
- [Review / rating schema](https://be1st.ai/glossary/ai-readiness/review-rating-schema): Review and AggregateRating schema are Schema.org structured data that describe the reviews and the average rating of a product or service.