About AuthorText
AuthorText looks at a text and tells you how many signs of machine writing it finds, and where. It does not tell you who wrote it. That difference matters, and most of this page is about it.
01/What it does
You paste a text, give a link or upload a document. A few seconds later you get a score from 0 to 100, a bar for each separate signal, and quotes from the places in the text that raised the score.
The same page can clean a text. It removes hidden characters, such as zero-width spaces, and replaces fancy typography with plain marks. Texts in Russian, Ukrainian, Portuguese and English work the same way.
02/What it does not do
The score is a hint about style. It is not proof of authorship, and no program can give you proof. Polished human writing can score high. Good machine text can pass without a flag.
If a decision about a person hangs on a result, such as a grade, a job or a contract, use the evidence places as a reason to talk to the author. Nothing more.
03/Who is behind it
AuthorText is a hobby project run by one person who lives in Portugal. It is not a company. There is no firm name, no registration number and no office, and this page will not pretend otherwise.
The person who runs the site is also the one who answers the mail and fixes what breaks. Replies can take a day or two.
04/Open source
The analyzer, the cleaning tools and the code of the site are public. You can read how a score is made instead of taking our word for it, and you can tell us where we are wrong. The repository is on GitHub.
05/How to reach us
Write to [email protected]. Questions about the score, a text that was judged unfairly, a bug, a payment problem, a request to delete an account: all of it goes to the same address. Please do not send passwords or other people's personal data.
MethodFAQSource code on GitHub
Contact: [email protected]