AI Writing Analyzer

Does this read human or machine?

Paste any company-produced text — a privacy policy, support doc, marketing page, or blog post — and get an honest, signal-based read on how consistently it matches AI-generated writing, plus the specific evidence behind the assessment. No account, no storage.

Read the score before you scan

Higher scores mean the text is more consistent with AI-written output. FairPrint checks style, repetition, specificity, voice, and disclosure signals — useful evidence, not a lie detector.

RhythmTemplatesSpecificityDisclosure
Helps the result route cleanly into certification or public nomination.
Import below can fetch readable public HTML/text from this URL.
Import from a public URL
Use the source URL above to pull public policy, support, or marketing text into the analyzer.
Paste or import the text. You need at least 120 characters. The analyzer reviews style and disclosure signals — it is not a lie detector and cannot read intention.
What we look for
Uniform rhythm & fluency
Repetitive phrasing & templates
Hedged, generic claims
Lack of concrete specifics
Flat / single-voice tone
Missing AI-use disclosure
How to read the score

Each signal is scored 0–10. A higher score means the writing is more consistent with AI-written output. The overall 0–100 score averages the six signals into a low / moderate / high confidence band. Some content is intentionally AI-assisted yet disclosed — that is not deception.

Why FairPrint built this

Undisclosed AI-generated legal and marketing copy is a transparency concern. The FTC’s AI-accuracy policy efforts put the spotlight on hiding AI use from the people it affects. FairPrint is about disclosure — this tool helps you spot writing that may not own its AI origins.