Use case

Detect AI-written text in submissions

The problem

Cover letters, applications, product reviews, guest posts — when the writing itself is the signal, undisclosed AI authorship changes what it tells you.

Most detectors return a single confident score with no reasoning, and false-positive heavily on templated or non-native writing.

How Stipple helps

Stipple returns a probability and the specific linguistic tells behind it, so a human can weigh the evidence. It abstains on forms, tables, and scans — cases where style detection is statistically unreliable — instead of guessing.

The probability is explicitly a confidence, not a verdict; the output says so.

The call

bash

curl -X POST https://www.stipple.sh/v1/detect-ai-text \
  -H "Content-Type: application/json" \
  -d '{"text": "The submitted essay text..."}'
# → { "probability": 0.92, "lean": "ai", "tells": [...], "applicable": true }

Use it as a triage signal in review queues — flag, don’t auto-reject.

Try the AI detector in the browser Read the docs