AI Text Detection

Analyse whether text was written by a human or generated by an LLM such as ChatGPT, Claude, or Gemini. Long text is automatically chunked for accuracy.

POST /ai-text/detect

Authentication

Bearer token in the Authorization header.

Request

Content-Type: application/json

Field	Type	Required	Description
text	string	Required	The text to analyse. Long text is automatically chunked into ~450-token segments.

Response

Field	Type	Description
score	number	Probability the text is AI-generated (0.0 – 1.0).
is_ai	boolean	Whether the text is classified as AI-generated.
verdict	string	Human-readable verdict: "human", "likely_human", "uncertain", "likely_ai", or "ai".

Example

Request

curl -X POST https://api.cacuda.com/ai-text/detect \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "The quick brown fox jumps over the lazy dog."}'

Response 200 OK

{
  "score": 0.12,
  "is_ai": false,
  "verdict": "human"
}

Error Codes

Code	Description
400	Missing or empty text field.
401	Missing or invalid API key.
402	Insufficient credits.
503	Detection service unavailable.

Notes

Text longer than ~450 tokens is automatically split into chunks. Each chunk consumes one credit.

Try it in the scanner

Test this endpoint interactively with real data. No code required.

Open Scanner