The Copyleaks AI Content Detection API returns a comprehensive response that includes classification results, probability scores, and detailed explanations of the AI detection analysis. This response structure provides both high-level insights and granular details about detected AI-generated content patterns.
Response Properties The version of the AI detection model used (e.g., “v9.0”).
An array of classification results for different sections of the text. An array of all the classifications. The sections contain their position, classification and the probability for that classification. Show result item properties
Indicates the text classification. Available Options:
1 - Human2 - AI Generated Text The classification confidence. Deprecated we plan to remove this value by July 2026.
Provides information about the match objects that have been found in the text. Describes the text positions for the result (textual version). Positions of the characters in that specific result. Character match start positions.
Positions of the words in that specific result. Words match start positions.
Describes the text positions for the result (html version). Default: null. Character match start positions.
Group identifiers associated with each word.
Words match start positions.
Group identifiers associated with each word.
Aggregated detection results. Indicates the confidence in which this text was written by a human.
Indicates the confidence in which this text was written by AI.
Metadata about the scan. Show scannedDocument properties
The unique scan id provided by you.
Unsigned Integer. Total number of words found in the scanned text.
Unsigned Integer. Number of excluded words in the submitted content.
Overall credits used for the scan.
The creation time of the scan. Timezone: UTC+0.
The submitted url after all http redirects.
Extracted canonical url from the scanned document.
Publication date of the scanned document.
Creation date of the scanned document.
Last modification date of the scanned document.
Scanned document organization.
Scanned document filename.
Example Response { "modelVersion" : "v9.0" , "results" : [ { "classification" : 2 , "probability" : 0.7316979 , "matches" : [ { "text" : { "chars" : { "starts" : [ 0 ], "lengths" : [ 554 ] }, "words" : { "starts" : [ 0 ], "lengths" : [ 73 ] } } } ] } ], "summary" : { "human" : 0.0 , "ai" : 1.0 }, "scannedDocument" : { "scanId" : "scan-id" , "totalWords" : 73 , "totalExcluded" : 0 , "credits" : 1 , "expectedCredits" : 1 , "creationTime" : "2025-08-10T08:33:05.22225Z" }, "explain" : { "patterns" : { "statistics" : { "aiCount" : [ 1.2066389 , 9.673915 , 34.41001 ], "humanCount" : [ 0.18481831 , 0.13894142 , 0.33555666 ], "proportion" : [ 6.5287843 , 69.625854 , 102.54607 ], "source" : [ 1 , 1 , 1 ] }, "text" : { "chars" : { "starts" : [ 10 , 96 , 136 ], "lengths" : [ 25 , 33 , 33 ] }, "words" : { "starts" : [ 1 , 12 , 17 ], "lengths" : [ 4 , 4 , 5 ] } } } } }
Classification Codes Code Classification Description 1 Human Content is likely written by a human 2 AI-generated Content is likely generated by artificial intelligence
Next Steps AI Content Detection Guide Learn how to use the AI Content Detection API to identify AI-generated text.
AI Detection API Reference Explore the complete API reference for AI content detection.
AI Logic Feature Understand how AI Logic provides transparency in AI detection results.
