Documentation

ProvenanceAI API docs: verify AI training data with SHA-256 & perceptual hashing, batch verify datasets, and generate compliance reports. Full REST API reference included. Get started in under 5 minutes.

Quick Start

Get started with ProvenanceAI content verification in under 5 minutes.

1 Create Account

Sign up for a free account — no credit card required.

Create Free Account

2 Upload Dataset

Upload your training dataset file. Any format is supported:

  • CSV, JSON, Parquet, TXT files
  • Images (PNG, JPG, WebP)
  • Videos (MP4, AVI)
  • Archives (ZIP, TAR)

3 Generate Verification Report

ProvenanceAI automatically:

  • Generates perceptual fingerprint in memory
  • Discards file content immediately
  • Generates verification report
  • Provides PDF for download

Core Concepts

Perceptual Fingerprinting

ProvenanceAI uses multi-algorithm perceptual hashing (SHA-256, pHash, dHash, aHash) to create unique fingerprints of your content. These fingerprints are collision-resistant — it is computationally infeasible to produce two different files with the same fingerprint.

e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855

Privacy-First Architecture

We never see, store, or have access to your training data. Only the fingerprint, filename, and timestamp are stored. File content is processed in memory and immediately discarded after analysis.

Provenance ID

Each verification gets a unique Provenance ID that serves as a permanent, auditable identifier. This ID can be used to reference the verification record at any time.

PROV-2025-01-15-ABC123DEF456

Timestamps

All events are timestamped using RFC 3339 format, providing a verifiable timeline of your content provenance. Timestamps are cryptographically bound to the verification record.

API Reference

Integrate ProvenanceAI directly into your content workflow or ML pipeline.

Base URL: https://app.provenanceai.network

POST /api/verify Authenticated

Verify a single content file and receive a verification report.

{
  "filename": "training_data_v1.csv",
  "hash": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
  "size_bytes": 1048576,
  "metadata": {
    "model_version": "1.0.0",
    "training_date": "2025-01-15"
  }
}

POST /api/batch-verify

Batch verify up to 10 files in a single request. Returns similarity scores and duplicate detection results for each file.

GET /api/status/:hash

Retrieve the verification status and provenance record for a given content fingerprint hash.

GET /api/statistics

Retrieve system-wide statistics including total verifications, duplicate detection rates, and compliance metrics.

Full interactive API documentation is available in the app.

Open API Docs