Forensic, AI-aware document fraud detection — built for risk engines, not onboarding funnels.
One POST returns AI-generation indicators, tampering regions, metadata drift, math reconciliation, and a verdict — designed for risk engines, not onboarding funnels.
Inspects EXIF, software signatures, edit history, and structural fingerprints.
Originals are processed in encrypted memory and removed after analysis. Reports stay redacted by default.
Most APIs sold as 'document verification' are identity-onboarding products — selfie capture, photo-of-ID matching, government-database lookups. They were built before generative AI made convincing fake bank statements, paystubs, invoices, and contracts trivial to produce, and they return identity decisions, not document-level evidence.
TrueDoc is the other half of the stack: a document fraud detection API. Every verdict comes with structured forensic evidence — AI-generation signals, tampering regions, MRZ checks, math reconciliation, metadata drift — that a risk engine can act on. Run it alongside Plaid, Persona, Jumio, or your internal KYC; it does not replace them.
AI-generated IDs and financial documents
PDF tampering and incremental edits
Recycled real documents across accounts
Synthetic identities engineered for fraud rings
X-API-Key header on every request — keys managed in the enterprise portal.
Send the document (≤10MB) plus optional metadata.
Trust score, findings, evidence, and explainability — synchronous or webhook.
Verdict and evidence stored in the enterprise portal for review and audit.
TrueDoc's document verification API is purpose-built for fraud detection: every /verify call runs forensic checks, structural inspection, and AI-generation analysis in a single pass, and returns the underlying signals — not just an opaque score.
Fraud teams use the document verification API to catch fake documents, tampered PDFs, AI-generated IDs, edited bank statements, and synthetic invoices before they hit a downstream decision. The same fake document detection API surface powers KYC, underwriting, claims, and marketplace trust workflows.
Each verification API call returns a normalized document trust score plus a structured verification report: ordered findings, per-field evidence, regions of interest on suspect areas, and an exportable PDF report for human review or audit.
The same document trust score API surface is available for engineering teams that just want one normalized number to feed into decisioning, with the full verification report kept on the side for reviewers and regulators.
Fintech and lender KYC: verify IDs, address proofs, paystubs, and bank statements inline during onboarding; auto-accept clean verdicts and route high-risk to manual review with evidence pre-attached.
HR and employment screening: batch-verify offer letters, paystubs, and education documents via the document verification API, with webhook callbacks back into your ATS.
Rental and tenant screening: verify income docs and IDs from applicants and partner platforms; signed webhooks feed verdicts into your decisioning UI.
Insurance claims: verify supporting documents on submission; flag tampered receipts, edited estimates, and AI-generated proof images before payout.
Marketplaces and platforms: re-verify seller and partner documents at trust-tier transitions, with webhooks driving automatic tier changes.
A trust score alone is not enough for production fraud or KYC decisions. Modern engineering teams need machine-readable evidence per field — MRZ checksum results, math reconciliation on financial documents, ELA regions on suspected tampering, font and metadata anomalies, and AI-generation signatures.
TrueDoc's /verify endpoint returns a structured verdict object: trust score, ordered findings, evidence map, recommended action, and a stable verdict ID that links to the immutable record stored in the enterprise portal. Every call is replay-safe and audit-ready.
No credit card. Redacted report in under a minute.