Koncile detects tampering in PDF and image files, and documents generated or edited by AI, then returns a fraud score from 0 to 100: hotel bookings, tickets, medical bills abroad and police reports checked before a travel claim is paid.
Create a free account to run the analysis.
Create a free accountFor document volumes or an API, talk to an expert.
The document checker used by insurers, banks and large companies
4.8 / 5
Check the booking against its issuer: a confirmation number the hotel or the platform can find, dates and amounts that agree with the tickets and the payment, a guest name that matches the traveler. Then check the file, which the eye cannot: the software that produced the PDF, its dates and its editing traces.
Hotel bookings, tickets, itineraries and documents from abroad: the checker runs four levels of checks. A changed date or amount leaves traces, either in the computer file or in the pixels.
Airlines, hotel chains and booking platforms send confirmations from their own systems. Koncile reads which software produced the file. A word processor, a PDF editor or a dummy booking generator is not how Delta or Booking.com issue a confirmation.
The most common edit is a real booking with other dates or a higher amount. An edited PDF often keeps its earlier versions: Koncile rebuilds them and shows the before and the after, even when the layout is perfect.
On a scan, a photo or a screenshot, K-Detect, our image analysis model, isolates edited zones: a total retyped over the original, a date changed on a medical bill. It also flags documents and damage photos generated in whole or in part by AI.
Booking references and ticket numbers must follow their formats, hotel nights must fit the flights, and amounts must add up and match the payment. Koncile flags every mismatch; your team confirms the booking with the issuer.
A trip cancellation, a medical emergency abroad or a lost bag rarely comes with a single document. Start from the file, not from one ticket: the fraud usually shows in what the documents say to each other.
They prove that the trip was booked and paid, for which dates and which amount. The insurer reimburses what was paid and not refunded, so the dates, the amount and the traveler's name are the fields most often edited.
A cancellation confirmation from the airline or the hotel, and the refund or credit that followed. A claim that ignores a refund, or a cancellation dated before the event that supposedly caused it, deserves a question.
Issued in another country, often in another language, and sent as photos. They carry the largest amounts. In the UK, the ABI's 2025 figures include a traveler who filed more than £300,000 of bogus medical emergency claims with forged medical paperwork.
Damaged luggage, a medical receipt photographed on a table, a screenshot of a booking page. Images arrive from phones and messaging apps, and some have been edited, including by AI: ALFA now lists edited photos among document fraud.
The first page of Google for fake hotel booking sells dummy reservations for visa applications. This page is about the other side: catching those documents, and their edited cousins, when they reach a claims or an expense team.
A reservation held for a few days and never paid, or a confirmation produced by a dummy booking website. In a cancellation claim, it shows as a booking with no payment behind it. Ask for the payment proof, and look at who produced the file.
A real booking where the dates, the amount or the traveler's name were changed in a PDF editor, to inflate a cancellation claim or fit the dates of a medical event. The old values often stay inside the file, and the file is rewritten after its issue date.
A screenshot of a checkout page or a trip summary proves that a trip was looked at, not that it was paid. Ask for the confirmation email or the airline's receipt as a PDF, which carries the ticket number and the file data the checker reads.
This page is about detection only. Koncile's checker is for businesses verifying the documents they receive, never for producing one.
Without a tool, these three checks stop clumsy fakes. Ask for the original confirmation emails and receipts as PDFs, not screenshots.
Call the hotel, or use the airline's manage booking page with the booking reference and the traveler's surname. An airline booking reference has six letters and digits. A reference nobody can find is the strongest signal there is.
Flights, hotel nights, cancellation date and the event that caused it must fit together, and fall within the policy dates. The amount claimed must match the payment, minus any refund or credit.
Use our free PDF metadata viewer and look at Creator, Producer and the timestamps. Confirmations come out of booking systems; a word processor, a PDF editor or a file created after the booking date are red flags.
Koncile's fraud detection API gives you a score from 0 to 100 and detailed explanations of manipulation risks, right inside your claims system.
Issuer, traveler, booking reference, ticket number, dates and amounts, ready to cross check.
High, medium or low level, with a one sentence explanation.
High risk from 80.
A one line summary ready for your claims system.
curl -X POST https://api.koncile.ai/v1/upload_file \ -H "Authorization: Bearer $API_KEY" \ -F "files=@hotel-booking.pdf"requests.post("https://api.koncile.ai/v1/upload_file/", headers={"Authorization": f"Bearer {API_KEY}"}, files=[("files", open("hotel-booking.pdf", "rb"))]){ "status": "DONE", "General_fields": { "Document_type": { "value": "Hotel booking confirmation", "confidence_score": 0.99 }, "Stay_dates": {"value": "2026-07-12 to 2026-07-19", "confidence_score": 0.97} }, "Fraud_detection": { "risk_score": 85, "summary": "Confirmation produced by a dummy booking site, reference format invalid", "signals": [ {"name": "Generator fingerprint", "severity": "high"}, {"name": "Inconsistent dates", "severity": "medium"} ] }}Wherever a booking or a ticket turns into a payment, a fake costs money. Here is where the checker fits in.
Trip cancellation, medical expenses abroad, baggage: every document of the claim is checked on receipt, from the booking to the medical bill and the photos, before the claim is paid.
Medical bills, receipts and records sent from abroad, often as photos from a phone. Extraction and fraud checks run in a single pass, so handlers only open the files that need them.
Hotel folios, airline receipts and taxi receipts in expense reports: an edited amount or a duplicate stay is flagged before reimbursement.
There is no reliable public figure on forged travel documents alone. Here are the closest sourced figures, from UK and French insurers, and from our own analyses.
UK insurers report more claims backed by AI generated images and doctored documents, and ALFA lists photos edited by AI among French document fraud. Insurers spot many of these photos by reading their metadata. A travel claim concentrates every risk at once: documents from abroad, photos sent from a phone, amounts in several currencies. We do not publish a travel specific rate of our own yet.
Source: ABI figures reported by Insurance Business UK, October 2026; ALFA, 2025 activity report published in July 2026; JeChange, October 2026; Koncile data 2026, all documents
A confirmation with a changed date or amount often shows no visible error. The checker relies on signals inside the file and the image, then on the consistency of the documents in the claim.
| Element of the document | What Koncile checks | What gives a fake away |
|---|---|---|
| Issuing software | Creator and Producer, compared with how a booking system or an airline produces a confirmation | A word processor, a consumer PDF editor or a dummy booking generator |
| File dates | Creation and modification dates, compared with the booking date | A file created or edited after the booking, or after the trip |
| File writing style | Object order, cross reference table, stream compression, saved versions | An earlier version with other dates or another amount |
| File nature | Native PDF, scan, photo, screenshot or hybrid file | A screenshot where a confirmation is expected, an image pasted into a PDF |
| Image and photo metadata | On a scan or a photo: edited zones and AI generated parts, with K-Detect, and camera metadata when present | A total retyped over the original, a damage photo generated or edited by AI |
| References, dates and amounts | Booking reference and ticket number formats, dates across documents, totals recalculated | A reference in an impossible format, hotel nights outside the flight dates, a total that does not add up |
| External sources | None: Koncile does not query airline or hotel reservation systems | A booking the hotel or the airline cannot find, confirmed by your team |
A travel claim holds passport details, travel dates, medical bills and card numbers. Founded by a former lawyer, Koncile was built with data protection at the core of its architecture: third party SOC 2 audit, GDPR compliance, secure hosting. Documents are encrypted, deleted after analysis and never used to train a model.
Confirm it with the hotel or the platform using the confirmation number and the guest's surname. Check that the dates fit the flights and that the amount matches the payment. Then check the file: the software that produced it, its dates and its editing traces. Koncile runs the file checks in about ten seconds.
They check that the trip was booked and paid, that the event happened within the policy dates, and that the amounts match what was paid and not refunded. They confirm bookings with airlines and hotels when needed, and more and more of them check the documents and photos themselves for edits and AI generated content.
A dummy booking is held without payment, or produced by a dummy booking website. It shows as a reservation with no payment proof, a reference the hotel or the airline cannot find, and a file produced by a tool no booking system uses. Ask for the payment proof and the original confirmation email.
It depends on the cover, but most claims ask for the booking confirmation and tickets, the payment proof, the cancellation and refund proof, and evidence of the event: a medical certificate or bill, a police report for a theft, an airline report for a lost bag, photos. Each of them can be checked for edits.
An airline ticket number has 13 digits, and the first three identify the airline that issued it. It should appear on the airline's receipt, next to a six character booking reference. The airline can confirm both with the passenger's surname.
Often, yes. Image analysis finds edited zones and AI generated areas, and camera metadata, when it is still there, says when and with what device a photo was taken. Photos sent through messaging apps often lose that metadata, so the image itself carries the check.
It does not query airline or hotel reservation systems, so it cannot confirm that a booking exists. A genuine booking cancelled after issue stays authentic. Screenshots and forwarded photos lose their metadata. A high score is a signal to verify with the issuer, never a verdict.
Yes. Koncile complies with SOC 2 and GDPR, and an independent firm certifies our security and data protection. Your documents are hosted securely with a provider that itself holds SOC 2 and ISO 27001 certifications, and are deleted after analysis.
Each document gets its own method: a proprietary ML model for photos, known metadata and forensics for native PDFs, and contextual cross checks.
Inflated balances, deleted lines, a PDF rewritten with software other than the bank's.
See the detection
Net pay changed, contributions that do not match gross pay, a layout that does not match the payroll software.
See the detection
Photo replaced, date of birth or number changed, an MRZ that no longer matches the fields.
See the detection
Edited amounts, wrong VAT, duplicates and made up suppliers.
See the detectionUpload your bank statements, pay stubs or IDs: Koncile extracts the data and flags fakes in seconds.
Talk to an expert
