Koncile detects tampering in PDF and image files, and documents generated or edited by AI, then returns a fraud score from 0 to 100: stop fake documents before they enter your pipeline.
Create a free account to run the analysis.
Create a free accountFor document volumes or an API, talk to an expert.
The proof of address checker used by insurers, banks and large companies
4.8 / 5
Utility bills, leases, bank statements and insurance documents: the fake utility bill detector runs five levels of checks. A replaced address or a newer date leaves traces, either in the pixels or in the computer file.
Con Edison, PG&E or Comcast Xfinity each produce bills through their own billing and print chain: same software, same fonts, same way of writing the PDF. Koncile compares your document with known bills from the same issuer. A PG&E bill made in Word or Canva looks like no real one.
The most common fake is a real bill where only the name or the address changed. An edited PDF often keeps its earlier versions: Koncile rebuilds them and shows the before and the after, old address included, even when the layout is perfect.
On a scan or a photo, K-Detect, our image analysis model, isolates edited areas: a white patch under the address, text laid over the original, cloned digits. It also flags proofs of address generated in whole or in part by AI from a real bill.
AI detected82/100The address on the bill can be checked against USPS reference data: does it exist, is it residential, is it a P.O. box or a commercial mail drop rather than a home? It is then matched with the address on the driver's license and the rest of the file.
A bill made more recent often keeps its old due date or service period. Koncile compares every date, checks recency against your own policy and matches the account holder and the service address with the ID document and the other documents in the file.
Without a tool, these three checks stop clumsy fakes in a few minutes. A tip: ask for the PDF downloaded from the utility's online account, not a screenshot or a scan, so you can read its metadata.
Statement date, service period, meter read dates and due date must follow each other and the utility's billing cycle. Fraudsters often fix the date at the top and forget the due date at the bottom. Then check that usage, rates, fees and taxes add up to the amount due.
Name and address must match the driver's license or state ID, apartment number included. The service address proves residence, not the mailing address, and a P.O. box is not a street address. An account number in a format the utility does not use should raise a question.
Use our free PDF metadata viewer and look in particular at Creator, Producer and the timestamps. A utility bill comes out of an industrial billing chain. Word, Canva or a PDF editor are often red flags, as is a file created after the statement date.
Koncile's fraud detection API gives you a score from 0 to 100 and detailed explanations of manipulation risks, right inside your own tools.
Utility, account holder, service address, 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 information system.
curl -X POST https://api.koncile.ai/v1/upload_file \ -H "Authorization: Bearer $API_KEY" \ -F "files=@utility-bill.pdf"requests.post("https://api.koncile.ai/v1/upload_file/", headers={"Authorization": f"Bearer {API_KEY}"}, files=[("files", open("utility-bill.pdf", "rb"))]){ "status": "DONE", "General_fields": { "Utility": {"value": "Con Edison", "confidence_score": 0.99}, "Service_address": { "value": "245 Example Ave, Apt 4B, New York, NY 10001", "confidence_score": 0.97 } }, "Fraud_detection": { "risk_score": 86, "summary": "PDF edited after issue: service address changed", "signals": [ {"name": "Traces of edits", "severity": "high"}, {"name": "Editing software", "severity": "medium"} ] }}Wherever a proof of address ties a person to a home, a fake opens the door to a synthetic identity, a lower insurance rate or a lease under a false name. Here is where the checker fits in.
The proof of address is read with the ID document, pay stubs and bank details. Koncile checks that it is in the customer's name, then cross checks the address with the one on the ID.
The proof of address comes with the proof of income. An address that changes from one document to the next, or an edited bill, is flagged before approval, not after the first missed payment.
The home or garaging address drives the rate. At underwriting as at claims, the proof of address is checked before validation, with a score the handler can explain.
Leases, bills and pay stubs arrive by the dozen in each application. The check runs on receipt, before the applicant is reviewed, and every document keeps the detail of its tests.
At onboarding, the address is backed up in seconds and at high volume. The detail of every test is kept, for internal audit and for the regulator alike.
Patient registration and coverage enrollment often ask for proof of residence in the state or the service area. The bill is checked before the file moves on.
An address replaced on a real bill often leaves no visible error: logo, layout and amounts stay in place. The fake utility bill detector relies on signals invisible to the naked eye: comparison with the utility's usual bills, and traces left by PDF editors and by AI.
| Element of the bill | What Koncile checks | What gives an edit away |
|---|---|---|
| Issuing software | Creator and Producer compared with the utility's billing chain | Word, Canva, a PDF editor or an online template instead of the utility's chain |
| File dates | Creation and modification dates, compared with the statement date | A file created or edited after the statement date, rewritten or missing dates |
| Saved versions | Versions appended at the end of the file, dormant objects | The old address or name stays in the file: Koncile rebuilds the before and after |
| Fonts | Embedded fonts, subsets and encoding, compared with bills from the same utility | An address retyped in another font, or embedded differently from the rest |
| File writing style | Object order, cross reference table, stream compression | pypdf, MuPDF or a Python script instead of the utility's chain |
| Text layer | Match between the displayed text and the encoded text | Visible text that differs from the extracted text, an old address still in the layer |
| Bill dates | Statement date, service period, meter read dates and due date | A newer statement date that no longer fits the service period or the due date |
| Account number and totals | Account number format, usage, rates, fees and taxes recalculated | A number in the wrong format, a total that does not add up |
| Holder and service address | Name and service address compared with the ID document and the rest of the file | An address that changes from one document to the next, a P.O. box or a mail drop |
| Image | On a scan or a photo: edited areas and AI generated parts, with K-Detect | A white patch under the address, text laid over the original image |
A utility bill holds a home address, account numbers and usage that shows when a home is occupied. 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. Bills are encrypted, deleted after analysis and never used to train a model.
It is a tool that combines several technologies to spot the anomalies that give away an edit or a fraud: machine learning that analyses the image, and analysis of the PDF file, which is written like code. The goal is to stop the few percent of fakes that slip into your onboarding, lending or leasing processes.
You never know for certain: you measure a risk. It becomes high when one fraud test goes over its threshold, or when several tests point the same way: an address edited in the image, metadata that does not match the utility, an earlier version hidden in the file. That body of evidence is what should raise the alarm.
It can help, mainly to spot inconsistencies in context. It lacks three things: models specialised in images edited or generated by AI, knowledge of the traces each editing tool leaves, and a base of each utility's own metadata, which brings the mismatches out. A language model is only one part of the machine.
It depends on your volume. The price combines a base subscription and a cost per page or per document processed. The plans are detailed on the pricing page.
It depends on who asks. Banks must collect a residential or business street address under the Customer Identification Program rule (31 CFR 1020.220), and many ask for a recent utility bill, lease or bank statement. For a REAL ID, states require two documents showing your principal residence (6 CFR 37.11). A P.O. box is not a street address.
Often, yes: banks, lenders and property managers commonly accept a recent statement showing the name and address. The same checks apply, recency, name and address, and the same file analysis, since a statement is a PDF that can be edited like a bill. Our bank statement checker adds the specific checks on balances and transactions.
Yes. Forgery and fraud are crimes under state law across the US. When a forged bill supports a loan or credit application to an FDIC insured bank, 18 U.S.C. § 1014 applies: up to 30 years in prison and a $1,000,000 fine. If you receive a fake, keep the original file, the analysis report and the date of receipt.
It does not prove the person lives at the address: an untouched bill from someone else stays authentic. On a rescan or a photo, the metadata is lost and only the image is judged. Our issuer base is deepest on European providers, so a US utility is compared with the bills already seen. A high score is a signal to verify.
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.
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
