Fabricated supporting documents are the fastest-growing fraud in visa, benefits and licensing. The hard part is not catching them. It is catching them without making life worse for the overwhelming majority who are honest.
One person can generate a thousand plausible supporting documents in an afternoon. Assessing a thousand takes an assessor weeks.
Tightening document checks lengthens processing for everyone. The applicant who suffers most is the one with nothing to hide and a deadline.
An adverse decision has to be defensible on review. "It looked wrong" is not a finding; it is an opinion that will not survive a tribunal.
Applicants capture supporting documents through a sealed flow. The provenance travels with the file rather than being asserted separately.
Every verdict is a signed proof object: the inputs, the checks, and what each check measured. A decision can be re-explained years later with the evidence that produced it.
A tribunal or an integrity unit can re-run the verification themselves, offline, without asking us and without taking our word for it.
CapSeal does not replace your assessment process - it adds two touchpoints to the one you already run. Documents are sealed at lodgement, a verdict attaches at assessment, and each file routes down one of three paths. Every outcome lands back in your case system with the proof stored on the file.
Note the middle path. An application lodged on paper, through an agent, or from somewhere the app is unavailable routes to normal officer assessment exactly as it does today - never auto-refused. Absence of provenance is not evidence of anything, and a program that treated it as adverse would fail on review.
CapSeal establishes where a piece of evidence came from, and produces a record of that finding which survives review.
The full position is in the Product Disclosure. We would rather you knew the boundary before you bought than after.
A bottom-up model for a program processing 120,000 applications a year, anchored to the average claim actually observed in Australia's largest refund-fraud event. Eight pages, every assumption on the page.
The honest boundary, stated in the brief itself: the model covers fabricated supporting documents, the category capture-time provenance directly prevents, on the share of a program carrying direct payment risk. It does not claim identity fraud that uses genuine documents, eligibility misstatement, or organised schemes that never submit a document. The figures are modelled from published ranges and ATO and DHA reporting, not results from a named program.
Try the verification half right now, with no signup, at capseal.ai/check.
We work with integrity teams on a bounded pilot against a single evidence type.
Book a pilot Check a file first Public-sector brief (PDF)