Completeness evidence for sponsor banks

Every record, accounted for.

Sponsor banks answer for AML controls their fintech partners run. Ellipe 3 gives them verifiable proof that every partner transaction reached monitoring, instead of a sample and a spreadsheet.

01 The problem

You carry the liability. Someone else runs the system.

The sponsor bank answers for BSA/AML controls its fintech partners run. When examined, the evidence on hand is screenshots, spreadsheets and attestations, produced by the party under examination, after the question was asked.

  • Jan 2024 · OCC

    Consent order

    A sponsor bank, over BSA/AML and third-party risk failures in its fintech partnerships.

  • Jun 2024 · Federal Reserve

    Cease-and-desist

    A sponsor bank, over AML, OFAC and risk-management gaps in its fintech partner program.

In both cases, the bank was the named party, not the fintech.

120+

US banking enforcement actions, fines and suits in 2024.

$61B

Annual financial-crime compliance spend across the US and Canada.

Sources: OCC and Federal Reserve enforcement releases (2024); American Banker / OCC (2024); LexisNexis Risk Solutions, True Cost of Financial Crime Compliance.

02 The blind spot

Controls rarely fail. Populations do.

Most failures involve records that never reached the control. A screening rule can run perfectly on the wrong population. Sampling can’t catch this, because the sample is drawn from the records that already made it in.

  • Unwired

    A product line never joined to the feed.

  • Parallel ledger

    A partner channel nobody connected.

  • Migration gap

    A date range silently dropped.

  • Changed predicate

    A status code whose meaning shifted.

  • Late arrivals

    Corrections that landed after the window closed.

Ellipe 3 makes the population itself the thing being checked.

03 From the field

What practitioners told us.

From our discovery conversations with people who run, test and examine these programs. Names withheld.

  • "I typically only test against what the partner supplies… I have to rely on the partner to provide me that record."

    Compliance assessor

  • Asked whether counts and sums ever tie out while individual transactions are missing: "Absolutely. That's why you have a team reconciling the individual transaction level to catch those things… at least that is the hope!"

    Bank reconciliation specialist

  • Asked whether anything checks that reconciled transactions reached the bank's monitoring: "If so, it's on their side. And I don't have granular visibility to what they do with the file."

    Payments operations leader

  • Examiners pushed hardest on "receiving full transparency on transaction details."

    Former bank BSA Officer

04 How VCEP works

Follow every record from source to outcome.

An examiner asks three questions about the records behind a control. Each layer of VCEP answers one, and each produces evidence the examiner can check with a standalone verifier, offline, without trusting the bank’s pipeline or us.

  • EVP-1Working

    Population coverage

    Did every record from the accepted source reach the control, or get excluded with a counted reason?

  • EVP-2Prototype

    Routing and attribution

    As records move through each step, does every one end up somewhere declared, under a stated rule?

  • EVP-3Active R&D

    Execution

    Did the bank’s or vendor’s own monitoring system actually run over the complete population?

    Banks won’t hand their monitoring engines to a vendor, so the evidence has to come from where the computation already runs. This is our core research problem.

Each layer extends the same evidence, so an EVP-1 result stays valid as the others are added.

Patent pending: two provisional applications filed.

05 Proof

Tested on 5 million records.

5,078,345

Records read, from IBM Research’s AMLworld synthetic AML dataset.

18 / 18

Daily epochs verified.

10 / 10

Tamper cases caught: dropped rows, altered totals, broken bindings.

From our EVP-1 proof of concept, built in Rust with a standalone verifier. See the evidence packet and verifier checks.

06 Pricing

How banks work with us.

  • For sponsor banks

    Sponsor bank license

    Starting at $60,000 / year

    per sponsor bank, plus $12,000 / year per fintech program verified

    Annual platform license per bank, plus a fee for each fintech program verified.

  • For fintechs

    Fintech partner tier

    Starting at $1,500 / month

    per fintech partner

    Fintechs produce completeness evidence for their sponsor bank, and become easier to keep as a partner.

  • To start

    Paid design-partner pilot

    Fixed fee, 90 days

    on one fintech program

    A fixed-fee 90-day pilot on one fintech program, converting to the annual license.

Indicative pricing.

Built for BSA/AML at sponsor banks first. The same evidence applies wherever a control depends on a complete population, such as insurance claims and healthcare billing.

07 Status

Status and support.

Regulatory

In conversation with the NYDFS Research and Innovation Division.

IP

Patent pending: two provisional applications filed.

Support

Lambda Research Grant

What’s built and what’s in progress is detailed on the technical page.

08 Team

Team

  • Kevin Mangroo

    Founder & CEO

    Cryptographic systems, RegTech infrastructure and BSA/AML compliance. Sole inventor on the VCEP work.

Based in NYC.

See it on your own controls.

For sponsor banks and fintechs that want completeness evidence on their BSA/AML data. 20-minute walkthrough, no prep needed.