earlyink Record 3/270 confirmed · median 3d early
270 watching first flagged26 Oct 2026

3-D Secure leaks a pre-authorization signal, opening a card-testing channel at the network layer

Audits and fixes for 3DS integrations that reveal card status before authorization. PSPs, issuers and large merchants pay.

Early · 3DS Oracle paper with live tests

Written for: Investor or analyst

Next step

Thesis: 3DS becomes an attack surface and integration security becomes a paid audit for PSPs; it is confirmed if EMVCo issues a bulletin, a network changes 3DS integration guidance, or a PSP reports BIN attacks through 3DS; already building: 3DS server vendors such as Visa's Cardinal Commerce and Netcetera, plus card-fraud platforms.

Why now

Fraud vendors largely treat 3DS as a defence, and this paper shows it can also be an oracle. AI tools are turning years-old bugs into active attacks, as the PHP episode showed, so a protocol-level leak documented in public is unlikely to stay theoretical. The fix sits across EMVCo, 3DS server vendors and merchant integrations, and no one owns the integration audit.

What became possible

The 3DS Oracle paper shows that the client-visible step EMV 3-D Secure requires in browser checkouts can, in some deployments, reveal routing and authentication status before authorization. That lets attackers classify cards at scale (BIN attacks). The authors back this with standards analysis, formal modeling, a documentation audit, disclosure history and a bounded live test.

Two ways in

  1. 3DS integration scanner

    Probe a merchant's checkout and 3DS flow for client-visible classification signals and generate fixes.

    Customer
    Payment service providers onboarding many merchants
    Model
    Per-merchant scan subscription
  2. 3DS-event card-testing detection

    Detect BIN enumeration from patterns in 3DS challenge and frictionless traffic, not just authorizations.

    Customer
    Card issuers and issuer processors
    Model
    Per-transaction fee

Competition

Early

The integration scanner for PSPs looks like the open part. No commercial or open-source tool turned up that probes a merchant's checkout and 3DS flow from outside for client-visible signals (distinct error messages, transStatus or reason codes reaching the browser, timing differences between frictionless, challenge and not-enrolled cases) and then produces fixes.

Full competition notes (8 competitors)

The integration scanner for PSPs looks like the open part. No commercial or open-source tool turned up that probes a merchant's checkout and 3DS flow from outside for client-visible signals (distinct error messages, transStatus or reason codes reaching the browser, timing differences between frictionless, challenge and not-enrolled cases) and then produces fixes. Visa's enumeration-ratio enforcement under VAMP, passed on through acquirers, gives PSPs a reason to run such a check during merchant onboarding. The weakness is that the main fix (one generic failure message, codes only in server logs) is simple and already published by GPayments. So this likely sells as a cheap compliance and onboarding check or a feature in a PSP risk suite, not a standalone platform. The 3DS-event detection idea for issuers is harder to enter. ACS vendors (Outseer, GPayments, Cardinal) and the networks (Visa VAAI, Mastercard) already sit on the 3DS and authorization data, and issuers buy from them. A newcomer would need either a cross-ACS, cross-issuer consortium view that single vendors lack, or a channel play selling detection models through ACS vendors and issuer processors such as FIS. These findings come from a limited set of searches, so smaller vendors may have been missed.

  • GPayments (blog: Where 3D Secure Helps Against Card Testing)

    A 3DS server and ACS vendor. Its October 2026 article describes this exact leak: checkouts that return a different message for each 3DS outcome let attackers find valid cards without authorizing. It recommends one generic error with reason codes kept in logs, and says issuer ACS risk engines are meant to spot near-identical 3DS request streams. The core insight is already public, free advice.

  • Visa Account Attack Intelligence (VAAI) Score

    Network-level, gen-AI enumeration scoring for issuers, launched 2024 in the US. Visa says it scores in about 20 ms and was trained on more than 15B VisaNet transactions. It is presented as authorization-focused, not 3DS-event-focused.

  • Mastercard Safety Net / card testing insights

    Mastercard runs network-level protections such as Safety Net and published a 2026 piece on card testing as a fraud signal. Older coverage of Safety Net: https://mobileidworld.com/mastercard-invests-20-million-in-cyber-security-2132

  • Outseer 3-D Secure ACS

    Commercial issuer ACS with consortium data and real-time risk scoring. It already sees all 3DS traffic for its issuers and is the natural home for 3DS-layer card-testing detection, though nothing found mentions that feature.

  • Cardinal Commerce (Visa) BIN Detection / 3DS

    Visa-owned 3DS platform. Its BIN Detection step collects device data before authentication and is not fraud detection, but Cardinal sits on the 3DS data path any detection product would need.

  • FIS card enumeration white paper

    Issuer processor that notes enumeration is spread across merchants, IPs and devices and is abusing authentication flows. It is a likely in-house builder or buyer for the issuer-side detection idea.

  • cside card testing prevention

    Merchant-side vendor that detects card testing from checkout session behavior before authorization. The nearest product to pre-auth checkout protection, but not specific to 3DS and not an integration audit.

  • Investigation of 3-D Secure's Model for Fraud Detection (arXiv)

    2020 academic analysis of 3DS fraud-detection weaknesses. Prior research exists, but no product was built from it.

Checked 2026-10-26

Risks

  • EMVCo or the big 3DS servers may fix it in a spec update, which leaves no market
  • Only some deployments are affected
  • Fraud platforms can add detection as a rule
  • This is a narrow security fix, not a category

Best on open space; market pull and team fit are the weak point (5).

tech 6 · pull 5 · space 7 · fit 5 · build 6

Exposed

VMAPYPLADYEN.AS

Companies that could benefit or be threatened. News and context, not investment advice.

Momentum of its main topic

102 developments this week +15 vs last week
27 Sep: 0 developments28 Sep: 0 developments29 Sep: 0 developments30 Sep: 0 developments1 Oct: 0 developments2 Oct: 0 developments3 Oct: 0 developments4 Oct: 0 developments5 Oct: 0 developments6 Oct: 0 developments7 Oct: 0 developments8 Oct: 16 developments9 Oct: 15 developments10 Oct: 9 developments11 Oct: 4 developments12 Oct: 9 developments13 Oct: 13 developments14 Oct: 16 developments15 Oct: 18 developments16 Oct: 13 developments17 Oct: 11 developments18 Oct: 6 developments19 Oct: 10 developments20 Oct: 11 developments21 Oct: 18 developments22 Oct: 21 developments23 Oct: 21 developments24 Oct: 16 developments25 Oct: 6 developments26 Oct: 9 developments
Stage: Coverage

Sources

  1. The 3DS Oracle: BIN Attack via the NetworkarXiv
  2. The PHP Meltdown Shows How AI Makes Bugs More ExploitableWired AI

Status history

  • 2026-10-26 Added to the watchlist