BNP Paribas Fortis just showed the industry where the game is moving. The winners will be those who connect AI, data, risk, fraud and customer decisions around a single view of every transaction.
A Historic Inflection Point
BNP Paribas Fortis recently announced something notable: for the first time, fraud cases at the bank are going down. In the first half of 2025, Belgium’s largest bank recorded a 12% drop in fraud cases compared to the same period in 2024, the first such decline in its history.
At the same time, the bank is clear that fraud is not becoming simpler. Some specific techniques rose by around 6%, as criminals increasingly exploit stolen data from other platforms and remote‑access tools to win trust and push transactions. In parallel, BNP Paribas Fortis has rolled out 27 new security features in its apps this year and is calling for more cross‑sector data sharing to fight phishing and fraud.
Leadership at BNP Paribas Fortis is explicit about where the focus goes next: AI‑driven transformation, customer experience, phishing and fraud prevention, and commercial strength; all underpinned by a tighter connection between AI, data, risk, fraud and customer decisioning across the business. That direction aligns exactly with where the market is heading and with why Vyntra was built.
The Problem: Siloed Controls, Connected Criminals
For years, banks have addressed fraud, AML, sanctions and payment operations through separate tools, separate data sets and, often, separate teams. That model is increasingly difficult to sustain.
- Fraudsters operate across channels and borders, exploiting gaps between systems.
- Money laundering and sanctions evasion ride on the same payments that fraud teams monitor.
- Operational issues (delays, failed steps, SLA breaches) create both customer pain and new attack surfaces.
- Regulators no longer accept “we ran the checks”; they want transaction-level proof, on demand, years later.
When each function sees only its own slice of a transaction, the connections between signals are lost precisely when they matter most.
The Answer: Transaction Intelligence as a Unifying Layer
Transaction Intelligence is the practice of turning raw transaction data into a structured, correlated view that people can search, understand and act on from initiation to settlement, across all rails, formats and systems.
At Vyntra, this takes the form of a single intelligence layer that connects:
- Payment fraud prevention: real-time scoring of scams, APP fraud and account takeover, with behavioral profiling and automated callbacks.
- Internal fraud prevention: continuous monitoring of insider risk and collusion, not just transactions but user behavior.
- AML transaction monitoring: risk-based scoring and explainable AI to reduce false positives while meeting regulatory expectations.
- Sanctions screening: fuzzy, phonetic and token-based matching at instant-payment speed.
- Transaction observability: end-to-end payment flow monitoring, SLA tracking, search and analytics across all systems.
All of these capabilities share the same detection engine, case manager and audit trail, so coverage grows without adding new systems to manage.
From “We Think We Complied” to “Here is The Proof”
BNP Paribas Fortis’ focus on connecting AI, data, risk, fraud and customer decisions mirrors what regulators increasingly expect: not just controls, but demonstrable, transaction-level evidence of those controls.
Vyntra’s Compliance Passport addresses this directly. It:
- Determines which compliance checks apply to each transaction (based on amount, currency, country, customer segment, risk score, etc.).
- Captures and correlates execution events for fraud detection, AML monitoring, sanctions screening and other controls, even when they arrive out of order or from different systems.
- Compares expected vs. collected “stamps” and alerts in real time if a required check is missing.
- Produces structured, regulator-ready exports in minutes instead of weeks of manual reconstruction.
The result is a single “passport” per transaction, showing all expected and collected checks in one view; exactly the kind of holistic, single-window evidence that compliance and audit teams need.
Regional Realities, Global Pattern
Europe: regulation, real-time payments and customer trust
In Europe, banks face:
- Tightening AML and sanctions expectations, with regulators asking for clear, reproducible evidence of controls.
- The shift to instant and near-real-time payments, where decisions must be made in seconds, not days.
- Growing customer expectations for transparency: “Where is my payment? Why was it delayed or rejected?”
Transaction Intelligence lets European institutions:
- Run fraud, AML and sanctions checks in real time on the same engine, with explainable AI that investigators and regulators can understand.
- Monitor instant-payment SLAs at a business layer, detecting stalls before cancellations spike.
- Provide customer-facing teams with clear status, ETA, fees and block/reject reasons across every rail.
BNP Paribas’ group-wide approach underlines this direction: AI is now embedded across businesses and functions to enhance customer experience, operational efficiency and risk management, especially in fraud prevention and cybersecurity.
Africa: scale, inclusion and resilience
In Africa, Vyntra (formerly NetGuardians) has spent a decade helping banks fight financial crime, screening more than two billion transactions annually across East Africa alone. The region combines:
- Rapid growth in digital and mobile payments.
- High exposure to fraud, scams and mule networks.
- Strong regulatory focus on AML/CFT and cross-border risk.
Transaction Intelligence supports African banks by:
- Unifying fraud, AML and sanctions monitoring in one platform, reducing the cost and complexity of multiple vendors.
- Providing end-to-end visibility into payment flows across legacy and modern rails, improving operational resilience and customer trust.
- Enabling community and network risk signals (via Community Scoring & Intelligence) to catch threats that no single bank could see alone.
APAC: speed, volume and cross-border complexity
APAC institutions deal with:
- Extremely high payment volumes and a mix of real-time, batch and cross-border rails.
- Complex correspondent banking chains and multiple intermediaries.
- Diverse regulatory regimes and increasing expectations around data and auditability.
Transaction Intelligence helps APAC banks and PSPs:
- Consolidate every financial message across formats and systems into a structured, searchable view.
- Apply consistent fraud, AML and sanctions logic across rails while respecting local rules.
- Trace payments end-to-end across issuers, CSDs and financial institutions, improving investigation speed and client retention.
North America: fraud losses, regulatory scrutiny and operational efficiency
In North America, the focus is often on:
- Rising APP and authorized push payment fraud, account takeover and mule networks.
- Regulatory expectations around BSA/AML, sanctions and operational resilience.
- Pressure to reduce false positives and improve customer experience.
Transaction Intelligence enables North American institutions to:
- Score every payment against fraud, insider risk, AML and sanctions in one layer, catching connections that siloed systems miss.
- Reduce AML alert overload with risk-based scoring and explainable AI, focusing investigators on what matters.
- Provide a single version of truth for transaction data, cutting investigation time and improving reporting quality.
Where The Competitive Edge Will Come From
As BNP Paribas Fortis’ leadership has highlighted, the next edge in banking will not come from one technology or one isolated use case. It will come from how well institutions connect AI, data, risk, fraud and customer decisioning around a shared, real-time view of every transaction.
That is exactly what Vyntra’s Transaction Intelligence platform is designed to do:
- Start with the priority that matters most today: payment fraud, internal fraud, AML, sanctions or observability.
- Build on a common intelligence foundation so each new capability adds context, not complexity.
- Move from disconnected monitoring to connected, preventive intelligence — without an all-or-nothing transformation.
For senior executives, the question is no longer “Which point solution do we buy next?” but “How do we ensure every payment is scored, explained and actioned in real time, with audit-ready proof, across our entire ecosystem?”
Key Takeaways
- BNP Paribas Fortis recorded a historic 12% drop in fraud cases in H1 2025 -the first-ever decline- even as some fraud techniques rose by around 6%.
- The bank’s leadership frames the next phase as AI-driven transformation that tightly connects AI, data, risk, fraud and customer decisioning across the business.
- Transaction Intelligence unifies fraud, AML, sanctions and operational visibility in one intelligence layer, with shared engine, cases and audit trail.
- Compliance Passport turns transaction-level compliance from a weeks-long reconstruction exercise into a minutes-long, regulator-ready export.
- Region by region, from Europe through Africa and APAC to North America, the pattern is the same: connected intelligence outperforms disconnected point solutions.
FAQs about Transaction Intelligence
Do we need to replace our existing fraud, AML and sanctions systems to adopt Transaction Intelligence?
No. Vyntra is designed for modular adoption. You can start with the use case that matters most: for example, payment fraud or AML monitoring and connect additional capabilities over time. The platform integrates non-intrusively on top of existing operational systems.
How does Transaction Intelligence help with regulators and auditors?
It provides transaction-level proof of which compliance checks applied to each payment and whether they were executed, with structured, time-stamped exports that can be reproduced years later. This shifts the posture from “reconstruct when asked” to “prove as you go.”
Does this work for instant and real-time payments?
Yes. All capabilities, fraud, AML, sanctions and observability, run on the same engine in batch and real time, so instant rails get the same coverage and explainability as end-of-day processing.
How does this differ from simply adding more dashboards?
Convergence here is architectural, not cosmetic. It means one transactional reality, shared context across fraud/compliance/operations, common intelligence with domain-specific workflows, and traceability by design, not just several dashboards behind a common login.
Can this scale across multiple regions and payment rails?
Vyntra already supports over 130 financial institutions in 60+ countries, covering retail, private, institutional and digital-native banks, PSPs and infrastructure providers across diverse rails and formats.
Sibos Miami - Sept 29 - Oct 1, 2026
Learn how banks are shifting from monitoring their IT systems to Transaction Observability, and what it implies for operational risk, resilience, and customer experience.
