Njordium Cyber Group AB has launched a new AI Fraud Detection Module integrated into its recently released Vendor Management System.
The launch comes against a backdrop of rapidly escalating invoice fraud across Europe. According to the Expert Group on Public Economics (ESO) report Svarta siffror (2026:1), Sweden’s criminal economy generates approximately SEK 352bn annually — equivalent to 5.5% of GDP — with the shadow economy alone accounting for around SEK 224bn per year, driven by undeclared labour, tax evasion, and fraudulent commercial activity.
In the European Economic Area, the EBA and ECB reported total payment fraud losses of €4.2bn in 2024, up 17% year-on-year. Credit transfers — the primary channel for fraudulent invoices — rose 24% to €2.5bn, with 85% of losses falling directly on victim organisations with little to no route to recovery.
Nearly half of all mid-market companies have encountered fake invoice scams in the past year, and nearly seven in ten fail to stop the payment in time.
Njordium Cyber Group AB provides cyber intelligence, governance, risk and compliance solutions across the Nordic region, helping enterprises manage vendor relationships, regulatory obligations and financial crime risk through an integrated technology platform.
The new module operates as a self-learning artificial intelligence engine, instantly detecting and neutralising fake invoices, phantom services or products, and inflated pricing. Fully transparent and compliant with the EU AI Act, it delivers a regulator-aligned solution to Europe’s growing invoice fraud problem.
Crucially, while fraudsters increasingly weaponise AI to craft convincing scams, Njordium’s engine counters with adaptive intelligence that improves daily through user feedback — something rule-based systems are unable to match.
Building on the Njordium VMS launched earlier this month, the module functions as a forensic auditor, extracting data from PDFs, OCR scans, Excel files, XML feeds and email attachments. It flags anomalies such as non-existent products, phantom services, or prices deviating sharply from market norms or historical baselines.
A dedicated Financial Anomaly Bridge automatically routes high-confidence detections into anti-money laundering (AML), ultimate beneficial ownership (UBO) and politically exposed persons (PEP) workflows, stopping fraudulent payments before funds leave the organisation.
The platform’s key AI-driven features include intelligent invoice extraction and validation, which processes any document format with continuous self-improvement for high-accuracy fraud screening; self-learning anomaly detection offering real-time monitoring of prices, services, products and spending behaviour with tunable sensitivity; and the Financial Anomaly Bridge, which routes invoice irregularities directly into AML, sanctions and financial crime workflows for preventive blocking.
The module also delivers transparent, explainable AI, ensuring every risk score, anomaly flag and decision is fully auditable, exportable and free of black-box models — meeting EU AI Act transparency and logging obligations. Additionally, AI-assisted multi-framework compliance allows a single vendor assessment to simultaneously satisfy NIS2, DORA, CRA, GDPR, ISO 27001 and the EU AI Act with automated, regulator-ready outputs.
Njordium Cyber Group CEO Mads Becker Jørgensen said, “With the VMS now live, we are delivering a defining moment for enterprise fraud defence. This self-learning AI module adapts faster than fraudsters evolve, explains every decision in clear language, and delivers ROI measured in millions of euros saved. One platform. Intelligent AI at enterprise scale. Zero tolerance for fake invoices. Fully transparent and compliant with the EU AI Act. Forward-thinking organisations adopting this today will not only protect their bottom line — they will lead the next wave of AI-driven risk intelligence.”
Njordium senior advisor Kim Haverblad added, “In an era where AI fuels both attacks and defences, our fully auditable, explainable engine meets the EU AI Act’s demands for transparency, human oversight and risk management — while cutting third-party fraud exposure at its source. Eighty-five percent of credit transfer fraud losses land directly on the defrauded organisation, with no recourse and no recovery. That is the number that should keep a CFO awake. We built a system that stops the payment before that number ever applies to them. This Nordic innovation is designed to attract customers seeking real protection, captivate media with its impact, and excite investors in Europe’s fastest-growing AI innovators.”
Keep up with all the latest RegTech news here
Copyright © 2026 RegTech Analyst
Copyright © 2018 RegTech Analyst
