The quick development of financial technologies and digital transactions has made fraud detection and regulatory compliance more difficult. This study introduces a Financial Digital Twin with Explainable AI over 6G (FinDT-XAI6G) to enhance real-time fraud detection and compliance monitoring in financial systems. The proposed method leverages the ultra-low latency, high bandwidth, and pervasive intelligence capabilities of 6G networks to enable seamless synchronization between digital twins and real-world financial activities. Common modeling approaches ensure interoperability across financial firms, regulatory bodies, and auditing authorities. Embedded artificial intelligence systems continuously examine vast amounts of transactional data, behavioral patterns, and contextual indications to spot anomalies that could be signs of fraud. Financial firms may now fully and transparently explain automated judgments to regulators thanks to Explainable AI (XAI) modules that enhance interpretability. Blockchain-based audit trails also guarantee data integrity, accountability, and traceability across distributed infrastructures. The integration of 6G connectivity allows for real-time monitoring, cross-border compliance validation, and instant anomaly reporting, even in scenarios with huge data volumes. Comparative studies reveal that our 6G-driven digital twin approach significantly improves detection accuracy, reduces false positives, and expedites compliance verification when compared to traditional methods. Additionally, its scenario modeling capabilities enable the proactive assessment of emerging compliance risks in dynamic regulatory and commercial contexts. Overall, this study demonstrates how financial fraud prevention and compliance assurance in next-generation digital economies can be revolutionized by 6G intelligence-powered standardized, AI-integrated digital twins.
With the increasing transaction volume, emerging fraud, complex money laundering schemes and a growing number of regulatory requirements, financial institutions have to deal with ever-increasing compliance pressures while dealing with legacy rule-based systems that create too many false positives and do not work in real-time and privacy-conscious or decentralized environments. To fill this void in integrated AI solutions, the authors introduce a novel solution a hybrid regulatory compliance monitoring platform that combines supervised ensemble models for detecting suspicious transactions, federated learning for privacy-preserving risk assessment, double-debiased machine learning for causal analysis and blockchain-based smart contracts for immutability, data integrity and auditability. The platform has 93.5% accuracy and 92.8% F1 score in transaction monitoring, with up to 4.8% systemic risk reduction with improved performance in tail risks and low latency of 3.7 seconds. These results are significantly better than the rule-based baselines and stand-alone ML methods and also are robust against heterogeneity tests and datasets. The platform’s ability to integrate centralized banking and DeFi markets through clear, scalable and secure solutions brings transformative opportunities for regulatory compliance, operational effectiveness and financial stability.
Sathish Kaniganahalli Ramareddy· International journal of com...· 0 citations
This study offers an evidence-based account of how AI in fraud detection has evolved and proposes a future research agenda emphasizing transparency, ethical assurance and global governance alignment, advancing financial risk management through conceptual clarity, methodological guidance and actionable pathways for responsible AI adoption.
Devansh Gupta, Priyanka Chugh, Poonam Mahajan· South Asian Journal of Busin...· 0 citations
Large organizations progressively adopt artificial intelligence (AI) to enhance fraud detection and ensure integrity across financial, operational, and digital systems. Despite its benefits, AI introduces challenges including data integrity risks, bias, clarity issues, and accountability gaps. Weak governance can lead to false alerts, model drift, and system vulnerabilities. This research proposes a multi-layer AI governance framework integrating technical, organizational, and controlling oversight to strengthen fraud detection and integrity assurance. The technical layer implements anomaly detection, machine learning, and explainable AI (XAI) models. The organizational layer sets up policies, accountability structures, and increase protocols, while the regulatory layer ensures compliance with GDPR, ISO standards, and AI-specific regulations. The framework adherence interoperability across layers, real-time monitoring, and adaptive responses to progressing fraud patterns. By combining ethical, procedural, and technical safeguards, the study offers a practical, scalable model that improves discovery accuracy, reduces false positives, and enhances organizational resilience and trust in AI systems. This study presents a multi-layer AI management framework that enhances fraud detection and uprightness assurance in large organizations. The framework combines technical, organizational, and regulatory layers to strengthen anomaly identification, audit completeness, and governance maturity. Adaptive thresholds and advanced AI techniques allow real-time detection with decreased alert fatigue, while blockchain and confederated learning ensure robust data integrity. Despite working challenges such as data imbalance, system latency, and explainability concerns, the framework demonstrates scalable and sustainable performance. The study offers actionable insights for policy formulation, stakeholder training, and ethical AI adoption, supporting resilient, accountable, and responsible organizational systems.
Nabeela Ehsan· Journal of Intelligent Decis...· 0 citations
This study contributes a trust-centered, infrastructure-aware AI adoption pathway specifically designed for emerging economies, offering policymakers, fintech developers, and financial institutions a pragmatic roadmap for responsible AI-enabled fraud management in Nepal.
Y. Pant, Aditya Pudasaini, R. Shrestha et al.· Islington Journal of Multidi...· 0 citations
The rapid expansion of Financial Technology (FinTech) platforms has fundamentally transformed different areas of delivery of financial services with business expansion into real-time payments, algorithmic trading, digital lending, decentralized financial products and so on. Today financial institutions face a compliance environment that has outgrown the tools built to manage it. Regulatory obligations increase and multiply across jurisdictions of operation; financial crime volumes continue to rise with advancement of technology, and data generation has grown exponentially to a scale where manual oversight is structurally impossible at the transaction level.
Legacy Governance, Risk and Compliance (GRC) frameworks were built around periodic audit cycles, static rule sets, and report heavy workflows designed for a different era. GRC frameworks are not failing because organizations run them poorly; they are failing because the environment has changed faster than the frameworks evolved. Artificial intelligence is the mechanism through which that gap is being closed in real-time, and this paper examines what that looks like in practice.
This paper covers what actually happens when financial institutions deploy AI in their GRC functions. Where it works, where it falls short, and what separates the two. The analysis draws data based on three peer-reviewed empirical studies covering over 1,155 publications and survey data from 564 financial sector professionals. Those studies are cited as evidence, not as subject matter. Some of the financial market scenarios are examined through the lens of what AI has demonstrably changed.
There are consistent findings those are worth stated across each domain from AML surveillance and KYC automation in retail banking, Credit underwriting in commercial lending, Regulatory reporting under Basel IV, and Board transparency in listed corporates.
AI returns are not determined by the sophistication of the algorithm. They are determined by the quality of the governance infrastructure surrounding it. Institutions that have learned this art early on are compounding the advantage. Those that have not are finding that AI investments disappoint not because the technology fails, but because the data and oversight conditions necessary for it to succeed were never put in place.
AI's contribution to GRC is real, statistically validated, and growing but it is not unconditional. Institutions that invest in data quality, lineage, metadata, model oversight, and ethical AI frameworks extract substantially stronger GRC returns from AI than those that deploy AI tools without the underlying governance infrastructure. This is not a theoretical nuance; it is the difference between an AI system that catches financial crime and one that floods investigators with noise.
Unknown authors· International Journal of Mod...· 0 citations
The rapid proliferation of instant settlement rails, decentralized architectures, and cross-border transaction protocols has heightened systemic vulnerabilities across modern financial infrastructures, introducing non-linear operational, liquidity, and financial crime risks. Traditional supervisory frameworks, characterized by periodic compliance reporting and static rule-based heuristics, exhibit prohibitive latency and fail to capture multi-hop illicit flows or sudden contagion across interconnected nodes. This study develops an adaptive Regulatory Technology (RegTech) framework engineered to monitor real-time risk dynamics within high-frequency digital payment ecosystems. Integrating dynamic graph topology analysis with unsupervised anomaly detection architectures, the proposed model captures latent counterparty interconnectedness, structural transaction velocity shifts, and automated money-laundering vectors. Empirical validation using high-dimensional payment settlement data demonstrates that this framework reduces systemic anomaly identification latency from hours to sub-second intervals while decreasing false-positive compliance noise by 38.6% relative to standard benchmark models. Furthermore, stress-testing under simulated liquidity shocks reveals enhanced predictive power regarding inter-institutional settlement bottlenecks. The results provide central monetary authorities and supervisory bodies with a scalable, data-driven architecture to transition from reactive ex-post audits toward continuous, proactive macro-prudential oversight, thereby preserving financial stability within digitized payment environments.
Vakhabov Bobur Alisherovich· EPRA International Journal o...· 0 citations
A new method, called CW-Net, translates the reasoning process of an autonomous vehicle’s AI system into understandable concepts that explain its behavior.