104 FEATURE These transactions can then run through generative AI reasoning models like Microsoft Azure OpenAI o3 and o4-mini to refine those that are more likely to be fraud, which can then be passed to human analysts who can focus on complex investigations.” To ensure effective collaboration between AI and human oversight, banks should establish clear rules for when and how human intervention is required. “Protocols should include setting thresholds for AI-generated alerts that necessitate human review and creating feedback loops where human analysts can provide input to improve AI models,” says Pichach. “Additionally, ongoing training and development for human analysts are essential to keep them updated on the latest AI tools and techniques, enabling them to work effectively alongside technology.” As financial criminals develop new tactics, AI is also helping organisations to adapt to emerging threats in real time. “AI-powered compliance solutions help financial institutions stay ahead of evolving financial crimes by enabling real-time threat detection, adaptive risk management and automated response mechanisms,” says Knox. “Machine learning models analyse vast amounts of transaction data to identify suspicious patterns and anomalies that may indicate fraud, money laundering or other illicit activities. Unlike rule-based systems, AI continuously learns from new threats, improving its accuracy over time. “Natural language processing (NLP) enhances screening processes by analysing unstructured data from various sources, such as news reports and regulatory updates, to identify emerging risks. Additionally, AI-driven automation reduces false positives, enabling compliance teams to focus on genuine threats more efficiently. By leveraging AI, financial institutions can strengthen their compliance frameworks, enhance resilience against financial crime and adapt swiftly to an ever-changing threat landscape.” To take advantage of AI, banks first need to ensure their data is in order. “Our customers are leveraging AI to unify data for advanced risk calculations, increasing customer acquisition, reducing false positives and ensuring regulatory control management,” says Pichach. Portuguese bank Novobanco used Microsoft Fabric and Quantexa’s entity resolution capabilities to unify its siloed data estate. “Novobanco’s obsession with being customerfirst is amazing and the idea of using Microsoft Fabric plus the entity resolution capabilities of Quantexa allows Novobanco to have a single platform to build new products and capabilities, create efficiencies and do that at once,” says Bill Borden, corporate vice president of worldwide financial services at Microsoft. The collaboration between Microsoft and Quantexa brings together powerful data integration and advanced analytics, enabling Novobanco to unlock new insights across its banking operations. “Microsoft Fabric provides this amazing gathering place for data,” says Dan Higgins, chief product officer at Quantexa. “When you then bring Quantexa’s scalability and accuracy “ The strategic use of AI can help banks differentiate themselves in a competitive market” TYLER PICHACH, MICROSOFT UBS has customised its own platform on Azure to support client advisors during customer interactions
RkJQdWJsaXNoZXIy NzQ1NTk=