An intelligent financial system starts first with trusted data

Organizations operating in financial services are being asked to do several difficult and often competing things at once: move faster while exercising greater control, all within a highly regulated system. Examples include the push toward real-time payments, the incorporation of AI-powered tools, and the shift toward digital assets. Together, these trends are increasing both the speed and volume of activity, while firms must simultaneously manage how fraud can move across multiple channels before traditional rules-based monitoring systems have a chance to detect it.

Why the pressure is increasing

Technology is changing the way the underlying market infrastructure works and will continue to do so. Capital markets are barreling toward the delivery of faster settlement and more integrated front-to-back platforms to facilitate it. Add tokenized assets, stablecoins, and autonomous financial workflows to the picture, and the challenge becomes even greater. 

To add to these pressures, finance functions are also expected to become more forward-looking rather than serving simply as record-keeping functions. Deloitte's recent Future of Finance report argues that "finance can no longer concentrate solely on transactions and retrospective reporting. AI-enabled automation and analysis could allow teams to spend more time anticipating change, supporting decisions and identifying new sources of value."[4] But none of this can be achieved simply by applying an AI layer to data that remains fragmented and sits on an aging technology infrastructure.

Better analysis of the data organizations hold can certainly help address these challenges, as can the incorporation of AI systems and the scalability and security offered by cloud-based platforms. The ideal solution though likely lies in connecting three distinct capabilities: trusted data, scalable cloud infrastructure, and responsibly governed AI.

Data: Solving the data issue must come before harnessing the AI opportunity

Most sound financial decisions are made through the application of high-quality data. It is clear that the financial services industry has no shortage of data. More often, the challenge is the quality of that data. Using data that is unreliable or unverified and connecting it directly to an AI model will not, by itself, solve the problem. An AI model may be technically sophisticated, but it cannot compensate for data that is missing, inconsistent, or poorly understood when grounding the model.

AI: From monitoring activity to anticipating risk

Once information can be combined and analyzed reliably, AI can help financial institutions move from retrospective review to earlier intervention through faster identification and tracking of anomalies. Machine learning systems can continuously analyze transactions, devices, locations, and behavioral patterns, surfacing anomalies that would be difficult to detect through static rules or manual review.
Silva says that AI-enabled anomaly detection is used by his organization to monitor more than 400 million transactions each day, identifying unusual client patterns and addressing potential issues before they escalate. The value lies not only in the automation itself or the efficiency gains it delivers, but also in the ability to identify risk earlier in the process and focus human attention where judgment is needed most.

Cloud: Building resilience through scale

The third element is the ability to better handle unpredictable demand and scale. Extreme trading days, cyber incidents, and changes in market structure can create sudden spikes in processing and analytics requirements. Fixed on-premises capacity can be expensive to maintain and support, and difficult to expand when demand increases unexpectedly. Cloud-based systems, when combined with AI capabilities and supported by high-quality data, provide a powerful foundation for addressing these challenges.

John Napoli, DTCC's Head of Digital Client Platforms and IT Business Management, describes the firm's cloud-first strategy as a response to the need for a scalable solution that can expand immediately when required. Moving workflows and data to cloud platforms allows operations to shift across geographic locations, providing the capacity to absorb surges in volume and helping services recover more quickly following a cyber incident or infrastructure outage.

For a financial market utility such as DTCC, these are not simply technology efficiencies. They help support operational resilience, continuity, and confidence across the broader financial system.

Cloud-based services can provide the flexible foundation on which data, analytics, and new digital asset services are connected. However, firms still require strong governance, human oversight, and security architectures specifically designed to address the unique risks of financial services.

Trust remains a key operating ingredient

Finally, financial institutions cannot pursue speed at the expense of accountability. Explainability, bias, privacy, cybersecurity and human oversight are all central requirements to any operating system, particularly for credit decisions, trading surveillance and compliance monitoring.In a recent article BNY adds governance, client-asset segregation, record keeping, capital and liquidity standards, anti-money-laundering protections and technology security to the foundations required for future state capital markets to operate succesfully

"Trust in data is a prerequisite for trustworthy AI, and is a core institutional obligation tied to DTCC’s role in market infrastructure."

John Napoli, Head of Digital Client Platforms and IT Business Management, DTCC

John Napoli links having trust in data to a better understanding of data lineage, governance and consistency. Those capabilities make it possible to understand where information came from, how it changed and whether it is appropriate for a particular model or decision. They also create a better basis for grounding AI models, testing outputs, assigning accountability and responding when something goes wrong.

For systemically important financial services like those provided by DTCC, resilience cannot be bolted on after deployment. It must be embedded right throughout data architecture, cloud design, cybersecurity, model governance and operational processes from the start.

The power of connecting data, cloud and AI

History is littered with examples of innovations that failed to deliver on their promise, not because they were inferior to the solutions they were designed to replace, but because they were never successfully integrated into the broader infrastructure and ecosystem around them.

The same could be true of AI, cloud, and data. Their value depends not on deploying each capability in isolation, but on connecting them as part of a broader, coherent operating model.

These technologies are often discussed through the lens of individual benefits: faster analysis, lower processing costs, greater personalization, or more automated workflows that improve efficiency, increase productivity, and reduce costs.

However, the larger opportunity is far more interconnected. Together, trusted data, AI, and cloud infrastructure can help financial institutions identify risk earlier, allocate capital with better information, support higher transaction volumes, improve liquidity, and adapt to new forms of market activity.

But the order in which these capabilities are deployed matters. Financial institutions should treat data as the foundation: a governed strategic asset that underpins every decision. From there, they can modernize the infrastructure that makes data accessible, resilient, and scalable, and then apply AI within clearly defined boundaries, with humans providing the objectives, guardrails, and oversight.

Data provides the context. Cloud delivers the scale and flexibility. AI helps turn information into action. Human governance makes the entire ecosystem trustworthy.