Cutting the Gordian Knot: AI, Data, and the Future of Financial Services

The Independent was not involved in the creation of this sponsored content.

The financial services industry has spent years trying to untangle its AI and data strategy. Banks have tightened budgets, layered on technology, and optimized legacy systems—but instead of simplifying operations, these efforts have only made the knot tighter.

Greek legend tells how this ends: a knot so intricate no one could unravel it, countless failed attempts, and then—Alexander the Great. Instead of struggling with what couldn’t be undone, he pulled out his sword and cut straight through it.

Financial institutions need their own Alexander moment now.

The challenge isn’t a lack of investment; it’s that AI and data are still being treated as separate initiatives instead of a single, sovereign platform. Institutions must stop layering AI onto outdated infrastructures and instead build a foundation that upholds AI and the data function as a single, intelligent ecosystem designed for sovereignty, agility, and scale.

So, how can financial institutions break free from constraints that have held them back? The path starts with four critical questions.

1. Can AI see the whole picture?

AI is only as powerful as the data it can access. Yet many financial institutions still operate in fragmented environments, where AI models must work with partial, outdated, or siloed information. Without complete observability, even the most advanced AI models are limited.

This challenge is especially critical in financial services, where 90% of enterprise data is unstructured. Traditional AI models built for structured data fail to capture key insights in fraud detection, risk analysis, and customer intelligence.

EnterpriseDB (EDB) research shows that 68% of financial services workloads in the U.S. are hybrid, while 48% of EMEA (specifically, the U.K. and Germany) operate similarly. AI is already embedded in these environments, but visibility remains the key challenge.

“For AI to drive real intelligence, it has to connect the dots across every part of the data ecosystem,” explains Andy Slater, global accounts director at EDB, a leader in Postgres^® database solutions and AI-driven innovation. “The institutions that are seeing success are the ones designing for full observability, so AI can continuously access and interpret data in real time, no matter where it lives.”

2. Is AI built into the foundation, or is it just another layer?

AI is not an application; it is the intelligence that should power the entire financial ecosystem. Still, too many institutions treat AI as a bolt-on feature rather than the foundation of their operations.

EDB research shows that 59% of U.S. banking and finance workloads already incorporate AI, a figure that rises to 68% in EMEA. But AI’s impact is only as strong as the infrastructure it runs on.

“AI bolted onto legacy systems is like putting a jet engine on a horse-drawn carriage,” says Slater. “Without integration into a modern, real-time data platform, AI can’t see—or act—fast enough. For things like fraud prevention, that difference isn’t academic. It’s the difference between stopping an attack and cleaning up after one.”

This distinction is especially evident in fraud detection and financial crime prevention. AI-driven security tools are already helping financial institutions detect suspicious activity, but how they are deployed makes all the difference. When AI is siloed from core data operations, it reacts to threats after they happen. But when AI is embedded within a real-time data platform, it can stop fraud before it occurs—an essential capability in an industry in which milliseconds matter.

3. Are humans building AI to learn, adapt, and evolve?

AI isn’t static, and the best models today won’t be the best models tomorrow. Financial institutions that fail to embed continuous learning into their AI infrastructure risk being left behind.

McKinsey found that firms spend up to 10% of their total OpEx on IT, nearly three times the industry average, yet AI-driven productivity gains remain elusive. A model that doesn’t evolve with every transaction, risk assessment, or compliance update quickly loses its edge.

Some institutions are designing for adaptability from the start. AI-driven compliance systems, for example, need to adjust in real time to regulatory changes. Without structured feedback loops, institutions risk relying on outdated insights, leaving themselves vulnerable to inefficiencies, security gaps, and compliance failures.

“The real question isn’t whether AI can learn, but whether organizations are structured to teach it,” says Slater. “The firms seeing the greatest impact have built feedback loops directly into their architecture, ensuring that AI evolves alongside their business, regulatory landscape, and customer needs.”

4. Are individuals fully committed to sovereign, secure, open-source AI?

AI is reshaping financial services, but long-term control over data and intelligence requires a commitment to open, secure architectures. According to EDB research, a majority of financial enterprises (75%) report they are actively moving toward an open source mandate for AI and data infrastructure. The reason? Agility, security, and the ability to maintain sovereignty over their most valuable asset: data.

AI models trained on proprietary, closed-off systems lack flexibility and scalability. As regulations shift and security threats evolve, financial institutions need AI architectures that can adapt in real time.

“AI-driven financial services depend on open, adaptable platforms that are both cost-efficient and secure,” says Slater. “Institutions that recognize AI and data as sovereign, differentiated assets aren’t just optimizing operations—they’re shaping the next decade of financial services.”

Cut through complexity and build for the AI era

Financial institutions have been stuck pulling at the same tangled threads—layering AI onto legacy systems, optimizing around constraints, trying to make the old ways work. But progress doesn’t come from untangling complexity; it comes from cutting through it.

The Gordian knot wasn’t solved by patience or incremental fixes. It was solved by seeing the problem differently and taking decisive action. Financial leaders now face the same choice: continue working within outdated models, or build something designed for the AI era.

Investing involves risk and your investment may lose value. Past performance gives no indication of future results. These statements do not constitute and cannot replace investment advice.