AI and Quantum Computing in Financial Services | 2025 Guide

AI and Quantum Technology Reshaping Financial Services

The convergence of artificial intelligence and quantum computing is poised to fundamentally reshape the global financial services landscape. As captured in a landmark report by Oliver Wyman and the Global Finance & Technology Network (GFTN) following the Singapore FinTech Festival 2024 (SFF2024), these technologies represent not merely incremental improvements but catalysts for transformative growth across banking, insurance, asset management, and payments. The festival convened 65,000 participants from 134 countries, creating an unprecedented platform for examining how AI and quantum computing will redefine financial services.

The financial services industry stands at an inflection point. AI has moved beyond its experimental phase, with organizations now demanding measurable returns on their technology investments. Generative AI applications are proliferating across customer service, risk assessment, and compliance functions, while predictive AI capabilities — though technically mature — remain surprisingly underexploited in most institutions. Simultaneously, quantum computing is transitioning from theoretical curiosity to practical concern, with its implications for both computational advantage and cryptographic security demanding immediate attention from financial leaders.

This comprehensive analysis examines six critical themes from the Oliver Wyman-GFTN report: AI value capture in financial services, the essential AI operating model, lessons from enterprise deployment, the evolving regulatory landscape, quantum technology’s dual potential, and the anticipated convergence of these two revolutionary fields. Together, they paint a picture of an industry that must navigate extraordinary opportunity alongside unprecedented technical and governance challenges.

AI Use Cases Driving Value in Banking and Finance

AI’s transformative potential to increase efficiency, improve effectiveness, and elevate customer experiences across financial services is already emerging through use cases of varying scale and sophistication. The report highlights a critical observation: while generative AI (GenAI) dominates headlines, the most significant value creation often comes from traditional predictive and prescriptive AI applications that have been refined over decades.

In retail banking, AI-powered personalization engines are tailoring product recommendations, credit offers, and financial advice to individual customer profiles in real time. Insurance companies are deploying machine learning models that assess claims in seconds rather than weeks, reducing fraud losses while improving customer satisfaction. Asset managers are leveraging natural language processing to analyze earnings calls, regulatory filings, and news sentiment at scales impossible for human analysts, generating alpha through information advantages measured in milliseconds.

The report emphasizes a notable shift toward value-focused approaches that deliver measurable business outcomes rather than pursuing GenAI experimentation for its own sake. Financial institutions that have moved beyond pilots report that the most impactful applications are often mundane — automating document processing, streamlining compliance reporting, and augmenting analyst workflows — rather than the flashy chatbot demonstrations that generate media attention.

Perhaps most striking is the acknowledgment that predictive AI remains insufficiently tapped for advanced applications, even within organizations at the forefront of technology adoption. Hyper-personalized products, dynamic pricing models, and predictive customer lifetime value calculations represent enormous untapped potential that existing AI capabilities could address if properly deployed within the right organizational frameworks.

The AI Operating Model for Financial Institutions

One of the report’s most important contributions is its emphasis that merely implementing the latest technology is not enough to unlock AI’s full potential. Achieving success requires a comprehensive operating model that prioritizes solving customer problems, supported by pragmatic prioritization, governance, effective risk management, and organizational readiness.

The AI operating model framework presented by Oliver Wyman begins with “user-first” and “customer-first” principles — ensuring that every AI initiative is grounded in genuine customer need rather than technology fascination. This philosophical foundation then cascades through several operational layers: data strategy and infrastructure, model development and validation, deployment and monitoring, and continuous improvement through feedback loops.

Governance emerges as the critical differentiator between organizations that successfully scale AI and those trapped in perpetual piloting. The report highlights the importance of clearly defining roles across model governance (who validates that models perform as intended), data governance (who ensures training data quality and regulatory compliance), and AI governance teams (who manage the broader organizational and ethical implications). These overlapping domains require careful coordination to avoid both gaps and redundancies.

Establishing fundamental controls from the outset proves crucial for building customer trust — an essential ingredient for AI adoption in financial services, where consumers entrust institutions with their most sensitive personal and financial information. The report cites examples of organizations that delayed governance frameworks only to face costly remediation when regulatory scrutiny intensified, contrasting them with institutions that invested early in responsible AI frameworks and benefited from faster regulatory approval and stronger customer confidence.

Generative AI Adoption From Experimentation to Scale

The trajectory of generative AI in financial services is following a pattern familiar from previous technology waves — initial excitement, followed by disillusionment, and now an emerging maturation phase focused on sustainable value creation. SFF2024 discussions revealed that leading institutions are transitioning from broad experimentation to disciplined scaling of proven use cases.

The most successful GenAI deployments in financial services share common characteristics: they target specific, well-defined processes with clear success metrics; they integrate with existing workflows rather than requiring wholesale process redesign; and they include robust human oversight mechanisms that maintain accountability while leveraging AI’s speed and scale advantages. Customer-facing chatbots have evolved from rudimentary FAQ responders to sophisticated conversational agents capable of handling complex queries about account structures, investment products, and regulatory requirements.

Internal applications are proving equally transformative. Compliance teams are using GenAI to draft regulatory reports, synthesize lengthy policy documents, and identify potential compliance gaps across thousands of transactions. Legal departments are accelerating contract review processes by orders of magnitude. Risk teams are generating scenario analyses and stress test narratives that would previously have required weeks of manual work. The common thread is that GenAI excels at tasks involving natural language — precisely the kind of work that consumes disproportionate amounts of highly skilled human labor in financial services.

However, the report cautions against premature declaration of victory. Many organizations report significant challenges in moving from successful pilots to enterprise-scale deployment, with technical infrastructure, data quality, change management, and regulatory uncertainty cited as the primary barriers. The gap between “works in a demo” and “works reliably at scale with appropriate controls” remains substantial and expensive to bridge.

Lessons Learned From Enterprise AI Deployment

Insights from organizations at the forefront of AI adoption reveal patterns of success and failure that provide valuable guidance for institutions at earlier stages of their journey. The report distills several key lessons that transcend specific technologies or use cases.

First, data quality consistently outweighs model sophistication as the primary determinant of AI success. Organizations that invested heavily in data infrastructure — cleaning historical records, establishing data pipelines, creating golden sources of truth — report dramatically better outcomes than those that pursued cutting-edge algorithms trained on mediocre data. The adage “garbage in, garbage out” applies with particular force in financial services, where regulatory and fiduciary obligations demand high accuracy.

Second, talent strategy must evolve beyond hiring data scientists. Successful AI programs require hybrid teams combining domain expertise (understanding of financial products, markets, and regulations), technical capability (machine learning engineering and data architecture), and business acumen (ability to translate technical capabilities into commercial outcomes). The report notes that the scarcest skill set is often not technical AI expertise but rather the ability to bridge the gap between what AI can do and what the business needs it to do.

Third, change management proves as important as technology development. AI implementations that fail typically do so not because the technology doesn’t work, but because the organization cannot adapt its processes, incentives, and culture to incorporate AI-generated insights effectively. The report highlights instances where excellent AI models were rendered useless by human operators who didn’t trust their outputs, didn’t understand how to act on them, or weren’t incentivized to use them.

AI Regulation and Public-Private Partnerships

The AI regulatory landscape is evolving rapidly, with regulatory bodies worldwide recognizing the need for strategies that balance innovation and risk management. The report identifies a notable shift toward viewing regulation as a catalyst for progress rather than an inhibitor of innovation — a significant departure from the adversarial dynamic that characterized earlier technology regulatory debates.

International cooperation on AI regulations is accelerating, signaling increased global attention on AI governance. The European Union’s AI Act, Singapore’s Model AI Governance Framework, and emerging frameworks from the United States, United Kingdom, and Japan are creating a complex but increasingly coherent global regulatory landscape. Financial institutions operating across jurisdictions must navigate these overlapping requirements while maintaining competitive agility.

Public-private partnerships have emerged as vital accelerators of safe AI adoption. Government-funded innovation sandboxes allow financial institutions to test AI applications under regulatory supervision, generating insights that benefit both the industry and regulators. Collaborative research initiatives bring together academic institutions, technology companies, and financial firms to address shared challenges such as algorithmic bias detection, model explainability, and systemic risk assessment.

The report highlights successful examples from Singapore, where the Monetary Authority of Singapore (MAS) has pioneered regulatory technology initiatives that simultaneously advance financial innovation and strengthen supervisory capabilities. These models demonstrate that regulation and innovation need not be zero-sum games, particularly when both sides invest in understanding each other’s constraints and objectives.

Quantum Computing Potential for Financial Services

Although still in its early stages, quantum technology presents both transformative opportunities and significant security challenges for the financial services sector. The SFF2024 discussions, which featured quantum computing as a major theme for the first time, underscored both the excitement and the uncertainty surrounding this technology’s trajectory toward commercial viability.

On the opportunity side, quantum computing’s ability to process multiple calculations simultaneously through quantum superposition and entanglement could revolutionize several core financial functions. Portfolio optimization — currently constrained by the combinatorial explosion of possible asset allocations — could be solved orders of magnitude faster, enabling truly dynamic portfolio management that adjusts in real time to market conditions. Monte Carlo simulations used for derivatives pricing and risk assessment could be performed in minutes rather than hours, fundamentally changing how institutions manage market and credit risk.

Fraud detection represents another promising application area. Quantum machine learning algorithms could identify patterns in transaction data that classical computers cannot detect, potentially reducing financial crime losses while improving the customer experience by reducing false positive rates. Credit risk modeling, regulatory stress testing, and high-frequency trading strategy optimization are additional domains where quantum advantage could prove decisive.

However, the report candidly acknowledges that most leading institutions are currently in the exploration phase, with limited production applications realized. Significant technological breakthroughs — particularly in quantum error correction and qubit stability — remain necessary to unlock the full potential of quantum computing for financial applications. Current quantum computers, while impressive as research tools, lack the reliability and scale required for production financial workloads.

Post-Quantum Cryptography and Security Preparedness

While quantum computing’s opportunities remain largely theoretical, its threat to current cryptographic infrastructure is immediate and concrete. The “harvest now, decrypt later” attack vector — where adversaries collect encrypted data today with the intention of decrypting it once quantum computers mature — means that sensitive financial data transmitted today could be compromised in the future. This makes post-quantum security preparation an urgent priority rather than a distant planning exercise.

The core vulnerability lies in public key cryptography, which underpins virtually all secure digital communications and transactions in financial services. RSA and elliptic curve cryptography — the mathematical foundations of current encryption — can be broken by sufficiently powerful quantum computers running Shor’s algorithm. While the timeline for such quantum computers remains debated, the consensus among experts at SFF2024 was that financial institutions cannot afford to wait for certainty before beginning their migration.

The post-quantum cryptography migration presents enormous logistical challenges. Financial institutions must inventory their cryptographic dependencies — a surprisingly difficult task given decades of system accumulation — evaluate and test quantum-resistant algorithms standardized by NIST, develop migration plans that maintain service continuity, and coordinate with counterparties and infrastructure providers who must undergo their own migrations.

The report recommends a phased approach beginning with cryptographic inventory and risk assessment, followed by pilot implementations of quantum-resistant algorithms in low-risk systems, and gradually expanding to production critical infrastructure. Financial institutions that begin this journey now will be better positioned when quantum computing threats materialize, while those that delay risk finding themselves in a scramble that could compromise both security and operational stability.

Convergence of AI and Quantum Technology

The intersection of AI and quantum computing — quantum-enhanced machine learning — represents perhaps the most tantalizing long-term opportunity identified in the report. Quantum computers could accelerate machine learning training by exploring solution spaces more efficiently than classical computers, potentially enabling AI models of unprecedented sophistication and accuracy.

Specific applications being explored include quantum kernel methods for classification tasks, variational quantum eigensolvers for optimization problems embedded in AI workflows, and quantum random number generators for improved model initialization and sampling. In financial services, these advances could translate to more accurate credit scoring models, better fraud detection, superior market prediction capabilities, and more robust stress testing frameworks.

However, the report maintains a measured tone, acknowledging that substantial challenges remain before this convergence can be realized at commercial scale. Quantum hardware limitations, the difficulty of designing quantum algorithms that outperform classical alternatives for practically relevant problem sizes, and the shortage of talent with expertise in both quantum computing and financial AI all constrain near-term progress. The so-called “quantum advantage” for real-world financial applications has not yet been definitively demonstrated.

Despite these challenges, the report encourages financial institutions to maintain awareness and selective engagement with quantum AI research. Early investment in quantum literacy — ensuring that technology leadership understands both the potential and the limitations — positions institutions to move quickly when breakthroughs occur. Partnerships with quantum computing companies and academic research groups provide low-cost options for remaining at the frontier of this potentially transformative convergence.

Future Outlook for AI and Quantum in Finance

Looking forward, the report identifies several critical trajectories that will shape the intersection of AI, quantum computing, and financial services through 2025 and beyond. The overarching theme is one of accelerating maturation across both technologies, accompanied by increasing urgency in governance and security preparation.

For AI, the next phase will be characterized by deeper integration into core business processes — moving from augmenting human workers to fundamentally redesigning workflows around AI capabilities. Financial institutions that have built strong operating models and governance frameworks will pull ahead of competitors still struggling with basic deployment challenges. The gap between AI leaders and laggards in financial services is expected to widen significantly.

For quantum computing, the focus shifts from “if” to “when” — with the primary strategic question being how quickly financial institutions can prepare their infrastructure for both the opportunities and threats that quantum technology presents. The report predicts that quantum-resistant cryptography migration will become a regulatory requirement in major jurisdictions within the next few years, making early preparation a competitive advantage.

The broader ecosystem implications are equally significant. Financial technology vendors, consulting firms, and regulatory bodies are all investing heavily in AI and quantum capabilities, creating a rich environment of tools, frameworks, and guidance for financial institutions to leverage. The SFF2024 discussions made clear that no institution can or should attempt to develop these capabilities entirely in isolation — partnerships, industry consortia, and regulatory collaboration will be essential for navigating the complexity ahead.

As the 10th anniversary of the Singapore FinTech Festival approaches, the financial services industry stands better equipped than ever to harness technology for customer value creation. The critical success factor will not be which institutions possess the most advanced technology, but which ones have built the organizational capabilities, governance frameworks, and strategic partnerships necessary to deploy that technology responsibly and effectively at scale.