[Columna de opinión] Artificial Intelligence and the Future of Financial Services Trade

Unlike previous waves of automation, which largely affected manufacturing jobs, the current wave of artificial intelligence (AI) is increasingly complementing—and in some cases substituting—not only routine tasks but also more complex activities involving reasoning, communication and coordination.

In this context, AI has rapidly moved from a supporting technology to a central strategic priority within the global financial system. Banks are investing billions of dollars in applications ranging from fraud detection and credit underwriting to trading analytics and customer-service automation. For example, JPMorgan Chase reported saving approximately 360,000 hours of annual work by lawyers and loan officers through the deployment of its COIN platform.

Despite these advances, deployment has been slower and more complex than expected, reflecting the challenges of integrating AI into highly regulated and risk-sensitive environments. This tension—between large-scale investment and gradual deployment—is often framed as a firm-level issue. However, it signals a broader transformation. AI is not only reshaping how financial institutions operate; it is also beginning to alter the global organization of financial services. This raises a central question for policymakers: How will AI affect comparative advantage and, consequently, the geography of international trade in financial services?

This question is increasingly urgent. Trade in services has expanded steadily over recent decades, outpacing trade in goods in many advanced economies. In the United States, for example, between 2020 and 2025, services exports and imports grew 13 and 45 percentage points more than their goods counterparts, respectively. Financial services—including commercial banking, capital markets, insurance, asset management and related activities—have been at the forefront of this expansion, supported by digitalization, regulatory convergence and the globalization of capital markets. At the same time, investment patterns reveal a rapid technological shift: Data from the U.S. Bureau of Economic Analysis (BEA) indicate that the financial-and-insurance sector roughly tripled its software stock between 2020 and 2025.

These trends point to a structural transformation. The International Monetary Fund (IMF) has estimated that nearly 40 percent of global employment is exposed to AI, rising to around 60 percent in advanced economies, reflecting their concentration in cognitively intensive occupations.

AI is likely to accelerate these dynamics by automating cognitive tasks and reducing the need for proximity between providers and clients. The financial sector is particularly exposed: Estimates suggest that between 50 and 70 percent of tasks may be automated or augmented. This implies not only job displacement but also job transformation, with new roles emerging in areas such as AI oversight, cybersecurity and model governance. Managing this transition is, therefore, a central challenge for both financial institutions and policymakers.

At a more fundamental level, AI expands the set of tradable financial services. Activities that once required highly specialized, locally embedded human capital—such as credit evaluation, fraud detection or portfolio allocation—can increasingly be performed remotely through algorithms trained on large datasets. In principle, this lowers barriers to entry and broadens participation in global financial-services markets.

However, the implications for comparative advantage are more nuanced.

Historically, countries with deep financial markets, strong institutions and large pools of highly skilled labor have dominated exports of financial services. This advantage was rooted in the concentration of expertise and trust within specific financial centers. AI, however, allows expertise to be codified, replicated and scaled, weakening the link between local human capital and service provision.

This perspective builds on my own research on automation and international trade in goods, in which I show that the adoption of labor-saving technologies weakens traditional comparative advantage based on low wages by enabling production to relocate toward more technologically advanced economies. The key mechanism is that as routine tasks are automated, relative costs depend less on labor endowments and more on access to technology and capital.

AI is likely to generate a closely related shift in services by codifying expertise and reshaping the basis of comparative advantage in financial services. The outcome is not convergence across countries, but a reallocation of activities in which advantage increasingly reflects differences in data availability, digital infrastructure and institutional capacity.

Recent patterns are consistent with this interpretation. In the United States, for example, financial-services imports grew by approximately 38 percent in real terms between 2020 and 2025, compared with around 14.5 percent for exports. While still preliminary, this asymmetry suggests that AI may be facilitating the international sourcing of certain financial-services tasks, reinforcing the idea that technology can reshape trade flows in non-trivial ways.

For policymakers, these developments have immediate and concrete implications.

First, the sources of competitiveness are changing. Attracting highly skilled labor remains important, but it is no longer sufficient. Access to high-quality data, advanced computational infrastructure and regulatory frameworks that support innovation are becoming decisive. Digital infrastructure and data governance should therefore be viewed not only as domestic priorities but also as core instruments of international competitiveness.

Second, the fragmentation of financial services is likely to intensify. Much like global value chains in manufacturing, AI enables a finer division of labor across countries. High-value activities—such as model design, governance and strategic oversight—are likely to remain concentrated in advanced economies, while standardized and scalable tasks may be performed across a broader set of locations. This creates opportunities for emerging markets but also raises the risk of a more hierarchical global system.

Third, regulatory frameworks will be central in shaping outcomes. AI introduces a fundamental tension between efficiency and accountability. While it can improve productivity and decision-making, it also introduces opacity, model bias and new forms of risk. Requirements for transparency, auditability and explainability—particularly in credit and risk decisions—are difficult to reconcile with complex “black box” models. The challenge for policymakers, therefore, is not whether to regulate AI, but how to do so in a way that preserves both innovation and trust.

Data governance is equally critical. AI systems depend on large volumes of high-quality data, yet financial data is often fragmented and subject to strict privacy constraints. Cross-border data flows add further complexity due to differences in legal frameworks. Countries that establish clear, credible and interoperable data-governance regimes will have a significant advantage in attracting AI-driven financial activity.

At the same time, policymakers must address new forms of systemic risk. These include model bias, cybersecurity vulnerabilities and the potential for correlated failures across institutions using similar algorithms. Given the global integration of financial markets, such risks are inherently cross-border, underscoring the need for international coordination.

Trust remains a central pillar. While AI can automate many aspects of financial services, it cannot substitute for the institutional frameworks that underpin them. Legal systems, regulatory quality and the credibility of supervisory authorities will continue to play decisive roles. If anything, AI adoption increases the importance of these factors.

The experience of automation in goods trade offers a useful parallel. Technological change did not eliminate cross-country differences; it reshaped them. Some economies adapted and strengthened their positions, while others lagged. A similar process is likely to unfold in financial services.

For emerging markets, AI presents both opportunities and risks. It lowers barriers to entry in certain segments of financial services, but it also raises the bar for technological and institutional capacity. Without adequate investments, countries risk being excluded from the most dynamic market segments.

For advanced economies and established financial centers, the challenge is to maintain their positions in a changing landscape. Their advantage will increasingly depend on their ability to integrate AI while preserving high standards of regulation and trust.

International cooperation will also be essential. Divergent regulatory approaches could lead to fragmentation or regulatory arbitrage. Coordinated frameworks for data governance, model transparency and risk management will be key to ensuring that the benefits of AI are realized while limiting systemic vulnerabilities.

AI is set to redraw the global map of financial-services trade. The key issue for policymakers is not whether this transformation will occur, but how it will be shaped. Comparative advantage is not fixed; it evolves with technology, institutions and policy. Countries that anticipate these changes—and act on them—will be better positioned not only to participate in this transformation, but also to shape it.

Fuente: Internationalbanker.com