AI and Digital Finance Raise Financial Stability Implications, BIS Warns

Artificial intelligence (AI) and the broader digitalisation of financial markets, including innovations such as tokenisation, are driving transformational change across the global financial system — but they also pose new and intensifying financial stability risks, according to senior executives at the Bank for International Settlements (BIS).

BIS officials made these comments during a keynote speech delivered by Tao Zhang, BIS Chief Representative for Asia and the Pacific, at International Financial Week in Hong Kong on 26 January 2026. Zhang stressed that the rapid adoption of AI and digital finance technologies is reshaping market behaviour, risk transmission mechanisms, and the way central banks identify and manage systemic risks.

How AI and Digital Finance Are Reshaping Financial Markets

AI is now embedded in many facets of modern finance. Banks, asset managers and trading firms apply AI for data processing, risk modelling, fraud detection, back-office automation and customer service. Meanwhile, breakthroughs in large language models (LLMs) and machine learning have expanded AI’s reach, enabling prediction, decision-making and automation at speeds far beyond typical human capabilities.

Digital finance, in BIS terminology, refers to the digitalisation of financial assets, processes and infrastructures — including distributed ledger technologies and the tokenisation of financial claims such as securities or deposits. Tokenisation enables faster trading, settlement and collateral management, and can reduce frictions in cross-border payments.

Together, these technologies promise efficiency gains, reduced operational costs and deepened market integration. But BIS officials cautioned that the very features that make AI and digital finance appealing also change how risks arise and propagate through the financial system.

Three Channels of Financial Stability Risk

1. Market Functioning and Liquidity Risks

AI systems can accelerate trading activity and portfolio adjustment — which, in normal conditions, can boost market efficiency but under stress may amplify price volatility and liquidity strains. For example, when multiple institutions use similar AI-driven strategies, rapid, correlated trading can exacerbate short-term price swings and contribute to disorderly market conditions.

Similarly, the tokenisation of assets means that digital claims can be transferred or redeemed faster than the underlying assets can be acquired or funded. During stress events, this mismatch between the digital and underlying layers can potentially strain liquidity and settlement capacity, contributing to dislocation in markets.

2. Operational Dependencies and Resilience Issues

AI and digital finance systems tend to rely heavily on a concentrated set of infrastructure providers — such as cloud computing platforms, specialised hardware vendors and pretrained model suppliers. This concentration increases operational risk; disruptions at a critical provider — whether due to cyberattack, technical failure, or regulatory action — could ripple across wide swaths of the financial system.

Tokenisation platforms and distributed ledger networks, too, often depend on shared protocols and service providers. If these systems experience outages or security vulnerabilities, the operational fallout could be systemic rather than isolated.

3. Amplification and Propagation of Stress

Widespread use of similar AI models, data sources or decision rules can cause institutions to respond to market shocks in highly correlated ways, raising the risk of collective behaviour that amplifies stress. This “herding effect” can lead to procyclicality — meaning institutions may collectively tighten risk exposures or withdraw liquidity at the same time, worsening stress conditions.

Tokenisation also creates complex chains of interdependencies across markets and intermediaries. In stressed conditions, these links can act as modern contagion channels, swiftly propagating stress across borders and asset classes.

Why the Risks Matter for Central Banks

BIS officials stress that the significance of these developments isn’t just theoretical; they complicate how central banks identify, assess and manage financial stability risks in real-time.

1. Increased Intensity: With more interconnected and automated trading and payment mechanisms, localized disruptions are more likely to have system-wide effects.
2. Greater Speed: AI-driven automation and tokenised settlement processes compress reaction times, giving regulators and institutions less time to respond during market stress.
3. Higher Complexity: The opaqueness of AI models and the intricacies of digital finance networks make it harder to map interconnections and monitor exposures — especially when activities span geographies and regulatory jurisdictions.

Policy and Governance Imperatives

Given the cross-border nature of these technologies, BIS officials underscore the importance of international cooperation and coherent regulatory frameworks. Disparate approaches to governance and regulation can create gaps that undermine financial stability, especially when markets and technologies extend beyond national borders.

Governance frameworks must evolve to address model risk management, data governance, third-party dependencies and interoperability standards for digital finance platforms — all while maintaining incentives for innovation. Central banks and global standard-setting bodies such as the Financial Stability Board (FSB) have roles in coordinating these efforts, according to BIS remarks.

Regulatory Attention and International Context

This BIS warning comes amid broader global regulatory focus on AI and digital finance risks:

• The Financial Stability Board and BIS have both highlighted the need for closer monitoring of AI’s impact on financial markets and institutions.
• The OECD has pointed out that AI’s deep integration into finance could introduce new systemic risk channels, including interconnectedness and third-party dependencies.
• Global regulators, including the Bank of England and European authorities, are developing frameworks to govern the use of AI and digital assets in financial services.

These collective efforts underline the shared recognition that technological innovation must be balanced with robust risk management and oversight to preserve financial stability.

Balancing Innovation With Stability

Though the risks are serious, BIS officials do not argue for halting innovation; rather, they emphasise the need for informed oversight:

“For central banks, the challenge is not to resist innovation, but to understand how it changes the nature and transmission of risks, and to ensure that governance frameworks and policy approaches remain fit for purpose.”

This balanced approach seeks to harness AI and tokenisation’s benefits — such as enhanced efficiency, liquidity, and cost-reduction — while safeguarding against systemic vulnerabilities that could lead to market shocks or erosion of confidence in financial infrastructure.

Conclusion: A New Frontier for Financial Stability

The BIS warning highlights a pivotal moment in the evolution of global finance. As AI and digital finance push the boundaries of efficiency and capability, they also introduce a new class of systemic risks — rapid, complex, and sometimes opaque — that traditional regulatory tools may not fully capture.

Central banks and international authorities must therefore modernise risk frameworks, deepen cross-border cooperation, and refine governance structures to keep pace with technological change. How well regulators adapt will shape whether these innovations strengthen — or strain — the future of financial stability.

#DigitalFinance#FintechRisk#GlobalRegulationAIBISCentralBankingFinancialStabilityTokenisation