Big data, blockchain and AI converge to reshape financial governance

While sustainability is frequently mentioned in policy discourse and regulatory frameworks, the study finds that it remains weakly embedded in the core of AI and machine learning research in banking. Environmental risk, climate transition risk, social impact, and long-term resilience appear as peripheral themes rather than central research drivers.

A new global study finds that while AI-driven banking research is expanding at record speed, its alignment with sustainability and environmental, social, and governance goals remains weak and fragmented.

The study, titled Mapping the Role of Artificial Intelligence and Machine Learning in Advancing Sustainable Banking, was published in the journal Sustainability. It provides a detailed analysis of how AI and machine learning are shaping banking scholarship and whether that momentum is translating into sustainable financial transformation.

AI and machine learning drive a post-2020 research surge

The analysis shows a sharp increase in academic output after 2020, coinciding with intensified digitalization across financial systems and growing demand for advanced analytical tools. AI and machine learning have become central to banking research, with publications increasingly focused on credit risk modeling, fraud detection, customer behavior analysis, algorithmic trading, and performance optimization.

These applications dominate keyword networks and thematic clusters, reflecting how AI is being positioned as a quantitative engine for improving speed, accuracy, and profitability in financial decision-making. Predictive analytics, classification models, and automated decision systems appear repeatedly as core research areas, reinforcing the idea that AI’s primary role in banking is operational enhancement.

The study finds that this focus is consistent across regions, though publication intensity varies. Europe, China, and the United States emerge as leading contributors, serving as hubs for both domestic research and international collaboration. Asia, particularly China and India, shows strong national research capacity, while European institutions demonstrate higher levels of cross-border collaboration.

Despite this global engagement, the research landscape remains uneven. Many developing and emerging economies contribute fewer publications, often reflecting limited research funding or priorities shaped by local financial challenges rather than global sustainability agendas. This uneven participation reinforces structural imbalances in how AI-driven banking knowledge is produced and disseminated.

Importantly, the study highlights that methodological sophistication has increased alongside publication volume. Advanced machine learning techniques, big data analytics, and hybrid modeling approaches are now standard across much of the literature. However, this technical maturity has not been matched by conceptual integration of sustainability objectives.

Sustainability and ESG remain peripheral in AI-driven banking research

While sustainability is frequently mentioned in policy discourse and regulatory frameworks, the study finds that it remains weakly embedded in the core of AI and machine learning research in banking. Environmental risk, climate transition risk, social impact, and long-term resilience appear as peripheral themes rather than central research drivers.

Keyword and network analyses show that ESG-related concepts are poorly connected to dominant AI research clusters. Instead of forming a coherent body of work, sustainability-related studies are scattered across the literature, often treated as secondary applications rather than foundational concerns. This fragmentation suggests that AI is still being used primarily to optimize existing banking models rather than to rethink them in response to climate and social risks.

The study also identifies a gap between technological capability and strategic intent. AI and machine learning offer powerful tools for climate risk assessment, stress testing, and sustainability reporting. Yet these capabilities are underrepresented in mainstream banking research compared to traditional risk and performance metrics.

Comparative analysis between databases reinforces this conclusion. Research indexed in Web of Science tends to emphasize methodological rigor, advanced modeling, and continuity in traditional financial risk research. Scopus-indexed literature displays greater thematic diversity, including customer-centric analytics and digital service innovation, but still lacks a consolidated sustainability-focused framework.

The absence of dominant sustainability-oriented authors or theoretical models further underscores the issue. While many researchers contribute to the field, no clear intellectual leaders or schools of thought have emerged around AI-driven sustainable banking. This suggests that sustainability remains an add-on rather than a structural priority within the research ecosystem.

Global collaboration grows as strategic integration lags

The study highlights strong internationalization in AI and machine learning research related to banking. Collaborative networks span continents, with increasing cross-border partnerships among universities and research institutions. This global exchange has accelerated knowledge diffusion and methodological convergence.

However, the authors argue that collaboration alone has not resolved deeper strategic shortcomings. Despite growing publication volume and international reach, the field lacks conceptual convergence around how AI should support long-term financial sustainability. Instead, research continues to prioritize efficiency gains, short-term profitability, and micro-level optimization.

This imbalance has practical implications. Banks increasingly face regulatory pressure to assess climate risks, disclose ESG impacts, and support sustainable development goals. Without stronger research foundations linking AI to these objectives, financial institutions risk deploying advanced technologies that improve operational metrics while leaving systemic risks unaddressed.

The study also notes that institutional research networks reflect this imbalance. While some institutions demonstrate sustained engagement in AI and machine learning research, their focus remains anchored in conventional financial analytics. Emerging sustainability topics appear episodically, often driven by specific projects rather than sustained strategic programs.