Predicting systemic financial crises with AI: A macroprudential approach in the U.S. Context

2026-03-08T12:40:31-05:00

The capacity to avert systemic financial crises remains a core determinant of financial stability and the attenuation of extensive macroeconomic distress. This paper evaluates routes for embedding artificial intelligence (AI) within the macroprudential framework of the United States to enhance the pre-emptive detection of systemic risk. The study uses a secondary-data methodology to synthesize peer-reviewed empirical evidence and authoritative policy documents, organizing the corpus around four interdependent pillars: AI modeling technologies, macroprudential policy instruments, systemic-risk signal metrics, and regulatory infrastructures. The analysis confirms that predictive architectures grounded in Recurrent Neural Networks, eXtreme Gradient Boosting, and Random Forest methodologies yield optimal predictive precision once supplemented by interpretative post-hoc frameworks, with Shapley Additive Explanations (SHAP) emerging as the most potent mechanism of explanatory power. The prevailing regulatory triad countercyclical capital buffers, judiciously calibrated loan-to-value thresholds, and progressively granular probabilistic stress-testing routines is the conduit that translates AI-generated risk signals into judiciously calibrated supervisory measures. Three recurrent structural anchors, the credit-to-GDP differential, the implied volatility gauge, and the configuration of interbank liabilities persistently surface across modeling coalitions, affirming their ongoing empirical significance. The proposed embedding draws additional support from the Dodd-Frank Act and the Basel III framework, which, when considered together, confer a resilient institutional foundation for the prudent incorporation of advanced machine-learning instruments within the supervisory apparatus. The argument posits that integrating advanced artificial intelligence, meticulously validated risk indicators, and a cohesive regulatory framework significantly enhances the robustness of early-warning mechanisms and macroprudential supervision across the entire financial sector.

A conceptual framework of the road safety index for motorcyclists among school students in Malaysia based on a systematic literature review

2026-03-09T00:36:25-05:00

Motorcycle accidents contribute substantially to road fatalities in Malaysia, particularly among school-aged motorcyclists. Despite the implementation of various road safety initiatives, accident statistics continue to rise, indicating the limited effectiveness of existing interventions. Road safety indices are widely used to assess safety performance and identify areas for improvement; however, most existing indices focus on general road users and lack specificity for school students. To address this gap, this study proposes a Road Safety Index (RSI) that measures road safety among school-aged motorcyclists in Malaysia. A systematic literature review (SLR) is conducted using Scopus, Web of Science, MyCite, and Google Scholar to identify relevant studies related to road safety and motorcycle safety in the context of school students. Following the PRISMA guidelines, 22 articles published between 2021 and 2025 are identified. Based on the review, a multidimensional conceptual framework for the proposed RSI is developed, grounded in the Safe System Approach advocated by the World Health Organization (WHO). The framework comprises five key components: safe users, safe vehicles, safe environments, policy and education, and post-crash management. The proposed RSI is aligned with national and global road safety visions, including the Malaysia Road Safety Plan 2022-2030 and Sustainable Development Goal target 3.6.

International Journal of Asian Social Science

Predicting systemic financial crises with AI: A macroprudential approach in the U.S. Context

A conceptual framework of the road safety index for motorcyclists among school students in Malaysia based on a systematic literature review