
Artificial intelligence can accelerate financial modeling, but it does not relieve the modeler of responsibility for the model or its conclusions. The central challenge is to use AI effectively without surrendering ownership of the model's purpose, structure, assumptions, logic, and evidence. This article presents the FRAME → DESIGN → BUILD → VERIFY framework as a practical approach to that challenge. FRAME defines the decision, intended user, scope, sources, and required outputs. DESIGN establishes the architecture, economic drivers, assumptions, scenarios, and controls. BUILD implements approved logic in manageable sections that can be understood and tested. VERIFY applies three distinct levels of review: technical verification, financial verification, and decision verification. The framework adds to established modeling practice by defining appropriate AI roles at each stage, identifying evidence of human ownership, and separating technical verification, financial verification, and decision verification. It applies to professional practice and finance education and is illustrated through private wealth management, capital budgeting, and financial forecasting. The modeler remains responsible for the integrity and defensibility of the analysis within the modeler's assigned role, while the decision-maker remains accountable for the decision. The framework does not certify accuracy or replace independent validation, organizational policy, or professional judgment.
Accounting firms are increasingly positioning generative AI as part of their service capabilities. Although this positioning may signal innovation, it may also affect how clients interpret later service failures. Drawing on attribution theory, this study uses an experiment to examine whether AI-focused promotional positioning influences clients' judgments after an ambiguous accounting service error. Participants read a scenario in which an accounting firm prepared a tax return that contained errors. The firm’s promotional message either emphasized investment in and use of generative AI-enabled tools or emphasized traditional professional services. The source of the error was left unspecified. Participants exposed to AI-focused promotional positioning attributed greater responsibility to the firm, perceived the error as more preventable, and judged similar errors as more likely to recur than participants exposed to traditional-services positioning. The study contributes to research on attribution in professional services by showing that pre-failure promotional positioning can shape how clients interpret accounting service failures. The findings suggest that AI-related communications may influence client expectations and accountability assessments when service-failure causes are uncertain.
This study investigates the interrelationships among generative AI risk assessment, AI policies, AI governance, and adoption practices using survey data from CPAs and corporate accountants in Japan. The analysis on a total of 372 valid responses show that AI governance, AI risk awareness, and training programs emerge as critical for mitigating risks and maximizing efficiency gains. Inadequate AI governance can exacerbate risks. Future responsible AI practices should focus on the need for structured AI policies and enough training systems. This study provides actionable insights for policymakers and practitioners seeking to integrate generative AI into accounting and auditing responsibly.
Currencies are studied using parameters usually associated with stocks by using two proxies for the market portfolio: the U.S. Dollar index (DXY) and the Global stock index (MSCI). The results show currencies relative risk to be influenced by global events, as reflected in their betas, more than by national events. Also, the relative risk of the currencies is more responsive to changes in the DXY index than to changes in the MSCI index. The relative risk of the currencies changes over time, but it is relatively stable overall. Last, stocks are better investment alternatives than currencies when compared using the Sharpe ratio.
Why do some informal firms remain persistently informal while others begin to formalize? Drawing on survey data from 120 informal SMEs in Lubumbashi, Democratic Republic of Congo, this article examines variation within the informal sector using a binary logistic regression model. Results identify perceived tax pressure, financial exclusion, and social obligations as the primary determinants of persistent informality, while administrative complexity and unemployment are not significant. These findings suggest that informality persists not because of regulatory barriers, but because formalization costs systematically outweigh its benefits under conditions of institutional uncertainty, financial constraints, and social embeddedness. Policy implications prioritize fiscal reform, financial inclusion, and differentiated intervention strategies.