Paper I This paper provides a structured review of the growing role of artificial intelligence (AI) in financial regulation, focusing on its applications, associated risks and emerging governance frameworks. It synthesizes insights from academic literature, policy reports and institutional analyses to examine how AI is transforming compliance, supervisory oversight and risk monitoring through technologies such as RegTech and SupTech. The study highlights that while AI enhances efficiency, accuracy and scalability in regulatory processes—particularly in areas like fraud detection and anti-money laundering—it also introduces new challenges, including model opacity, algorithmic bias and systemic vulnerabilities arising from model homogeneity and third-party dependencies. By integrating technical, governance and economic perspectives, the paper identifies a critical gap in existing research regarding the broader economic and systemic implications of AI adoption. It emphasizes the tension between micro-level efficiency gains and potential macro-level risks to financial stability, and calls for more comprehensive, risk-based and adaptive governance frameworks. Overall, the paper contributes to the discourse on AI governance in financial systems by framing AI not merely as a technological tool, but as a system-level transformation with significant implications for regulation, market structure and financial stability. Paper II This paper provides a comparative review of global artificial intelligence (AI) governance frameworks, examining principles-based models, regulatory approaches, and national strategies across major jurisdictions. It analyzes key dimensions including ethical foundations, accountability structures, enforcement mechanisms, ESG integration, and institutional oversight. The study finds that while global frameworks share common normative principles such as fairness, transparency, and accountability, they diverge significantly in implementation, enforcement, and institutional design. Principles-based approaches emphasize flexibility and voluntary compliance, whereas regulatory models introduce binding obligations and risk-based classifications. National strategies further reflect differing economic priorities and political systems, contributing to fragmentation in the global governance landscape. The paper highlights critical challenges in implementation, including technical complexity, resource constraints, weak enforceability, and cross-border inconsistencies. It also identifies structural gaps in accountability, coordination, and operational clarity, limiting the effectiveness of current governance models. Overall, the study emphasizes that AI governance remains fragmented and incomplete, underscoring the need for more integrated, adaptive, and coordinated frameworks that align ethical principles with practical implementation and economic realities. Paper III This paper provides a structured review of the relationship between artificial intelligence (AI), corporate governance and firm performance, focusing on how governance mechanisms shape the outcomes of AI adoption. It examines key dimensions including productivity, profitability and risk management, alongside governance factors such as board structure, diversity and oversight mechanisms. The findings indicate that while AI adoption is generally associated with improved firm performance through efficiency gains, enhanced decision-making and better risk prediction, these benefits are highly contingent on governance quality. Strong governance frameworks enable effective implementation, strategic alignment and risk mitigation, leading to more sustainable performance outcomes. In contrast, weak governance can limit the value of AI or introduce new operational and ethical risks. The paper emphasizes the role of corporate governance as both a mediator and moderator in the relationship between AI and firm performance, highlighting the interaction between technological capabilities and institutional structures. It also identifies key limitations in existing research, including methodological constraints, narrow measurement approaches and limited consideration of long-term economic impacts. Overall, the study underscores the importance of aligning AI adoption with robust governance mechanisms to maximize firm-level benefits and ensure sustainable value creation. Paper IV This paper provides a critical review of algorithmic bias and accountability failures in artificial intelligence (AI) systems, with a focus on their economic and systemic implications. It develops an integrated framework that examines how bias emerges across three interconnected layers—data representation, model construction and institutional deployment—and how these dynamics are shaped by economic incentives and governance structures. The analysis shows that algorithmic bias and accountability failures are not isolated technical issues but systemic phenomena rooted in historical inequalities, institutional fragmentation and incentive misalignment. These dynamics lead to distorted decision-making, misallocation of resources and the reinforcement of economic inequality at the micro level, while contributing to systemic risk, market instability and concentration of power at the macro level. The paper critically evaluates existing AI governance frameworks, highlighting their reliance on technical fixes, procedural compliance and static oversight mechanisms. It identifies key structural limitations, including weak enforcement, fragmented accountability, lack of incentive alignment and insufficient attention to dynamic and system-level risks. Overall, the study reframes AI governance challenges as embedded in broader economic and institutional systems, emphasizing the need for more integrated, adaptive and incentive-aware governance approaches capable of addressing systemic bias and long-term economic consequences. Paper V This paper examines the role of artificial intelligence (AI) in public sector governance, focusing on its impact on state capacity, economic outcomes and institutional structures. It develops a conceptual framework that models the state as an algorithmic allocation system, where AI functions as a core infrastructure shaping the distribution of resources, risks and rights across populations. The analysis shows that AI-driven governance introduces structural transformations rather than merely improving efficiency. Through interconnected layers of data, model-based decision-making and institutional embedding, these systems generate feedback loops that can reinforce disparities, reduce transparency and shift decision-making authority away from traditional governance processes. The paper identifies key trade-offs inherent in AI adoption, including tensions between efficiency and due process, scale and surveillance, and objectivity and embedded bias. It further explores how these dynamics influence economic inequality and global asymmetries, particularly in contexts with uneven technological capacity and data infrastructure. A critical evaluation of existing governance frameworks reveals significant design failures, including lack of allocation visibility, fragmented accountability, misaligned incentives and absence of adaptive oversight mechanisms. Overall, the study reframes AI in governance as a systemic allocation problem embedded within economic and institutional structures, emphasizing the need for more integrated, adaptive and context-aware governance models capable of addressing the complexities of the algorithmic state. Paper VI This paper examines the economic externalities of artificial intelligence (AI) regulation, focusing on its impact on innovation, competition and financial stability. It conceptualizes regulation as a market-shaping force that influences firm behavior, market structure and systemic risk across AI-driven economic systems. The analysis shows that regulatory frameworks impose asymmetric compliance costs that disproportionately affect smaller firms, raising barriers to entry and contributing to market concentration. At the same time, regulation can enhance trust, reduce uncertainty and support the adoption of AI in high-risk domains, highlighting a non-linear relationship between regulation and innovation. The paper further explores how AI regulation interacts with financial systems, identifying risks related to algorithmic trading, model dependence and concentration in critical infrastructure. It demonstrates that while regulation can mitigate these risks, it may also introduce new vulnerabilities through standardization and regulatory-induced homogeneity. By integrating firm-level, market-level and system-level perspectives, the study highlights the complex and context-dependent nature of AI regulation. It identifies key gaps in existing research, including limited attention to distributional effects, dynamic interactions and global regulatory fragmentation. Overall, the paper emphasizes the need for adaptive, incentive-aligned and system-aware regulatory frameworks that balance innovation with market stability in increasingly complex and interconnected AI-driven economies. Paper VII This paper analyzes human–AI decision interaction as a critical yet underexplored dimension of algorithmic governance, emphasizing how cognitive behavior shapes the effectiveness, risks and legitimacy of AI-supported systems. It develops a conceptual framework integrating insights from behavioral economics, cognitive psychology, human–computer interaction and governance theory to explain patterns such as automation bias, algorithm aversion, cognitive offloading and decision delegation. The analysis shows that these interaction dynamics systematically influence how decisions are made in practice, often diverging from formal governance design. Drawing on empirical evidence from welfare administration, policing, judicial systems and healthcare, the paper demonstrates that human actors frequently rely on, selectively override, or tacitly defer to algorithmic outputs. This produces hybrid decision systems in which formal human authority coexists with de facto algorithmic influence, blurring the boundary between decision support and decision-making. The study identifies a set of interconnected cognitive risks, including error propagation, over-reliance, deskilling and loss of expertise, which together can degrade institutional capacity and weaken accountability over time. It shows that these risks emerge not from algorithms or humans alone, but from their interaction within specific institutional contexts and incentive structures. The paper critically evaluates existing governance mechanisms—such as human-in-the-loop systems, algorithmic audits and explainability—and finds that they often fail to address behavioral dynamics, resulting in symbolic rather than substantive oversight. It highlights fundamental tensions between efficiency and deliberation, accuracy and legitimacy and formal authority and functional control. By reframing human–AI interaction as a structural determinant of governance outcomes, the paper argues for a shift toward “cognitive governance,” emphasizing the need to design systems that actively shape human engagement, preserve critical judgment and manage dependency in adaptive, AI-driven decision environments.