This study explores how cloud computing influences operational efficiency and cost performance in Small and Medium-sized Enterprises (SMEs) compared to traditional IT systems. Under the digital transformation, SMEs face increasing pressure to improve flexibility and efficiency, reduce costs, and respond quickly to market changes. Though revisit academic journals, industry reports, and case studies from 2013 to 2025 in a systematic way. According to the research, cloud computing can significantly lower investments during the early stage and costs of continuous maintenance while promoting scalability, operational agility, and collaboration. It also shows improved data analysis, product development, and market responsiveness. Case studies about Marais USA and SMEs in Kampala show that the benefits of cloud adoption mainly depend on technological standard, strategic countermeasure, and regional infrastructure, with noticeable differences between developed and developing areas. However, there are still a lot of risks that need to be taken seriously. Not only data security and privacy remain major points, but system reliability and cross-platform interoperability also cause technical risks and skills gaps and financial investment also limit SMEs' ability to fully utilize cloud technologies. Although emerging approaches such as edge-cloud computing paradigm, multi-cloud strategies, and green generative AI show opportunities to enhance efficiency, resilience, and sustainability, but still require careful implementation and adequate technical capacity. Ultimately, cloud computing can provide substantial benefits to SMEs in cost reduction and flexibility of operations. The proof is that cloud computing is more significant to enhance the operation of SMEs than the traditional IT system although there are differences on its impact based on enterprise size, region and digitally degree. SMEs must take a strategic and professional view of the use of cloud computing potential realization if they are to play to their full, with the support of policy and technological progress.

The frequency of mobile terminal consumer credit usage has significantly increased, but some users exhibit signs of over-indebtedness or failure to fully recognize hidden costs. According to behavioral finance, it particularly focuses on two psychological aspects: people's fear of losing money and their susceptibility to the first number they see. For instance, borrowing apps like Huabei and Weilidai examine how they exploit these psychological tendencies to encourage users to borrow more, while also evaluating whether such practices comply with regulations. Research reveals: Huabei claims interest-free borrowing, while Weilidai breaks down monthly repayments into smaller amounts without clearly stating the total interest, effectively enticing users to borrow more—an approach that lacks regulatory compliance and deviates from the well-intentioned goal of making borrowing more accessible. Finally, recommendations are provided for borrowers, app developers, and regulatory agencies to enhance the reliability of mobile lending. The recommendations emphasize enhancing users' financial risk awareness, improving platform compliance and ethical responsibility, and strengthening regulatory supervision to prevent irrational borrowing and ensure responsible consumer credit practices.

This research was completed collaboratively by the three authors, aiming to systematically assess the development status and industry differentiation characteristics of Chinese enterprises in the field of Environmental, Social and Governance (ESG), and explore the evolutionary path of ESG practices from compliance disclosure to value creation. The study adopts a systematic data collection method, based on ESG reports, sustainability reports, annual reports published on corporate official websites and public data from authoritative rating agencies, conducts a cross-year tracking of a number of Chinese Fortune Global 500 enterprises, and converts textual information into structured datasets to support multi-dimensional comparative analysis. The results show that industries present significant differentiation in financial performance, ESG rating coverage, adoption of international disclosure frameworks and achievement rate of core indicators; leading enterprises in some industries have initially institutionalized ESG governance and aligned with international standards, but the overall situation still has common shortcomings such as "high actions with low commitments" and insufficient gender diversity; the number of ESG reports continues to grow, and the disclosure paradigm is integrating towards systematic frameworks, yet there remains a gap between standardization and substantive actions. The research conclusion points out that ESG in China is undergoing a critical transformation from social responsibility response to strategic value creation. Enterprises urgently need to internalize material issue management into core competitiveness, convert compliance costs into unique customer value and sustainable business models, so as to seize the competitive advantage in the next stage.

The institutional surge into digital assets has pushed classical derivative models to their structural limits. Traditional Gaussian-based valuation frameworks, while analytically elegant, systematically ignore the extreme statistical properties inherent in Bitcoin and Ethereum returns. This research investigates this fundamental disconnect, focusing on pricing biases and tail risk management failures in non-standard cryptocurrency options. The analysis specifically examines the "kurtosis trap", where standard models fail to reflect the leptokurtic distribution and frequent price jumps observed in crypto markets. To address these deficiencies, the F-Model is introduced—a heavy-tailed alternative utilizing Student-t distributions and robust numerical risk-neutral calibration to preserve martingale properties. By applying Monte Carlo simulations optimized with Common Random Numbers (CRN) to high-frequency transaction data from the Deribit exchange, the breakdown of standard pricing metrics is quantified. The results reveal that standard frameworks exhibit a systematic underpricing bias of nearly 40% for deep Out-of-The-Money (OTM) contracts during volatile regimes. Conversely, the F-Model significantly sharpens tail risk estimation through Expected Shortfall (ES) metrics, offering a more realistic safeguard against "black swan" events. These findings underscore that incorporating non-normal parameters is no longer a theoretical luxury but a functional necessity for market stability in a 24/7 trading environment. Ultimately, ignoring these biases threatens the integrity of Decentralized Financial (DeFi) protocols, necessitating a transition toward more coherent risk measures.

As global regulatory pressure on corporate sustainability intensifies, manufacturing enterprises face growing challenges in building effective Environmental, Social, and Governance (ESG) internal control systems. Traditional approaches, reliant on manual data collection and fragmented reporting, are increasingly inadequate for the scale and complexity of modern manufacturing operations. This paper investigates how Artificial Intelligence (AI) technologies can systematically optimize ESG internal control in manufacturing enterprises. Employing a combination of longitudinal case study analysis, literature review, and panel regression, this study uses Prince Holdings (Oji Holdings Corporation) as the primary case and draws on a panel dataset of 15,623 firm-year observations from 3,358 A-share manufacturing companies over 2018–2023. The findings demonstrate that AI adoption is significantly and positively associated with ESG internal control quality (β = 1.051, p < 0.001), with data governance capability identified as a partial mediator and organizational readiness as a positive moderator. Based on these findings, a five-layer AI-ESG optimization model aligned with the Committee of Sponsoring Organizations of the Treadway Commission (COSO) framework is proposed as a replicable blueprint for the manufacturing sector.

In the context of the digital wave sweeping across all industries and the increasingly homogenized competition in the information product market, how enterprises can build differentiated competitive advantages has become a key issue. This article takes Xiaomi Group as the research object and systematically explores its differentiated competitive strategy based on an ecological perspective. The aim of the research is to reveal how technology enterprises in the digital era can create core competitiveness through ecological construction and differentiated layout in the era of digitalization, providing theoretical references and practical inspirations for information industry-related enterprises. This study adopts the methods of literature analysis and case study, and comprehensively utilizes secondary data such as public financial reports, academic papers, and industry data to systematically sort out and analyze the competitive strategy evolution and differentiated practices of Xiaomi Group. The research finds that Xiaomi's differentiated competitive strategy has undergone a dynamic evolution process from the early stage of "cost-effectiveness—online channels" to the construction of the Artificial Intelligence of Things (AIoT) ecosystem, and then to the implementation of the "human-car-home full ecosystem" strategy, and has formed a three-dimensional differentiated system covering business model, ecosystem, product technology, channel marketing, and brand users. In the digital era, enterprise competition has shifted from single-product competition to comprehensive game at the ecosystem level. Building a "product, user, data" virtuous cycle is the key path to achieving sustainable differentiated advantages.

Owing to the low occurrence frequencies of extreme financial risk events, risk prediction models are prone to being dominated by the majority of the normal samples encountered during the pretraining process when using financial risk data. This weakens the ability to effectively learn abnormal characteristics and thus reduces the sensitivity and stability of the developed predictive model. To overcome these limitations, a data class-balancing model, Financial Wasserstein Generative Adversarial Network with Gradient Penalty (FinWGAN-GP), is constructed on the basis of the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) framework. Experimental results reveal that the FinWGAN-GP model achieves high-fidelity generation and expansion for high-risk samples, and a balanced dataset is conducive to improving the training effect of the early warning model. Through the integration of multisource heterogeneous data and Artificial Intelligence (AI) methods, the proposed model can identify potential financial risks earlier and more accurately, helping enhance the forward-looking and scientific nature of financial regulation.

Based on an analysis of the research background, objectives, and significance, this paper employs a variety of research methods to examine the manifestations of credit risk prevention and control in China's banking industry. It further identifies the underlying causes of these risks from four perspectives: the external economic environment, internal banking operations, borrowers, and institutional factors. In response to the emerging trends and challenges facing bank lending, the paper proposes a series of countermeasures, including strengthening risk awareness, enhancing risk management, improving early warning systems, making effective use of credit risk mitigation instruments, implementing sound credit authorization policies, optimizing corporate governance structures, making appropriate dynamic adjustments to credit limits, and reinforcing governance functions, with the aim of improving credit risk prevention and control across China's banking sector.