Innovation plays a central role in driving China's high-quality development. However, China remains heavily reliant on international sources for key core technologies and components. In recent years, global economic fluctuations and geopolitical tensions have constrained technological innovation activities, with the risk of innovation disruption persisting and potentially increasing. It is urgent to optimize innovation practices and activate innovation resilience. Technological innovation often has high risks due to its large cost scale and high output uncertainty, and is easily constrained by corporate resources. Patient capital can provide long-term and stable financial support for corporate technological innovation. Furthermore, through its risk diversification function, patient capital can effectively mitigate the high-risk nature of technological innovation. Therefore, patient capital has the typical characteristics of highly matching with the intrinsic needs of corporate technological innovation, and is gradually becoming a new quality driving force to enhance corporate innovation resilience.
However, due to the inherent profit-seeking and risk-taking nature of capital, the excessive penetration of patient capital may intensify the “goal conflict” caused by investors chasing short-term self-interest. This could lead to an imbalance of interests between investors and innovative subjects and ultimately weaken the sustainability of the innovation ecosystem. Additionally, patient capital may also lead to the development of a comfortable mentality and innovation inertia among managers. These issues raise a series of key questions that deserve deeper inquiry. Can patient capital become the core element to stimulate corporate innovation resilience And how to effectively ensure that patient capital can continue to promote corporate innovation resilience and help enterprises overcome technological blockades and continuously realize technological breakthroughs.
Using the data of A-share listed companies in Shanghai and Shenzhen from 2009 to 2023, this study empirically examines the influence of patient capital on innovation resilience. Meanwhile, from the horizontal strategy dimension and vertical time dimension, this study constructs a framework for the transmission path of patient capital's influence on innovation resilience. Additionally, by analyzing the characteristics of internal innovation decision-makers within enterprises and the characteristics of the external innovation environment, this study explores how patient capital influences innovation resilience under different circumstances.
The results show that the influence of patient capital on corporate innovation resilience presents an inverted U-shaped relationship and the optimal allocation for patient capital is 34.36%. Mechanism tests reveal that patient capital affects corporate innovation resilience through three dimensions: innovation cooperativeness, innovation ambidexterity, and innovation sustainability. Further research based on innovation decision-makers and environmental characteristics shows that the inverted U-shaped impact of patient capital on corporate innovation resilience is further strengthened when managers exhibit higher levels of patience, innovation decision-making power is more concentrated, and intellectual property protection is higher.
Compared to existing research, the possible marginal contribution of this study is three-fold. First, this study reveals the inverted U-shaped relationship between patient capital and corporate innovation resilience from a nonlinear dual perspective, which not only deepens the theoretical understanding of the duality of patient capital, but also provides a scientific basis for corporate capital allocation by identifying the “optimal” threshold. Second,this study explores the role of patient capital in enterprise innovation resilience from three dimensions: innovation cooperativeness, innovation ambidexterity, and innovation continuity. It not only expands the connotation of innovation resilience research at the theoretical level, but also provides new ideas for enterprises to optimize capital allocation and enhance innovation effectiveness in practical application. Third, the study characterizes the innovation decision subjects within the firm in terms of the degree of managerial patience and the allocation of innovation decision-making power, and applies the degree of intellectual property protection to characterize the external innovation environment. From the perspectives of internal decision-making subjects and external environment, it can provide new perspectives and insights for academics to understand the role of the boundary of patient capital, and help promote the in-depth integration of relevant theoretical research and corporate practice.

In recent years, the impact of "black swan" events such as the decoupling of Sino-US technology, the conflict between Russia and Ukraine, and the new round of Israeli-Palestinian conflict has intensified the trend of "anti-globalization", accelerated the reconstruction of the global industrial chain, and increased the risk of chain breaking and blocking in China's industrial chain and supply chain. Consequently, these events pose substantial challenges to the innovation-driven growth of businesses, the establishment of a modern industrial system, the cultivation of new quality productive forces, and the pursuit of high-quality development. In particular, the sudden global public health crisis in early 2020 has exacerbated the turbulence of the world's economic and political landscape. The uncertainty of the market environment and the severity of the crisis have caused a huge impact on enterprises. An important emerging issue in management has emerged: In the face of escalating volatility, uncertainty, complexity, and ambiguity (VUCA, "VUCA") in organizational management situations, as well as the risks of "chain breaking and blocking" in the industrial chain and supply chain, and frequent extreme events, how can manufacturing enterprises, as the core entities of the industrial chain and supply chain, quickly adjust to overcome difficulties and achieve resilient growth amid crises and adversities?
This study, grounded in the theory of dynamic capabilities, delves into the effects and underlying mechanisms of digital transformation on the organizational resilience of Chinese manufacturing firms. Drawing on a dataset encompassing 3 446 manufacturing enterprises listed on the Shanghai and Shenzhen A-share markets from 2009 to 2023, the study employs empirical analysis leveraging big data text mining and a bidirectional fixed effects regression model. The findings indicate that digital transformation substantially bolsters organizational resilience. Regarding the mechanism, digital transformation promotes open innovation, thereby improving organizational resilience. Furthermore, an economic model based on multi-period difference-in-differences is employed to test the impact effect, revealing that major unexpected global public health crises have significantly impacted the organizational resilience of Chinese manufacturing enterprises. However, companies with a higher degree of digital transformation demonstrate stronger crisis response capabilities. Various robustness tests support these conclusions. This study provides theoretical and practical insights for enterprises to accelerate digital and intelligent transformation, cultivate dynamic capabilities, and enhance organizational resilience to uncertainty, thereby fostering new productive forces in the context of accelerated technological-economic paradigm shifts.
The research contributions of this paper are mainly reflected in three aspects. While the majority of existing literature on organizational resilience centers on its conceptualization and precursors, there is a dearth of studies examining the mechanisms underlying the development of organizational resilience.This study explores the mechanism of digital transformation on organizational resilience, expands the theoretical mechanism research on organizational resilience, and provides empirical evidence for the promotion effect of digital transformation on corporate organizational resilience. Second, the existing research rarely pays attention to the impact mechanism of digital transformation on corporate organizational resilience in crisis situations. This study addresses this gap by examining, through the lenses of dynamic capability theory and open innovation, how digital transformation fosters open innovation and subsequently enhances organizational resilience; from the perspective of digital innovation, it opens the "black box" of the process of digital transformation affecting organizational resilience. Third, the existing research on major global public health emergencies mainly focuses on the impact of such major public crisis events on the macro-economy and industrial chain security, but few studies focus on how micro-entities of enterprises can effectively respond to the impact of major crises. This study uses the multi-period double difference method to empirically examine the impact of major global public health crises on micro-enterprise entities, and concludes that digital transformation can significantly enhance the ability of enterprises to cope with crises, and provides important theoretical and practical references for accelerating the digital transformation of manufacturing enterprises in the new journey of China's modernization, gaining sustained competitive advantages in the VUCA era, accelerating the promotion of new industrialization, building a modern industrial system, and accelerating the cultivation of new quality productive forces.

Taking the awarding of the 2025 Nobel Prize in economics by Joel Mokyr, Philippe Aghion, and Peter Howitt as an important starting point, this study systematically explores the historical evolution, theoretical core, and contemporary relevance of the "creative destruction" mechanism in Schumpeterian growth theory. Against the backdrop of a new global context characterized by rapid scientific and technological transformation, industrial upgrading, and great-power competition, this study aims to address the central question:how can China achieve high-quality and sustainable economic growth through "destructive innovation" in the new development stage?
At the theoretical level, this study traces the paradigm evolution of growth economics: from Joseph Schumpeter′s concepts of "Innovation-driven Growth" and "Creative Destruction", to Robert Solow′s exogenous growth theory, Paul Romer and Robert Lucas′s endogenous growth theory, and further to the modern Schumpeterian growth model constructed by Aghion and Howitt which incorporates "Creative Destruction" into a general equilibrium framework. Mokyr′s "Institutionalization of Knowledge" supplements the institutional prerequisites for its operation. The core contribution of this study lies in clarifying the hierarchical relationship between "Creative Destruction" at the macro-economic level and "Disruptive Innovation" at the micro market level. As the concretization of the former, the latter explains how latecomer enterprises break the competitive advantages of incumbents through low-end market penetration, new market development, or technological subversion, thereby promoting industrial and economic upgrading.
At the practical level of China′s economy, the evolution of "Creative Destruction" is divided into three stages:(1)The embryonic stage (1978—2001):It provided space for disrupting the old system through institutional deregulation, and relied on the "Disruptive Innovation" of microeconomic agents such as farmers and township and village enterprises to facilitate the materialization of a new equilibrium.(2)The acceleration stage (2002—2012): Driven by the dual engines of "Factors and Innovation", China integrated into the global value chain to absorb and re-innovate technologies, breaking the monopoly of foreign capital in the manufacturing sector. (3) The deepening stage (2013—present):A framework of "proactive government, efficient market, and innovation ecosystem" has been formed. The "Destruction" involves the elimination of backward production capacity and the digital transformation of traditional industries, while the "Creation" focuses on disruptive technological innovation, the cultivation of emerging industries, and the construction of an innovation ecosystem.Empirical cases illustrate the micro-level mechanisms: BYD achieved a high-end breakthrough in manufacturing through "technological iteration", catalyzing the automotive industry′s shift from foreign dominance to indigenous control; Pinduoduo (PDD) unlocked demand in underserved markets via "low-end market penetration",contributing to the balanced development of a unified national market; and TikTok exported digital ecosystems through "overseas market creation," reshaping the global digital services landscape.
This study makes the following contributions: it constructs a three-dimensional collaborative framework of "institutional reconstruction, market evolution, and innovation-driven growth", clarifies the theoretical linkage between micro- and macro-level dynamics, and validates its applicability in the Chinese context through cross-sectoral case studies. Looking ahead, technology-driven development, upgraded globalization, green transformation, and the integration of the digital and real economies will shape the future trajectory of creative destruction. Policy efforts should prioritize core technology R&D, social cost mitigation, talent incentives, intellectual property protection, and international cooperation. China′s economic miracle is fundamentally the outcome of evolving "creative destruction" and invigorated "disruptive innovation", offering practical insights and "Chinese wisdom" to other latecomer nations.

Key industrial technologies are the core areas for cultivating technological competitive advantages and seizing opportunities for industrial development, playing a pivotal role in driving breakthroughs in industrial technology. Strategic emerging industries are knowledge- and technology-intensive sectors rooted in major technological breakthroughs and evolving development needs. Their key technologies are the essential and irreplaceable technologies (or links) that play an important role in the industry, reflecting current technological hotspots, difficulties, or future technological breakthroughs. As China’s technological and industrial competitiveness advances rapidly, Western countries have intensified their technological blockades against China. Strategic emerging industries now face the "small courtyard, high fence" technological dilemma, resulting in unbalanced and inadequate development of key technologies. Against this backdrop, it is essential for China's strategic emerging industries to focus on enhancing their independent innovation capabilities and breaking technological barriers. Therefore, in the new round of technological revolution and industrial transformation, identifying and clarifying the development direction of key technologies, and making forward-looking layouts are of great significance for enhancing independent innovation capabilities, achieving high-level technological self-reliance and self-improvement, and accelerating the development of new quality productive forces.
Traditional technology identification methods mostly rely on single-source data or static analysis, making it difficult to reveal the cross-disciplinary relevance and dynamic evolution process of key technologies. Furthermore, they lack systematic consideration in data source construction and dynamic analysis, resulting in insufficient accuracy and agility in technology identification. The methods of multi-source data fusion and knowledge association aggregation provide a new direction for solving the above problems. The former can integrate multi-source data and statistical methods into a unified framework, adapting to the intelligence analysis needs for addressing uncertainty and complexity. The latter functions through the reorganization, association, aggregation and presentation of multi-dimensional and multi-granularity information objects (including knowledge association, knowledge aggregation, etc.).
Drawing on a three-stage intelligence analysis model, this study integrates multi-source data fusion, knowledge association and aggregation methods to construct a key technology identification model for strategic emerging industries.Specifically, first, multi-source data fusion provides a comprehensive data foundation for key technology identification; second, knowledge association is applied to reveal potential correlations between technical topics in multi-source data, which in turn helps identify key technological hotspots; and finally, guided by the theme of knowledge aggregation and integration technology, core themes in the development of generic technologies and industrial-specific technologies are extracted, and on this basis, the development direction of key technologies in strategic emerging industries is further explored.
To verify the scientific validity of the key technology identification method, this study conducts a horizontal analysis by selecting the BERT terminology model and the Gompertz patent model. Three quantitative indicators for knowledge network topology are adopted: Technology Potential Index (TPI), Technology Influence Degree (TID), and Cross-domain Relevance (CDR). The study finds that incorporating multi-source data fusion and knowledge association aggregation into a unified framework to construct an industry key technology identification model can help identify strategic emerging industry key technologies with uncertainty and complexity. In addition, the effectiveness of the proposed method is verified by comparing it with relevant authoritative documents.
The contribution of this paper includes two aspects: Firstly, it enriches the identification methods of emerging technologies. In response to the uncertainty and complexity of key technologies in strategic emerging industries, it integrates multi-source data fusion and knowledge association aggregation into the intelligence analysis model, reduces the difficulty of identifying key technologies in strategic emerging industries, expands the ideas of emerging technology identification, and deepens the identification methods of emerging technologies. Secondly, it reveals key industrial technologies with higher granularity. The study conducts a deep analysis of common and hot technologies in the industry, and compares them with authoritative and public literature to obtain finer grained key technologies. The insights offer theoretical support for government authorities in formulating targeted industrial policies and provide valuable guidance for enterprises to advance their technological innovation strategies.

In recent years, the VUCA characteristics including volatility, uncertainty, complexity, and ambiguity have become increasingly prominent, posing severe survival challenges for enterprises. However, some firms have managed to survive, recover, and even thrive through adverse shocks, largely due to their organizational resilience—a critical capability that enables firms to maintain stability and rebound swiftly from unexpected disruptions. This resilience has become indispensable for corporate survival and sustainable development in today's volatile environment. Meanwhile, the digital wave has swept through, with emerging technologies like artificial intelligence establishing key competitive advantages for enterprises. These technologies are being integrated into daily operations and management models, significantly enhancing stability and recovery capabilities during crises, thereby forming a close connection with organizational resilience.
While existing research has examined the determinants of resilience in manufacturing enterprises from various perspectives, it has yet to fully account for the complex interplay among digital technologies, strategic positioning within industrial chains, and internal resource allocation decisions, all of which are crucial in dynamic environments. It is worth noting that the enabling effect of AI technology on enterprise resilience and its realization mechanisms have not been systematically clarified and empirically tested in the existing literature, which will become a key bottleneck to deepen the understanding of the formation of enterprise resilience in the digital era.
Therefore, to address the gap, this study draws on resource-based theory and utilizes panel data from Chinese listed manufacturing firms (2011-2023). By applying text analysis to measure AI application intensity based on the frequency of AI-related keywords in corporate disclosures, the study quantifies the development level of AI technology adoption and examines its impact on enterprise resilience. Meanwhile, this study explores the specific impact channels of AI technology application on corporate resilience from the perspective of multi-dimensional effects across the upstream, midstream, and downstream of the industrial chain, and analyzes the differentiated impacts of AI technology application under different scenarios.
The research findings indicate that, first, AI technology application significantly enhances enterprise resilience. AI technology positively strengthens corporate resilience through three key mechanisms: empowering digital innovation in the upstream sector, optimizing digital operations management in the midstream sector, and providing digital marketing services in the downstream sector. Heterogeneity analysis reveals that both external factors (industry competition intensity) and internal resource foundations (digital transformation speed, factor intensity, and fixed asset investment levels) significantly amplify AI's positive impact on corporate resilience. However, the analysis of strategic resource allocation patterns shows that improved ESG performance creates "resource crowding-out" effects, while excessive resource reserves lead to "resource surplus," thereby diminishing AI's catalytic role. In summary, the study innovatively constructs an industrial chain collaborative analysis framework, revealing the dynamic law and influence boundary of artificial intelligence technology penetration enhancing enterprise resilience.
Thus, this study proposes the following practical management recommendations for both the Chinese government and enterprises. The government should focus on building an AI technology application support system and strengthening industrial chain coordination policies, with a focus promoting the integrated application of digital technologies represented by AI across the upstream, midstream, and downstream segments of the industrial chain. Enterprises should actively adopt digital technologies represented by AI, optimize industrial chain structures, enhance digital intelligence capabilities, and proactively leverage external environmental pressures to forge their own resilience.
This study contributes to the literature in three innovative ways. First, it introduces an industry-ecosystem perspective to systematically examine how AI application enhances resilience in China's manufacturing sector, bridging a theoretical gap between AI and organizational resilience. Second, it unpacks the mechanisms through which AI affects resilience via digital innovation, operational optimization, and marketing enhancement, while also highlighting how external environments, internal resources, and strategic allocation patterns shape these effects. Third, it offers a comprehensive theoretical lens for understanding the evolutionary path of AI-enabled resilience, providing actionable insights for both scholars and practitioners in the digital transformation era.

Artificial intelligence (AI) is rapidly reshaping national governance, social operations, and economic structures, while simultaneously generating multifaceted risks relating to safety, privacy, discrimination, and systemic uncertainty. In an era characterized by rapid technological iteration and the wide proliferation of AI application scenarios, constructing a scientific, systematic, and forward-looking risk-based regulatory framework has become an essential legislative task for China. This article conducts a comparative examination of AI regulatory regimes in the European Union, South Korea, Canada, and the United States, analyzing their respective approaches to risk identification, classification models, and the allocation of regulatory obligations, with the aim of informing China′s future AI legislation.
At the comparative level, the European Union adopts a four-tier framework—unacceptable, high, limited, and minimal risk—and imposes stringent, lifecycle-wide compliance obligations on high-risk systems. Although comprehensive, this framework produces excessively heavy compliance burdens and lacks flexibility in responding to technological dynamics. South Korea employs a horizontal regulatory model centered on high-impact AI, characterized by concise provisions and a streamlined structure, yet its taxonomy does not adequately differentiate among distinct categories of risks. Canada distinguishes between biased-output systems and harm-based systems, but ultimately applies uniform obligations to both categories, resulting in a disconnect between risk classification and regulatory practice. The United States adopts the most flexible structure: a dual-track model distinguishing safety-impacting AI from rights-impacting AI. On the basis of a unified minimum compliance baseline, the U.S. model adds differentiated requirements for rights-impacting systems and incorporates deferral, exemption, and dynamic adjustment mechanisms that enhance regulatory adaptability.
Despite differences in legislative traditions and policy objectives, the four jurisdictions share a common governance logic: risk identification as the regulatory starting point, risk classification as the core organizing principle, and differentiated obligations as the primary regulatory tool. The proportionality principle underlies these systems. The U.S. dual-track model distinguishing safety and rights impacts offers notable advantages in proportionality, precision, and institutional flexibility, thereby providing a valuable template for China in building a multi-level governance structure.
In light of these comparative insights, this article proposes that China optimize its AI risk-governance framework along three dimensions. First, China should establish a layered legislative structure combining common rules + sector-specific rules. A national-level Artificial Intelligence Basic Law should articulate overarching governance principles, risk-classification methods, and baseline regulatory obligations. Sectoral regulatory authorities should then develop technical standards and regulatory rules tailored to specific application scenarios, thereby balancing systemic coherence with operational flexibility and avoiding fragmented governance.Second, China should adopt a dual-track classification framework that distinguishes between safety-impacting AI and rights-impacting AI. Systems involving life safety, critical infrastructure, or public security should be categorized as safety-impacting AI, whereas systems that affect fairness, fundamental rights, or vulnerable groups should be regulated as rights-impacting AI. For systems that present hybrid or overlapping risks, a primary-risk identification mechanism should be introduced to classify them according to their dominant risk attributes. In addition, a combined-obligations mechanism should be implemented to allow both sets of obligations to apply where necessary, thereby enhancing the precision of risk identification and strengthening the applicability of regulatory tools.Third, China should develop a comprehensive system of dynamic adjustment and flexible exemptions. Through presumed-strict classification, application-based exemptions, periodic review, and cross-departmental feedback mechanisms, regulatory measures can be dynamically aligned with technological evolution and sector-specific characteristics. Such mechanisms help prevent regulatory rigidity and excessive compliance burdens, ensuring that governance tools remain adaptive to emerging risks and evolving industrial practices.
In sum, through comparative analysis of foreign regulatory models and the construction of a localized governance pathway, this article argues that the core of China′s AI risk-governance framework lies in risk-based classification as its organizing principle, a layered legislative structure as its institutional foundation, a dual-path classification model as its methodological approach, and dynamic adjustment mechanisms as its regulatory toolset. A governance system guided by unified national principles, supported by differentiated regulatory rules, and coordinated between central and sectoral authorities can achieve a dynamic balance between safeguarding safety and fostering innovation, thereby forming a Chinese model of AI risk governance capable of addressing the complexities of the digital era.

Amid the global digital wave, artificial intelligence (AI) technology, with its powerful data processing and intelligent decision-making capabilities, is increasingly becoming the core driving force for innovation and transformation across various industries. However, the introduction of AI transcends mere tool replacement; it reshapes the working modes of organizations and employees, giving rise to a series of management challenges. A prominent challenge is how to effectively leverage AI to empower employees and stimulate their innovative behavior. A review of the literature reveals that scholars' research on stimulating AI-driven employee innovation behavior mainly focuses on two perspectives: exploration of the double-edged sword effect on AI-driven employee innovation behavior from individuals' perception of AI, and examination of how AI collaborates with employees to promote individual innovation. Nevertheless, these studies overlook the significant role of leadership in stimulating AI-driven employee innovation behavior. Leaders' AI symbolization refers to leaders' explicit expression of support, acceptance, and promotion of AI by taking actions closely related to AI and displaying items that reflect their preference for AI. In light of this, this study investigates the mechanism and applicable boundaries of leaders' AI symbolization on AI-driven employee innovation behavior, aiming to provide guidance for both theory and practice.
Drawing on social cognitive theory, this study reveals the mediating mechanisms of leaders' AI symbolization on AI-driven employee innovation behavior through two aspects: technical cognitive trust (AI trust) and self-cognitive efficacy (AI innovation self-efficacy). Leaders' AI symbolization reflects their recognition, support, and trust in AI. This influences employees to trust AI more, increasing their willingness to accept AI and boosting AI-driven innovation. Additionally, employees' experience with leaders' AI symbolization helps them recognize their own innovative abilities. They gain confidence in solving complex problems and completing tasks innovatively with AI assistance. This enhances their AI innovation self-efficacy and provides psychological support for AI-driven innovation. Furthermore, individuals' cognitive processes towards leadership behavior are not only influenced by leaders' traits and behaviors themselves but also depend on how individuals understand and interpret these traits and behaviors. The inherent complexity of AI has left many leaders with insufficient expertise to fully grasp its implications, often resulting in a tendency to offer only surface-level support without the critical resources, training programs, or strategic direction needed to effectively implement AI solutions and address genuine organizational requirements. This can easily trigger employees' attribution analysis of leaders' AI symbolization motives. Therefore, this study explores the boundary conditions of leaders' AI symbolization influencing AI-driven employee innovation behavior through the dual-mediating cognitive mechanism from the perspective of employees' attribution of leaders' AI symbolization motives.
The analysis of matched data from 488 employees in two stages indicates that leaders' AI symbolization positively affects AI-driven employee innovation behavior through AI trust and AI innovation self-efficacy. Moreover, when employees attribute leaders' AI symbolization motives to performance improvement, the positive impact of leaders' AI symbolization on employees' AI trust is enhanced, thereby boosting AI-driven employee innovation behavior. Conversely, if attributed to impression management, the positive effects of leaders' AI symbolization on employees' AI trust and AI innovation self-efficacy are weakened, thereby reducing AI-driven employee innovation behavior. However, when employees attribute leaders' AI symbolization motives to performance improvement, the moderating effect of leaders' AI symbolization on AI innovation self-efficacy is not significant, and the moderated mediation hypothesis is also not significant.
The theoretical contributions of this study are as follows: First, it enriches the research on leadership factors in the antecedent mechanism of AI-driven employee innovation behavior and expands the influence of leaders' AI symbolization, providing a new perspective for understanding how leaders can effectively stimulate employees' innovative potential through AI symbolization. Second, this study innovatively analyzes how leaders' AI symbolization influences AI-driven employee innovation behavior and its effects through the dual-mediating paths of technical cognitive trust (i.e., AI trust) and self-cognitive efficacy (i.e., AI innovation self-efficacy), offering insights from a social cognitive perspective. Third, drawing on attribution theory, this study explores the boundary conditions of leaders' AI symbolization influencing AI-driven employee innovation behavior from the perspective of employees' attribution of leaders' motives, making an important supplement to the research on leaders' AI symbolization.

In recent years, a growing number of organizations have rapidly adopted artificial intelligence (AI) technology, aiming to build a sustainable competitive advantage through automation and intelligence. Despite substantial investments in AI, many organizations have yet to realize satisfactory returns. A key reason for this shortfall lies in the disproportionate emphasis on the technical implementation of AI, while neglecting the management of AI use at the employee level. Consequently, AI often fails to be fully integrated into business processes or to realize its full potential. Although AI can effectively reduce costs and improve organizational efficiency, its growing role in tasks previously performed by humans has led to a squeeze on employees' professional value and livelihood. This has given rise to a phenomenon known as “AI disruption awareness”which refers to employees' perception of the threats posed by AI applications. Such awareness may trigger resistance to AI and become a significant barrier to its adoption and effective use. Therefore, as organizations undergo AI-driven transformation, it is essential to understand and address the impact of AI disruption awareness on employees' use of AI.
While academic discussions around AI disruption awareness have grown in recent years, relatively little attention has been paid to its influence on employees' use of AI. Existing research has primarily focused on employees' willingness to use AI, with little attention paid to how AI disruption awareness affects employees' actual AI usage behavior. Moreover, the majority of prior studies have emphasized the negative effects of AI disruption awareness on AI usage, neglecting its potential positive effects. Against this backdrop, this study focuses on employees' AI usage behavior in AI application scenarios, exploring how they adjust their use of AI in response to AI disruption awareness, in order to expand the study of the impact of AI disruption awareness on AI usage. Drawing on the cognitive appraisal theory of stress, the study constructs a model to explore how AI disruption awareness differentially impacts employees' innovative and avoidant use of AI, and the moderating role of strengths-based leadership in this process.
By analyzing two-wave survey data collected from 317 employees, the study yields the following conclusions: AI disruption awareness triggers two distinct strategies of innovative use and avoidant use of AI, and the choice of strategy is influenced by strengths-based leadership. Under the influence of strengths-based leadership, employees tend to make a challenge appraisal of AI disruption awareness, which drives them to adopt innovative usage strategies toward AI. Conversely, in the absence of strengths-based leadership, employees will make a threat appraisal of AI disruption awareness, leading them to adopt avoidant usage strategies toward AI.
The theoretical contributions of this study are as follows: First,it shifts the analytical lens from intention to actual behavior, foregrounding employee agency in AI application. By addressing the key question of “how AI disruption awareness influences employees' AI usage behavior”, this study offers new insights into how employees use AI under the influence of AI disruption awareness. Additionally, by revealing the impact of AI disruption awareness on creative use—a positive AI usage behavior—this study addresses the limitations of previous research, which often adopted a singularly negative perspective. Second, unlike previous studies that mostly explored employees' behavioral performance under AI disruption awareness from a single positive or negative perspective, this study integrates previous research perspectives based on a dialectical perspective, incorporates employees' positive and negative responses into the same framework, and proposes a dual behavioral mechanism of employees' AI disruption awareness, which provides a more comprehensive theoretical explanation for understanding the effects of AI disruption awareness. Third, this study proposes that strengths-based leadership is an important conditioning factor in determining the effect of AI disruption awareness. This not only bridges the gap of past studies' understanding of the boundaries of the differential impact effects of AI impact awareness from a leader's perspective, but also provides effective clues to reconcile the controversy of existing studies on the differential impact effects of AI disruption awareness.

Green innovation(GI) has become crucial for the high-quality development of enterprises. With the impact of digital technology on the real economy, digital transformation(DT) has risen to a significant national development strategy. The process of DT implies introducing advanced productivity, which can empower enterprises to transform the development mode to realize GI. However, this process depends to a certain extent on the impact of enterprise agility and ESG information disclosure. Based on the theory of organizational information processing, DT can help enterprises respond to changes in the external environment promptly to promote GI by improving the information processing capability and effectively realizing the agility response of enterprises at the industrial, production, and sales ends. In addition, DT also exists in the interactive communication between enterprises and society. Accroding to the signaling theory, high-quality ESG information disclosure can play a "signaling effect" in conveying a positive enterprise image to society, and it is easy for enterprises to obtain external support and create conditions for GI. Green innovation (GI) is crucial for high-quality enterprise development, while digital transformation (DT) has become a key national strategy, driving advanced productivity. DT enhances enterprise agility and information processing capabilities, facilitating GI through efficient responses across industry, production, and sales. Additionally, DT fosters interactive communication between enterprises and society. High-quality ESG (Environmental, Social, Governance) disclosure acts as a positive signal, attracting external support for GI. Meanwhile, there are multiple interactions between the GI process of enterprises and the government. In recent years, local governments have been paying more and more attention to new quality productive forces. Local governments' focus on new quality productive forces supports industrial upgrading and provides a favorable environment for GI. However, further research is needed to explore the mechanisms of enterprise agility, ESG disclosure, and government attention within the DT-GI framework.
Using the data of Chinese listed enterprises from 2015 to 2023, this paper introduces the mediating variables of agility responsiveness and ESG disclosure and the moderating variable of local government's attention to new quality productive forces and empirically explores how
DT empowers GI of enterprises from the perspective of new quality productive forces. The results show that DT has a positive and significant impact on GI; agility responsiveness and ESG disclosure play a mediating role between DT and GI; local government's attention to new qualitative productivity positively moderates the relationship among DT, ESG disclosure and GI, and at the same time strengthens the mediating role of ESG disclosure between DT and GI. Further analysis reveals that the local government's concern on new quality productive forces can positively moderate the relationship between agility responsiveness and GI, and strengthen the mediating role of agility responsiveness in the relationship between DT and GI only when the degree of enterprise capital redundancy is high.
This paper makes several significant contributions to the literature on digital transformation (DT) and green innovation (GI). Firstly, it shifts the focus from the economic impacts of DT to its role in driving green development and green innovation dividends under the framework of new quality productive forces. This approach bridges gaps in current research and offers new empirical evidence on value creation through green-oriented digital transformation. Secondly, the study reveals the dual mediating roles of agility responsiveness and ESG disclosure quality between DT and enterprise GI. Unlike previous research that examined resource input, information sharing, and governance, this paper explores the theoretical mechanisms underlying these relationships, expanding the understanding of how DT empowers GI. Thirdly, it examines the moderating role of local governments' attention to new quality productive forces in influencing the GI process through DT. This research clarifies the contextual boundaries of DT's impact on GI and aligns with China's emphasis on developing new quality productive forces. It provides new empirical insights into how local government focus can shape enterprises' DT and GI, contributing to both theoretical and practical advancements in the field.

In the context of global economic restructuring and rapid technological advancement, innovation has become the core driver of high-quality development. However, China's innovation system exhibits a structural imbalance in ambidextrous innovation: exploitative innovation, which focuses on incremental improvements, dominates, while exploratory innovation, aimed at breakthrough technologies, lags behind. This imbalance stems from micro-level corporate dilemmas, including short-termism in resource allocation and path dependency, which are exacerbated by traditional financial systems favoring short-term returns. Against this backdrop, the concept of "patient capital" has emerged as a strategic solution to foster long-term innovation. Although prior studies confirm that patient capital fosters corporate innovation, they leave two critical gaps: the differential impact on exploratory versus exploitative innovation has not been disentangled, and the cognitive pathway through which patient capital shapes managers' resource-allocation decisions—turning long-term funds into ambidextrous R&D choices—remains theoretically underdeveloped.
This study examines how patient capital influences corporate ambidextrous innovation, with a focus on the mediating role of managerial myopia. Using the data of A-share companies listed in China's Shanghai and Shenzhen Stock Exchanges from 2008 to 2023, the study employs fixed-effects models and instrumental variable methods to address potential endogeneity issues. In this study, patient capital is set as the independent variable, which is jointly measured by the proportion of relational debt and the proportion of stable equity. The dependent variable is corporate ambidextrous innovation, where exploitative patents refer to those with the first four digits of IPC (International Patent Classification) that have appeared in the past five years, and exploratory patents refer to those with the first four digits of IPC that have never appeared before. Managerial myopia is measured by the proportion of short-term perspective words in the MD&A (Management's Discussion and Analysis) section of annual reports, with a linear transformation of multiplying by 100 twice, and is used to test the mediating effect.
The findings reveal three key insights: First, patient capital has an asymmetric effect on ambidextrous innovation—it linearly promotes exploratory innovation but exhibits a U-shaped relationship with exploitative innovation. Initially, patient capital suppresses exploitative innovation due to resource reallocation toward exploratory projects, but beyond a critical threshold of 167.77, it synergistically enhances both types of innovation. Second, managerial myopia acts as a mediator: patient capital mitigates short-termism, which in turn positively affects exploratory innovation and exhibits an inverted U-shaped relationship with exploitative innovation. Moderate myopia may temporarily boost exploitative innovation, but excessive myopia harms both innovation types. Thirdly, the mediating path of patient capital's impact on ambidextrous innovation is heterogeneous. For exploratory innovation, patient capital significantly enhances it by alleviating managerial myopia. For exploitative innovation, however, there exists a dual mechanism: when the scale of patient capital is below the critical threshold, alleviating myopia weakens exploitative innovation; once it exceeds the threshold, alleviating myopia simultaneously drives the synergistic development of both types of innovation.
This study contributes to the literature in three ways: (1) It uncovers the non-linear and asymmetric effects of patient capital on ambidextrous innovation, enriching the understanding of how capital heterogeneity influences innovation strategies. (2) It introduces managerial myopia as a cognitive mediator, bridging the gap between financial resources and innovation outcomes. (3) It provides empirical evidence for policymakers and firms to optimize capital allocation and governance mechanisms, emphasizing the need to cultivate long-term-oriented investment ecosystems.
The implications are twofold. For policymakers, expanding patient capital,particularly in strategic industries,is crucial to cross the tipping point for synergistic innovation. For firms, aligning managerial incentives with long-term innovation metrics and dynamically adjusting resource allocation can mitigate short-termism and unlock dual innovation potential. Limitations include the exclusion of non-listed firms and reliance on textual measures for managerial myopia. Future research could explore cross-industry heterogeneity and alternative measures of capital patience.

With the widespread adoption of AI in critical fields such as credit and justice, algorithmic fairness has become a key issue in digital governance. While developers often try to quantify fairness using specific metrics, the selection of these metrics is not merely a neutral technical setting. Rather, it reflects a strategic trade-off where companies should balance compliance costs, efficiency, and social responsibility. This paper analyzes the Impossibility Theorem of Fairness, which demonstrates mathematically that different fairness metrics are often incompatible. This incompatibility forces developers to prioritize certain ethical values over others. Consequently, the selection of a metric becomes the actual source of algorithmic bias or fairness, making legal scrutiny essential.
This paper identifies significant gaps in the current legal framework when regulating this choice behavior. Through a critical analysis of existing anti-discrimination laws and algorithmic governance regulations, the paper reveals two primary structural dislocations. The first is the dislocation of regulatory focus: current laws predominantly focus on observable discriminatory results while neglecting the upstream process of metric selection. This retrospective focus allows enterprises to engage in regulatory arbitrage and fairness washing, where developers select technically easier or commercially favorable metrics to create a veneer of compliance, thereby evading substantive scrutiny. The second is the dislocation of regulatory standards. There is an inherent conflict between the legal paradigm of individual causal accountability, which seeks specific causation for individual harm, and the algorithmic logic of group statistical correlation, which operates on probability distributions across populations. This mismatch renders traditional legal remedies ineffective, as individual plaintiffs struggle to prove causation against black-box statistical decisions, leading to formalism in algorithmic auditing.
In light of these challenges, this paper argues that a mere optimization of result-based metrics is insufficient. Instead, legal regulation must undergo a paradigm shift from Result-Based Regulation to Process Accountability. The law should not attempt to prescribe a single “correct” metric, which is theoretically impossible, but should instead regulate the decision-making process itself. The core thesis is that the act of choosing fairness metrics must be transformed from an opaque technical decision into a legally reviewable, transparent, and accountable procedure.
To implement this shift, the paper proposes a comprehensive regulatory framework centered on Procedural Due Diligence and a full-lifecycle Algorithmic Impact Assessment (AIA) system. Specifically, the framework advocates for the establishment of Procedural Due Diligence, where algorithm developers, similar to fiduciaries in corporate law, bear a duty of care. They should document the rationale behind their metric selection, demonstrating that they have weighed potential conflicts, assessed societal impacts, and considered the rights of vulnerable groups. Furthermore, the paper details the design of a robust AIA regime that mandates pre-deployment evaluation to identify data biases and justify the final choice of metrics. To ensure checks and balances, the framework suggests a multi-layered evaluation structure involving internal ethical committees, external independent audits for high-risk algorithms, and regulatory filing. Finally, recognizing the varying risks of different algorithms, the paper advocates for a dynamic grading mechanism to balance regulatory effectiveness with the burden on innovation.
By anchoring regulation in the procedural propriety of metric selection, this framework bridges the gap between abstract legal principles and concrete technical implementations. It effectively addresses the issue of fairness washing by making the value trade-offs visible and traceable. Ultimately, this paper contributes to the discourse on algorithmic governance by providing a concrete pathway to embed Process Accountability into the lifecycle of algorithmic development, ensuring that the pursuit of digital efficiency does not come at the expense of substantive social justice.Meanwhile,a process-based accountability framework, while theoretically compelling, is not self-executing. Its operationalization hinges on a constellation of institutional enablers and practical safeguards: regulatory agencies must cultivate the technical capacity to substantively review algorithmic impact assessments; firms must navigate the delicate balance between procedural compliance and innovation incentives; and courts must progressively refine doctrinal standards to treat procedural due diligence as a legally cognizable benchmark for assessing algorithmic legitimacy. These interlocking challenges represent critical frontiers for future scholarship as a roadmap for embedding meaningful accountability into the governance of algorithmic systems.

In the rapidly evolving landscape of the digital economy and under the imperatives of innovation-driven development strategies, breakthrough innovation has emerged as a critical lever for enterprises to transcend technological trajectory lock-in and achieve discontinuous advancements that disrupt existing market paradigms. This study investigates how digital leadership—a paradigm characterized by technological acumen, transformative vision, and data-driven strategic decision-making—catalyzes breakthrough innovation within organizations. Rooted in dynamic capability theory, which emphasizes the continuous cycle of sensing, seizing, and reconfiguring resources to sustain competitive advantage, this study explores the mechanisms through which digital leadership fosters breakthrough innovation, with resource orchestration as a mediator and strategic flexibility as a moderator. The significance of this study lies in its response to a pressing gap in the literature: while prior studies have emphasized resource-based views and absorptive capacity, they have largely overlooked the pivotal role of leadership in navigating digital disruptions and reconfiguring organizational capabilities for non-incremental innovation.
The study aims to explore how digital leadership drives breakthrough innovation which is defined as radical technological leaps or business model transformations that break from established trajectories. It proposes that digital leadership, as a higher-order dynamic capability, enhances firms′ ability to overcome path dependencies by reshaping strategic cognition and resource allocation. To this end, the study constructs a theoretical framework proposing that digital leadership exerts a positive impact on breakthrough innovation, with resource orchestration defined as the systematic process of identifying, integrating, and activating organizational resources to function as a pivotal mediating mechanism. Additionally, strategic flexibility, the capacity to swiftly adapt strategic directions and resource configurations, is hypothesized to amplify the effect of digital leadership on resource orchestration, thereby enhancing innovation outcomes in dynamic environments.
This study empirically employs a robust two-wave longitudinal survey design , collecting data from 341 enterprises in Eastern China—a region globally recognized for its advanced digital economy and dynamic innovation ecosystems. The sample predominantly comprises manufacturing firms, reflecting the industrial composition of the area, and data were gathered through validated scales adapted from prior literature, ensuring psychometric rigor. Analytical methods, including correlation analysis and hierarchical regression, are employed to test the proposed relationships, with additional robustness checks conducted across firm subgroups to validate the findings′ consistency.
The results reveal several key insights. First, digital leadership significantly and positively impacts breakthrough innovation, underscoring its role in broadening innovation horizons through technological foresight and strategic agility. This finding highlights how digital leadership enables firms to leapfrog traditional innovation constraints. Second, resource orchestration partially mediates the relationship between digital leadership and breakthrough innovation. This mediation elucidates how digital leadership translates into tangible innovation outcomes by dynamically reconfiguring organizational resources, bridging the gap between strategic intent and execution. Third, strategic flexibility positively moderates the linkage between digital leadership and resource orchestration, amplifying the effect in firms with high adaptability. However, robustness tests indicate that this moderating effect weakens in older firms and larger enterprises, where organizational inertia and structural complexity may attenuate the impact of flexibility.
In practice, the findings provide actionable guidance for for leaders and policymakers during digital transformation. Firms should develop digital leadership skills in senior management, integrating technology foresight into decision-making. They should also emphasize resource orchestration with modular designs and data-driven valuation to boost resource deployment efficiency. Policymakers can support these efforts by developing digital leadership programs, creating data-sharing ecosystems, and establishing adaptable regulations. Enhancing strategic flexibility through agile governance structures further enables firms to capitalize on digital leadership′s transformative potential.

The current view that enterprises apply data elements to gain potential benefits has become a consensus in the academic community; however, in practice, many enterprises face the dilemma of continuously increasing investments in data elements without achieving significant innovation outcomes. This necessitates considering the issue of the deep integration between data elements and enterprise innovation. Unlike traditional innovation resources with relatively static and proprietary attributes, data elements have stronger mobility and non-proprietary attributes such as self-growth and non-competitiveness, and the change of such resource attributes leads to a predicament of insufficient theoretical explanatory power of traditional innovation theories to explain the underlying logic of data elements empowering enterprise innovation capability. How to effectively integrate data elements with innovation processes has emerged as a critical challenge for enterprises seeking to survive and thrive in the digital-intelligence era.
Drawing on the resource-based view and dynamic capabilities theory, this study explores the core mechanisms through which data elements empower enterprise innovation capability from the perspectives of knowledge accumulation and dynamic capabilities. Following the logical pathway of “data-information-knowledge-innovation”, it first investigates whether data elements promote enterprise knowledge accumulation. Simultaneously, grounded in dynamic capabilities theory, it examines whether data elements enhance enterprise dynamic capabilities. Furthermore, the study seeks to unveil the dual-path mediating effects of knowledge accumulation and dynamic capabilities in the relationship between data elements and enterprise innovation capability. Using listed enterprises as the research subject, the study analyzes data from a balanced panel data-set comprising 1239 A-share listed enterprises spanning 2014 to 2023, which was ultimately obtained by matching corporate annual reports with patent data. Employing machine learning and text analysis word frequency methods, it constructs enterprise data element indicators, and measures enterprises' knowledge breadth and depth based on patent data. Through constructing a dual-path chain mediation model, this study applies two-way fixed-effects panel regression analysis to empirically investigate the relationships among enterprise data elements, knowledge accumulation, dynamic capabilities, and innovation capability. In addition, this paper investigates the variations in different regions and types of enterprises utilize data elements to enhance their innovation capability.
The results show that (1) the application of data elements has led to significant changes in corporate knowledge accumulation and dynamic capabilities, but not all of these changes are positive; (2)the changes in corporate knowledge accumulation and dynamic capabilities brought about by data elements have a significant impact on corporate innovation capability; (3) data elements enhance corporate innovation capability by improving knowledge depth; however, the increase in knowledge breadth brought about by data has a negative impact on corporate innovation; (4) data elements enhance organizational agility and flexibility, which are beneficial for corporate innovation;(5) dynamic capability and knowledge accumulation play a chain mediating role in the process of data elements empowering enterprise innovation.
This paper theoretically enriches the understanding of the enterprise knowledge effects generated by data elements, and provides theoretical support for knowledge management in digital environments. The dual-path chain mediation model constructed in this study not only supplements the mechanism of data element innovation effects from the perspectives of knowledge accumulation and dynamic capabilities, but also provides an explanatory basis for the causal relationships of knowledge effects generated by data elements through the lens of dynamic capabilities. By analyzing the heterogeneous impacts of different knowledge dimensions and changes in dynamic capabilities on corporate innovation capabilities, the study explores the micro-level implementation mechanisms that effectively connect data elements with corporate innovation, offering decision-making references for enterprises to formulate efficient digital innovation strategies in the digital-intelligent environment. The managerial implications of this study are as follows: In the context of digital-intelligent transformation, enterprises should not only incorporate data elements into the core of their innovation strategies and build data-driven innovation systems, but also objectively examine the new knowledge management challenges triggered by data elements, prioritize the efficient alignment between knowledge resources and innovation transformation, actively establish agile digital organizational structures, and explore dynamic adaptation mechanisms and efficient conversion mechanisms through which data elements empower enterprise innovation.

Entity List sanctions serve as a pivotal tool for the United States in its crackdown on China. A notable instance was in April 2018 when the U.S. government imposed a technology embargo on ZTE, marking the beginning of a broader technology blockade against China. Under these sanctions,targeted enterprises face restrictions on purchasing U.S.-controlled technologies, products, and services, and even with license applications, they often face strict scrutiny and rejections, leading to operational instability and higher uncertainties. As economic globalization and specialization deepen, enterprises form close supply chain networks. However, the sanctions on midstream enterprises can trigger a chain reaction, transmitting risks to upstream and downstream enterprises, increasing supply chain risks, and weakening overall competitiveness. This situation raises the question of what strategies upstream and downstream enterprises should adopt to deal with the supply chain risks caused by the technology blockade.
From the perspective of supply chain, this paper examines the two-way spillover effect of midstream enterprises on the collaborative innovation of upstream and downstream enterprises after the U.S. technology blockade. It draws on the CNRDS database to compute patent and supply chain matching data to gauge enterprise collaborative innovation. After rigorous sample processing, 2 679 valid samples emerge, comprising 1 466 "year-midstream-upstream enterprise" matches and 1213 "year- midstream-downstream enterprise" matches. Subsequently, a benchmark model is constructed, followed by in-depth empirical analysis. In the mechanism analysis, the study respectively measures the operational risk of upstream and downstream enterprises based on the annual stock daily return volatility of these enterprises, and the higher the value, the greater the operational risk. It then adopts the proportion of the sum of the top 3 executive compensations to all executive compensations to measure managerial overconfidence, and the higher this value is, the higher the degree of managerial overconfidence.
The study finds that after the midstream enterprises are subject to the U.S. technology blockade, the supply chain spillover effect is generated, which not only promotes the collaborative innovation of upstream enterprises, but also promotes the collaborative innovation of downstream enterprises, thus highlighting the symmetry of the U.S. technology blockade supply chain spillover. Operational risk and managerial overconfidence are the internal impact mechanism of U.S. technology blockade on the collaborative innovation of upstream and downstream enterprises. The heterogeneity analysis shows that the spillover effect of the U.S. technology blockade on the supply chain is more obvious in the samples with closer supply chain distance, higher supply chain stability, lower supply chain concentration and higher market status.
According to the empirical results, the following countermeasures and suggestions are proposed respectively. For enterprises, on the one hand, when making collaborative innovation decisions, they not only need to assess the risks they face directly, but also need to comprehensively assess the potential threats in the entire supply chain network, especially the potential risks brought by other enterprises in the supply chain under the sanctions of the U.S. Entity List, so as to adjust the collaborative innovation strategy in a timely manner. On the other hand, enterprises should set up a full-time risk management department, strengthen the voice of independent directors and set up a special audit committee to conduct a third-party evaluation of managers ' major decisions, so as to curb managers ' overconfidence tendency and unblock the channels through which U.S. technology blockade affect the collaborative innovation of upstream and downstream enterprises. For the government, on the one hand, it should take into account the spillover effects on the supply chain caused by the U.S. technology blockade,and support not only the midstream enterprises that have been directly sanctioned but also provide appropriate resource allocation to the upstream and downstream enterprises within the supply chain, so as to maximize the effect of policy support by improving the overall competitiveness of the supply chain ; on the other hand, the government should tailor innovation support policies to different enterprises based on their unique circumstances, so as to enhance the level of enterprise collaborative innovation.

In the rapidly evolving digital economy, data assets have emerged as a pivotal strategic resource, driving corporate transformation and playing an increasingly vital role in creating enterprise value. As businesses grow more reliant on data-driven decision-making, the disclosure of data assets has become a key factor influencing corporate performance and market competitiveness. Existing literature has explored either the direct impact of data asset disclosure on enterprise value or its indirect role through the lens of cost of equity capital in enhancing capital market efficiency in resource allocation. However, there remains a lack of systematic analysis regarding the underlying mechanisms through which data asset disclosure drives enterprise value creation. Most studies have yet to unpack the "black box" linking disclosure to enterprise value, failing to identify the specific pathways through which such disclosures exert their influence or the contextual conditions under which these effects are more pronounced.
To address this gap, this study conducts a comprehensive empirical analysis to investigate the impact of data asset disclosure on corporate value creation. The sample consists of 27 141 firm-year observations of Chinese A-share listed companies from 2008 to 2023, primarily drawn from the CSMAR database and firms′ annual financial reports. The dependent variable is Tobin′s Q, used to measure enterprise value creation, as it captures both current market valuation and expectations of future growth, and offers superior data availability and comparability in the context of China′s capital markets. Data asset disclosure constructed from firms′ annual reports is selected as the key explanatory variable.
The empirical findings yield several key conclusions. First, data asset disclosure exerts a statistically significant and economically meaningful positive effect on corporate value creation. This suggests that firms proactively disclosing their data assets are better positioned to unlock value, as such disclosures enhance transparency and signal the firm′s capacity to leverage data for competitive advantage. Second, a detailed mechanism analysis identifies two primary pathways: (1) improving surplus levels by optimizing operational efficiency and profitability via the strategic use of data assets; and (2) reducing the cost of equity capital, as increased transparency mitigates information asymmetry and lowers investor risk perceptions. These findings underscore the dual financial benefits of data asset disclosure, reinforcing its importance in corporate strategy. Furthermore, heterogeneity analysis reveals that the positive impact of data asset disclosure is not uniform across firms but varies with contextual factors. Specifically, the value-enhancing effect is more pronounced in firms operating in highly competitive industries, where data-driven insights provide a critical edge in sustaining market positions. Additionally, the benefits are stronger for firms located in regions with higher data marketization, where institutional support and infrastructure facilitate the effective utilization of data assets. Finally, firms with robust internal governance structures exhibit a more significant valuation premium from data asset disclosure, suggesting that strong governance amplifies the strategic benefits of data transparency.
This study makes several important contributions to theory and practice. Theoretically, it advances understanding of data asset disclosure as a strategic tool in the digital economy, highlighting its role in shaping enterprise value through financial and market-based channels. Practically, the findings offer actionable insights for corporate managers seeking to optimize disclosure strategies, emphasizing the importance of data asset transparency in enhancing competitiveness. Moreover, the study provides policymakers with empirical evidence supporting the promotion of data element marketization, suggesting that regulatory frameworks encouraging data disclosure can foster more efficient capital allocation and economic growth.
In summary, this study underscores the critical role of data asset disclosure in the digital era and provides a foundation for future studies exploring the intersection of data strategy, corporate finance, and market dynamics. Its implications extend beyond the Chinese context, offering valuable lessons for firms and regulators worldwide navigating the challenges and opportunities of the data-driven economy.

Strengthening urban innovation systems and optimizing the efficiency of innovation factor allocation are critical for overcoming the "middle-technology trap" and realizing a Chinese-style modernization. Against this backdrop, the digital revolution has positioned data as a crucial production factor. In China, public data accounts for over 80% of the national total data resources, with an estimated value of 10-15 trillion yuan in 2020. The nationwide open datasets have exceeded 2 million items, covering livelihood areas such as transportation, healthcare, education, and environmental protection. Open government data policies can break down information silos, enhance inter-departmental coordinated development, and expand socio-economic benefits. Both domestic and international research consensus holds that enhancing innovation capacity is a key impact of public data openness, though systematic empirical verification remains insufficient. Domestic research mostly focuses on the enterprise perspective, with less research on urban innovation.Additionally, existing literature varies significantly in its research angles, resulting in no unified conclusions regarding how public data openness specifically influences urban innovation. Therefore, it is essential to explore their mechanisms in promoting urban innovation.
This study utilizes annual panel data from 271 cities in China from 2009 to 2023 to construct a multi-period difference-in-differences model to explore the impact of open public data on urban innovation. The establishment of open government data platforms in cities is treated as a quasi-natural experiment, with the number of patent grants per city as the dependent variable. Mechanism tests are conducted via effective market and proactive government paths. In addition, the paper studies the moderating role of digital inclusive finance and human capital stock, and finally uses a spatial Durbin model to investigate whether the positive effect of open government data on urban innovation has spatial spillovers. The study tests the spatial autocorrelation of the dependent variable—urban innovation capacity. Following the existing literature, it computes the global Moran’s I for each year under three weighting schemes: the contiguity matrix (W1), the economic-distance matrix (W2), and the nested economic-geographical matrix (W3).
Empirical results indicate that the implementation of open government data policies significantly promotes urban innovation by stimulating market vitality and enhancing marketization levels to make the market effective, reducing government intervention, optimizing government-business relations, and improving the business environment to promote proactive government action, thus fostering urban innovation through the dual pathways of an effective market and a proactive government. Digital inclusive finance plays a positive moderating role by alleviating financing constraints and providing funding support, while human capital stock amplifies the innovative advantages of open government data by strengthening knowledge spillover effects. Moreover, there is heterogeneity among cities: cities southeast of Hu Huanyong’s line, cities with a higher level of industrial structure upgrading, and large cities exhibit a stronger innovation promotion effect, likely due to more complete digital infrastructure, a more advantageous industrial structure, and stronger human capital, which further confirms the positive moderating effect of human capital stock and digital inclusive finance. In addition, open government data has a significant spatial spillover effect on urban innovation, positively impacting other cities through spatial interactions, and this effect is jointly driven by geographic proximity, economic similarity, and the interaction between the two. Lastly, in the transmission mechanism of public data openness promoting urban innovation capacity, the synergy of an effective market and a proactive government has produced a significant promotion effect.
The theoretical contributions are threefold. First, by conceptualizing open government data as a quasi-experiment, we extend existing innovation theory to encompass data governance frameworks. Second, the integration of effective market and proactive government mechanisms addresses a critical gap in understanding state-market synergies within innovation ecosystems. Third, our spatial and moderating analyses provide nuanced insights into cross-city disparities and multiplier effects. Practically speaking, this study offers valuable guidance for policies related to data openness roadmaps, digital finance integration, and intercity innovation networks, providing an actionable blueprint for fostering high-quality urban development.

With the rapid development of information network technology, enterprises face growing uncertainty in the digital economic environment. To address this challenge, enterprises' innovation networks are expanding outward, creating an urgent need to collaborate with other actors in these networks to establish value co-creation mechanisms. Enterprises, as the micro-foundation of economic operation, are the key subjects of digital innovation. However, for traditional enterprises, there may be many barriers to digital innovation. Therefore, many enterprises form digital innovation networks by cooperating with other enterprises, particularly their digital service providers. They integrate digital innovation resources by embedding in digital innovation networks and accelerate the construction of new value creation models. This phenomenon is defined as digital innovation network embeddedness. Digital innovation networks represent a collaborative ecosystem where different participants, including different network participating enterprises, come together to generate shared value using digital innovation resources. Traditional value co-creation theories emphasize dyadic relationships between enterprises and consumers. Yet, the rapid advancement of digital technology has expanded this concept of value co-creation from dyadic interactions to complex networks involving multiple participants. The scope of value co-creation has thus broadened from firm-consumer interactions to inter-firm collaborations. As a result, digital service providers and client enterprises actively exchange and integrate resources, enabling value co-creation among enterprises embedded in digital innovation networks. Simultaneously, the widespread application of digital technologies has raised concerns about corporate digital responsibility (CDR), such as data privacy and algorithmic ethics. CDR emphasizes the fair and ethical use of digital technologies by organizations when engaging with stakeholders in the digital ecosystem. However, research on CDR remains in its nascent stage, with existing literature primarily comprising conceptual qualitative studies and lacking quantitative investigations into how CDR influences corporate development.
Following social network theory, this study divides digital innovation network embeddedness into two dimensions: digital innovation network structural embeddedness and digital innovation network relationship embeddedness. A theoretical framework is constructed to explore how digital innovation network embeddedness facilitates enterprise value co-creation, with CDR introduced as mediating variables to reveal the new mechanisms for multi-stakeholder value co-creation. Through hierarchical regression and Bootstrap test on 317 sets of corporate questionnaire data, the research findings are as follows. First, digital innovation network structural embeddedness and digital innovation network relationship embeddedness not only directly promote value co-creation but also indirectly enhance it by elevating CDR. Digital innovation network structural embeddedness expands firms' access to heterogeneous digital resources and data elements, strengthening their centrality within the network and thereby fostering resource exchange and risk-sharing. Digital innovation network relational embeddedness improves the transparency of tacit knowledge sharing and reduces collaboration risks through trust and reciprocity mechanisms. Second, CDR partially mediates the relationship between digital innovation network embeddedness and value co-creation. By fulfilling corporate digital responsibility and establishing ethical norms and trust mechanisms, enterprises attract more stakeholders to participate in resource integration and thus achieve value co-creation.
　This research contributes to the literature in the following ways. First, this study extends research on the impact mechanisms of digital innovation network embeddedness, and extends the focus from individual-firm performance to value co-creation among multiple stakeholders. Second, this study deepens the micro-foundations of value co-creation under service-dominant logic and elucidates the mechanisms through which actors embed themselves in digital innovation networks. Third, this study enriches empirical research on CDR and verifies its critical mediating role in how digital innovation network embeddedness empowers value co-creation.At the practical level, this study offers the following insights for enterprises. Enterprises should actively embed themselves in digital innovation networks, occupying structural holes and strengthening partnerships to access non-redundant resources and emotional support. Furthermore, they should prioritize fulfilling CDR and enhance reputation and trust through transparent digital practices.

With generative artificial intelligence (AIGC) becoming deeply integrated into human life, people are increasingly unable to distinguish between AI-generated synthetic content and human-created material. China's Measures for the Identification of AI-Generated Synthetic Content (hereinafter referred to as the “Measures”) officially came into effect on September 1, 2025. In addition to specifying the identification obligations of providers of AI-generated synthetic services, the Measures also clarify the responsibilities of users of such content, online information dissemination service providers, and application distribution platforms. Through the “Exception to Explicit Identification” clause in Article 9, the Measures aim to strike a better balance between facilitating the use of generated content and maintaining oversight of the information ecosystem.
This study conducts an interpretive analysis of the regulatory cluster on identification obligations, including the "Measures", to address the urgent compliance needs arising from the current implementation of the "Measures". Moreover, it fills the gap in the current research regarding the lack of law and economics analysis, clarifies the functional limitations of the identification system in terms of efficiency, and expounds that the interpretation of identification obligations should follow the principle of "cost reduction and efficiency improvement". What's more, it reflects on some theoretical viewpoints that currently support the legitimacy of identification obligations and examines the interpretive limits of "the theory of product information disclosure obligation". Lastly, it conducts a detailed analysis of two interpretive approaches to the "Exception to Explicit Identification" in Article 9 of the "Measures", providing solutions for the implementation and optimization of the "Measures".
At present, there is an overly optimistic perception of the functional effectiveness of the identification system, leading to a tendency to expand the scope of supervision. However, the progress of substantive transparency in China's identification system remains limited. The identification of generated content is prone to problems such as devaluation of information value and weakening of the contributions of human co-creators, and undermines users' and platforms' willingness to comply with identification requirements. Meanwhile, the identification of generated content has functional limitations: it cannot replace judgments on information quality, authenticity, or copyright, and is easily tampered with, which affects the function of traceability supervision. From the cost-benefit perspective, if the social welfare brought by identification is limited by "formal transparency" and "watermark tampering", and is lower than social costs such as technology construction, value devaluation, and dispute resolution, the legitimacy of the system will be difficult to justify. In an environment of "prevalent non-compliance", high rigid penalties are not only ineffective but also incur high law enforcement costs.
To promote the long-term development of the identification system, it is necessary to establish a legal and economic analysis mindset, adhere to the principle of maximizing functional effectiveness, promote the transition to substantive transparency, and conduct substantive labeling of the extent and methods of AI participation. It is essential to reflect on the limitations of the product information disclosure theory and emphasize the collaborative fulfillment of identification obligations by multiple subjects. The principle of differentiated interpretation for cost control should be upheld: requirements should be adjusted according to differences in subjects and risk scenarios, and explicit identification can be reduced or canceled in low-risk scenarios. The identification obligations of platforms shall be supervised by administrative authorities, relying on the four-dimensional framework of "legal provisions - departmental regulations - national standards - platform rules"; the obligations of users shall be mainly supervised by platforms, which need to provide convenient tools and prompts, and establish a proportional sanction and appeal mechanism. The interpretation of Article 9 of the "Measures" needs to shift from the "supervision-oriented perspective" to the "development-oriented perspective". As a common exception to relevant provisions, explicit identification should only be mandatory in high-risk scenarios, while in other scenarios, it may not be mandatory and can be determined in accordance with user agreements.
Further efforts should be made to improve empirical research and quantitative analysis on the implementation effects of the AIGC identification system, refine the standards for AIGC risk classification, and improve the specific provisions of penalties such as administrative penalties for identification obligations applicable to platforms and users. In addition, the relationships between AIGC identification, copyright, and fair use of data need to be further clarified.

Against the background of global economic slowdown and geopolitical frictions, enterprises are facing increasing pressure on survival and development. Organizational resilience has become a key capability for firms to cope with adversities and achieve sustainable development, and firms are placing more emphasis on building organizational resilience to ensure survival and long-term growth, yet balancing resilience and cost advantage constitutes a significant dilemma for enterprises. From the perspective of the Resource-Based View (RBV), enterprises choose strategic alliances mainly due to the lack of key internal resources. However, enterprises with strong organizational resilience can independently adjust resource allocation and respond to crises, which may reduce their dependence on strategic alliances. Although strategic alliances offer advantages such as external resource support and synergy (which can help enterprises maintain cost advantages), they also pose problems like high governance costs and collaboration risks. Therefore, enterprises with high resilience may tend to rely on internal resource optimization rather than external alliances when weighing costs and benefits.
Organizational resilience exerts a two-sided impact on cost advantage. On one hand, it can help enterprises avoid high-cost losses and create cost advantages. On the other hand, cultivating and maintaining resilience requires substantial resource investment, which may increase corporate costs and impact the implementation of cost leadership strategies.Scholars hold divergent views on the relationship between organizational resilience and cost leadership strategy. Additionally, the specific impact of the interaction between organizational resilience and strategic alliances on cost advantage also needs further exploration. Controversies exist in relevant fields, calling for more in-depth research.
Therefore,drawing on the resource-based view (RBV), this study examines the effect of organizational resilience on cost leadership strategy. It also explores the underlying mechanisms using panel data from Chinese A-share listed companies between 2010 and 2022. The results show a significant negative effect of organizational resilience on cost leadership strategy. This reveals a possible “dark side” of resilience. Mechanism analysis shows that highly resilient firms rely less on strategic alliances. This limits their access to external cost-optimizing resources and makes it harder to maintain cost leadership. Further analysis divides alliances into equity-based and contractual forms. Resilience significantly reduces participation in equity-based alliances, which help create cost advantages through capital investment, economies of scale, and resource sharing. In contrast, resilience has no significant effect on contractual alliances, which have limited potential for deep resource integration. Instead, resilient firms focus on internal redundancy and adaptive capacity. While this improves risk resistance, it can reduce resource efficiency and increase operating costs.
The study further uses a multi-level contingency framework to explore boundary conditions. At the micro level, managerial myopia strengthens the negative effect of resilience on cost leadership, as short-term oriented managers avoid long-term cost control investments. At the meso level, strategic orientation plays a key moderating role. Growth orientation weakens the negative effect by improving resource acquisition and legitimacy. Profit orientation strengthens it by increasing cost pressures and resource hoarding. At the macro level, market competition weakens the negative effect of resilience on alliance formation. In competitive markets, resilient firms are more likely to form alliances for survival.
This study makes three contributions. First, it challenges the idea that resilience is always beneficial by showing its trade-offs with cost leadership. Second, it extends the RBV by examining both the presence and type of strategic alliances as mediators. This shows that even strong firms may forgo alliance-based cost advantages. Third, it enriches the contingency view by showing how managerial cognition, strategic priorities, and market dynamics together shape the effect of resilience strategies.
The findings have practical implications. Firms should avoid over-investing in resilience without considering cost efficiency. They should adopt alliance strategies suited to their resilience level and match resilience investments to their strategic orientation. In competitive markets, balancing internal flexibility with external cooperation is essential for staying competitive without losing efficiency. This study offers a nuanced view of resilience by revealing its trade-offs, boundary conditions, and governance implications, and lays a foundation for future research on how firms can achieve both resilience and efficiency.

In recent years, global climate change has become increasingly severe, with China actively assuming major-country responsibilities as one of the principal contributors to worldwide carbon reduction. Under the "Dual Carbon Goals" framework, the government vigorously promotes green transformation in the manufacturing sector, advances new industrialization, and cultivates innovative green development models. To achieve win-win outcomes in both economic and environmental performance, manufacturing enterprises have begun experimenting with business model innovations (BMI), such as servitization transformation and circular economy practices. Given its strategic and systemic nature, BMI not only provides manufacturing enterprises with more adaptable production methods and operational patterns but also serves as a potential breakthrough in driving green development and reducing carbon emissions through reconstructing value creation and delivery processes.
Existing studies predominantly focus on the economic impacts of BMI through questionnaires or case studies, while rarely investigating its role in curbing corporate carbon emissions from an enterprise perspective. However, BMI can optimize resource allocation and reduce resource consumption during value creation and delivery processes, making it crucial to explore whether BMI effectively inhibits corporate carbon emission intensity. This study addresses three core questions: (1) Through what pathways does BMI influence carbon emission intensity? (2) How does external environmental regulation affect the relationship between BMI and carbon emission intensity? To answer these questions, the study conducts empirical research using data from Chinese listed manufacturing enterprises between 2012 and 2022.
Grounded in Resource-Based View (RBV) and Dynamic Capability Theory (DCT), this study develops a composite BMI measurement system incorporating subjective and objective indicators using the entropy weight method. It further examines BMI's impact on carbon emission intensity, while investigating the mediating roles of green innovation and digital transformation, along with the moderating effect of local environmental regulation intensity. The findings demonstrate the following: (1) BMI significantly reduces corporate carbon emission intensity, and this conclusion remains valid after a series of robustness tests; (2) The emission-reduction effect is more pronounced in state-owned enterprises, heavily polluting industries, and large-scale firms; (3) Green innovation and digital transformation partially mediate the BMI-emission intensity relationship. Moreover, the mediation role of green innovation is primarily evident in high-quality innovation, namely the output of green invention patents; (4) The moderating effect of local environmental regulation intensity on the relationship between BMI and firm carbon emission intensity is context-dependent. This negative moderating effect is contingent upon firms’ resource endowments: while statistically insignificant in the full sample, it manifests significantly among small-sized firms or firms with low R&D investment.
In light of these findings, the study derives managerial and policy implications as follows. For businesses, it is essential to prioritize the role of business-model innovation on the road to sustainable development by integrating it into strategic dimensions such as carbon-emission governance. By building an innovation matrix of ecological value propositions—intelligent value creation— value delivery circulation, enterprises can achieve the co-evolution of economic and environmental benefits. For governments, the task is to leverage macro-regulatory capacity, balancing the stringency of rules with room for innovation, so as to accelerate corporate green transition and digital upgrading. Policies should be fine-tuned for specific targets through differentiated supervision, guiding enterprises to convert compliance pressure into long-term drivers of low-carbon transformation.
This study contributes theoretically in three dimensions: First, by employing the entropy weight method to measure corporate BMI, it enriches interdisciplinary research on BMI and carbon emission drivers at the micro level. Second, this study constructs the transmission chain of "business model innovation→green innovation/digital transformation→carbon emission intensity", and validates the critical mediating role of substantive green innovation. Third, regarding the moderating effect of local environmental regulation intensity, this study reveals its pronounced context-dependency. It highlights the central role of enterprises’ resource endowments within the regulatory mechanism triggered by external environmental pressures, thereby broadening the theoretical perspective for environmental institutional theory in the context of business model innovation.

Artificial intelligence has been deeply embedded in the network of social operations, serving as the underlying architecture and infrastructure of the digital society. Academia has conducted in-depth typological analyses of data risks in generative artificial intelligence from various perspectives and methodologies, including the life cycle perspective, rights theory perspective, and data security perspective, and has proposed corresponding measures to address these data risks. However, a common limitation of the aforementioned life cycle perspective, rights theory perspective, and data security perspective lies in the disconnection between theoretical logic and practical dilemmas, specifically, they fail to fully recognize the unique characteristics of data risks in generative AI and systematically examine the complex feedback mechanism between training data and generated content. Consequently, these perspectives are trapped in an internal perspective of data risk governance, neglecting the spillover effects, transmissibility, and interconnection inherent to generative AI data risks themselves. Overall, academic analyses of generative AI data risks have long been confined to the cognitive limitation of "centering on governance objects",that is, overemphasizing the attribute characteristics of "data" and the morphological manifestations of "data risks" while ignoring the core attributes and process mechanisms of "governance" itself. This has rendered existing analytical approaches and models unable to respond to the complexity of governance practices, ultimately reducing governance schemes to mere "paper compliance" that cannot be effectively implemented. Therefore, it is necessary to take governance theory as the starting point, analyze the practical dilemmas in generative artificial intelligence data risk governance, and propose targeted optimization schemes for such governance.
Professor Gerry Stoker has further refined a "governance" theory that can provide an organizational framework. "Governance as theory" emphasizes four key aspects: the clarity of governance goals, the synergy of governance subjects, the completeness of governance basis, and the flexibility of governance means. The governance dilemmas in generative AI data risks can thus be categorized into four dimensions: ambiguity of governance goals, fragmentation of governance subjects, absence of governance basis, and rigidity of governance means. Accordingly, the governance dilemmas in generative artificial intelligence data risks can thus be categorized into four major types: the dilemma of misaligned governance goals, which includes difficulties in balancing security and development, reconciling national and corporate goals, and aligning short-term and long-term objectives; the dilemma of fragmented governance subjects, which involves complex games among public authorities, the binary opposition between the state and enterprises, and international competition for rule-making power; the absence of governance basis, which covers the lack of laws and regulations, industry standards, and ethical norms; and the rigidity of governance means, which includes the path dependence on traditional administrative tools, institutional obstacles to technological empowerment, and the breakdown of synergy among multiple tools.
In response, system synergy, dynamic adaptation, risk communication, and multi-stakeholder co-governance shall serve as core concepts. Goal calibration is to be achieved through a risk-classified dynamic balance mechanism, a synergistic coupling mechanism of government-enterprise interests, and a temporal cohesive mechanism for short-term and long-term goals; subject synergy is to be realized through a collaborative linkage mechanism among public authorities, a co-governance operation mechanism for government-enterprise symbiosis, and a collaborative mechanism for international rule-making; the governance basis is to be improved through a hierarchical legislative legal guarantee mechanism, a multi-stakeholder co-governance standard-setting mechanism, and an institutionally embedded ethical constraint mechanism; and governance means are to be innovated through a precision-oriented reform mechanism for administrative tools, an institutional activation mechanism for technological tools, and a synergistic coupling mechanism for multiple tools. Furthermore, through in-depth coupling of governance goals, subjects, foundations, and means, as well as systematic reconstruction of interrelated mechanisms, continuous innovation in the generative artificial intelligence data risk governance system and a qualitative leap in governance mechanisms can be ultimately promoted, laying a solid foundation for the healthy and sustainable development of generative artificial intelligence.

In the era of the digital economy, data has evolved from an auxiliary information resource into a core production factor driving technological innovation and sustainable development. Understanding how data factor empowers firms to achieve balanced exploitative and exploratory innovation,thus enhancing their ambidextrous innovation capability,has become an essential question in both academic research and policy practice. Existing research primarily examines corporate innovation through the lenses of digital economy, digital transformation, and digital technology application, spanning macro, industrial, and digital capability levels. Recent studies have begun to explore the nexus between data factor and corporate innovation, highlighting their promotional effects, potential risks, and relevant theoretical models. However, significant deficiencies persist regarding mechanisms, contextual factors, and theoretical integration. Specifically, research on data factor and corporate ambidextrous innovation exhibits three major gaps: first, a lack of systematic characterization based on micro-level firm data; second, insufficient elaboration on how data factor embeds within the innovation process and exert differential impacts; and third, inadequate empirical testing that incorporates external environmental dynamics and firm heterogeneity, thereby limiting the generalizability of conclusions and practical implications.
Thus, this study aims to systematically explore the mechanisms and boundary conditions through which the application of data factor influences firms' innovation performance in both exploitative and exploratory dimensions. Grounded in the resource-based view and innovation management theory, this paper constructs a dual-path framework of “stock resource activation” and “incremental resource integration”. It argues that data factor application can stimulate exploitative innovation by identifying and activating slack internal resources, while promoting exploratory innovation by strengthening external R&D collaboration networks. To empirically test these mechanisms, a two-way fixed effects model is established using panel data from 3 076 Chinese A-share listed firms between 2008 and 2023, yielding 29 900 firm-year observations. The level of data factor application is measured through text mining of corporate annual reports across four dimensions: data stock, data development capability, data-driven business applications, and data monetization. Exploitative and exploratory innovation are distinguished by patent classification continuity, with slack resources and R&D collaboration as mediating variables.
Empirical results show that the application of data factor significantly promotes both exploitative and exploratory innovation, with a stronger effect on exploitative innovation, suggesting that data utilization primarily enhances incremental rather than radical innovation. Mechanism analysis confirms two distinct transmission paths: data factor application reduces slack resources by improving resource recognition, allocation, and utilization efficiency, thereby enhancing exploitative innovation; simultaneously, it facilitates exploratory innovation by optimizing partner identification, collaboration modes, and knowledge integration efficiency, underscoring the role of data in enabling heterogeneous resource synergy. Heterogeneity analysis shows that these effects are more pronounced under favorable institutional and regional conditions. Specifically, the establishment of national big data pilot zones and the economic advantages of eastern China significantly strengthen the positive relationship between data factor application and dual innovation outcomes.
This study makes three primary contributions. First, it advances micro-level research on data factor by integrating them into the analysis of firm-level innovation behavior, extending the theoretical frontier of data economy studies. Second, it deepens the understanding of how data-driven processes reshape firms' resource allocation mechanisms, providing empirical evidence that data applications can transform slack resources into innovation assets and foster cross-organizational knowledge recombination. Third, it identifies the heterogeneous effects of institutional environments and regional development levels, offering new insights into how data infrastructure and policy design can enhance firms' innovation capability.
From a managerial perspective, the findings suggest that firms should strengthen their data management and analytics capabilities to embed data-driven decision-making throughout innovation activities. For exploitative innovation, data can be used to optimize internal processes and accelerate incremental improvement; for exploratory innovation, big data and collaborative platforms can help identify emerging opportunities and reduce innovation uncertainty. Policymakers are advised to accelerate data market reforms, promote cross-regional data circulation, and build supportive institutional frameworks,especially in underdeveloped areas,to ensure equitable access to data-driven innovation opportunities.

The promotion of carbon emissions reduction has become an urgent priority. Currently, it is essential to transform the economic growth model from the traditional, resource-intensive approach to a more intensive, sustainable one. This shift relies heavily on green innovation to enhance economic quality and global competitiveness. While studies have demonstrated that China's carbon market has played a positive role in reducing carbon emissions, the extent to which it can improve the green innovation efficiency of micro-enterprises remains an area that requires in-depth exploration.To address this question, the theoretical mechanisms through which carbon emissions trading policies impact a firm's green innovation efficiency are examined. On one hand, the carbon allowance system has increased the operating costs for enterprises. To reduce carbon emissions, enterprises must undergo a green transformation and sell their surplus carbon allowances to offset their R&D investments. On the other hand, enterprise behavior is also influenced by the reward and punishment mechanisms associated with carbon emissions trading. First, the pressure of carbon allowance has forced some high-energy-consuming enterprises to transform or abandon their operations due to financial pressure. Thus, some high-energy-consuming industries are gradually replaced by low-energy-consuming industries. Second, under the incentive policy, to obtain resources tilted by the Government, enterprises take the initiative to improve the quality of infrastructure service units through green innovation and realize the quality improvement of processes and products. Third, the reward and punishment mechanism not only help to release investment but also increases the mobility of production factors such as personnel and technology, which helps to optimize the allocation of resources and coordination capacity among industries. Therefore, carbon emissions trading can also enhance the enterprises' green innovation efficiency by optimizing the transformation of industrial structure.
This paper selects industrial enterprises listed on China's A-share market from 2006 to 2023 as research samples, with 7 596 observations from 422 listed companies. The green innovation efficiency of industrial enterprises is measured by the Super-SBM model, the carbon emissions trading pilot policy is taken as an exogenous shock variable, and a multi-period double-difference model is constructed to explore the impact of the carbon emissions trading pilot policy on the enterprises' green innovation efficiency. This paper selects industrial enterprises listed on China's A-share market from 2006 to 2023 as research samples, with 7 596 observations from 422 listed companies. The green innovation efficiency of industrial enterprises is measured by the Super-SBM model, the carbon emissions trading pilot policy is taken as an exogenous shock variable, and a multi-period double-difference model is constructed to explore the impact of the carbon emissions trading pilot policy on the enterprises' green innovation efficiency. Meanwhile, industrial structure transformation is introduced into the model as a mediating variable to further explore the impact mechanism. In addition, the heterogeneity of the impact is discussed.
The findings are presented as follows. First, the carbon emissions trading pilot policy significantly enhances enterprise green innovation efficiency, and this conclusion remains robust across a series of sensitivity analyses. Second, both industrial structure upgrading and rationalization play a positive mediating role in the relationship between the carbon emissions trading pilot policy and enterprise green innovation efficiency. Third, heterogeneity analysis indicates that the impact of carbon emissions trading on green innovation efficiency varies across different contexts. Specifically, the effectiveness of the policy is influenced by factors such as whether the firm is located in a key environmental city, and the ownership structure of the enterprise. These factors contribute to differentiated policy outcomes. Collectively, these findings provide empirical evidence to objectively assess the green effect of the carbon trading market.
To further advance the green and low-carbon transformation of enterprises, several key measures are proposed. First, it is essential to continuously improve the safeguard measures for enterprises during their green and low-carbon transformation. This includes enhancing market stability, expanding financing channels, and providing macro-level information and technical support.Second, regions should further optimize carbon allowance pricing mechanisms to reduce market uncertainties caused by price volatility. When formulating carbon allowance allocation systems, regions must balance the survival and development of enterprises with the effectiveness of carbon emission governance. Third, mechanisms for incentivizing and supervising the green and low-carbon transformation of non-state-owned enterprises should be strengthened to ensure their active participation and compliance.Finally, policies should be continuously refined to promote industrial structure transformation, thereby enhancing the overall efficiency and sustainability of green innovation.

Led by the US, developed countries have modified supply policies towards China, seeking technological decoupling via a "de-Chinaized" supply chain. In response, China has introduced a “double-cycle” development strategy. Enterprises are the core forces to cope with changes in the trade pattern, and ensuring the resilience of enterprise supply chains and security is the top priority for realizing high-quality economic development. Currently, there are few studies in the academic community that utilize U.S. policy of technology decoupling from China as an empirical target to investigate the resilience of enterprise supply chains. This research gap could potentially hinder China's ability to manage the risks inherent in a volatile global market. Therefore, in the context of China's new round of opening-up policy, it is important to explore how U.S. policy of technology decoupling from China affects the resilience of enterprise supply chains, which provides a basis for how enterprises can enhance supply chain resilience in the complex international situation. This research is of great theoretical significance and practical value for the promotion of high-level opening to the outside world in line with the double-cycle development strategy.
To explore the impact of U.S. policy of technology decoupling from China on the resilience of enterprise supply chains, this study utilizes a sample of A-share listed firms from the period 2010-2022 to investigate this impact. The starting time of the US-China technology decoupling policy is identified as 2018. The impact of the US-China tech-nology decoupling policy on the cooperation between enterprises and their upstream and downstream partners is significant. Specifically, customer sales (Y1) and supplier procurement (Y2) are considered as the dependent variables.
The analysis yields some findings . First, the U.S. policy of technology decoupling from China significantly enhances the resilience of enterprise supply chains, and this finding remains reliable after the robustness test. Second, the heterogeneity analysis shows that the resilience of enterprise supply chains is more significantly positively affected by U.S. technology decoupling from China for the non-state-owned enterprises, eastern region, and high-tech industry categories. Third, further research has found that the U.S. policy of technology decoupling from China promotes corporate supplier procurement volumes by driving supply chain efficiency. Additionally, this policy enhances the resilience of enterprise supply chains by increasing supply chain concentration. However, the U.S. policy of technology decoupling from China also inhibits the improvement of resilience of enterprise supply chains through the complexity of supply chain collaborations.
This article makes novel contributions to the research on enterprise supply chain resilience. Firstly, it offers fresh empirical evidence, a significant advancement since the current literature is heavily reliant on qualitative analyses and lacks empirical studies on the relationship between the U.S. policy of technology decoupling from China and supply chain resilience. By integrating both qualitative and quantitative analysis methods, this paper uncovers this relationship, thereby providing valuable empirical evidence for management science research on supply chain resilience. Secondly, it expands and enriches the theoretical research on U.S. policy of technology decoupling from China and resilience of enterprise supply chains. Most of the existing studies examine this issue from the macro level of industry and national strategy. In this paper, from the micro level of enterprises, we discuss how the enterprise supply chains can cope with the unfavorable impacts brought by the changes in the international situation, so as to enrich the theoretical content of the relevant research. Lastly, the paper provides actionable strategies to bolster the resilience of Chinese enterprise supply chains. Given the frequent occurrence of unforeseen events in science and technology policies and international relations, and the scarcity of studies on the influence mechanisms and operational frameworks behind these uncertainties, this paper elucidates the mechanism of U.S. policy of technology decoupling from China affecting the resilience of enterprise supply chains, and provides effective practical insights for enterprises to enhance the resilience of supply chain.

With the in-depth development of globalization and digitalization, companies face increasingly complex and dynamic competitive environments. Knowledge, as the core resource of modern enterprises, has rendered the innovation of its management model crucial to sustainable development. Technological innovations not only greatly enhance the efficiency of knowledge acquisition, transformation, and application but also drive knowledge management toward intelligence, dynamism, and collaboration. As a result, how companies can drive the innovation of knowledge management models has become a pressing issue for both academia and industry. Existing literature mainly studies knowledge management model innovation from a linear perspective and case studies, but it fails to clarify the causal differences of multiple factors in the coordinated and unified process of achieving knowledge management model innovation, nor does it reveal the driving pathways of knowledge management model innovation. Therefore, exploring the antecedent variables and driving pathways of knowledge management model innovation is of practical value. This paper seeks to answer the following questions: How are the antecedent conditions of knowledge management model innovation interlinked? Which condition among the antecedent factors contributes the most to knowledge management model innovation?
This study adopts actor-network theory (ANT) to construct a research framework for knowledge management model innovation. Taking 202 valid questionnaire responses from high-tech enterprises as the research data, it employs fuzzy-set qualitative comparative analysis (fsQCA), importance-performance map analysis (IPMA), and artificial neural network (ANN) analysis to examine the impact of factors at different levels on enterprises' knowledge management model innovation and the configurational effects among these factors.The FsQCA identifies conditional combinations between variables and reveals different causal paths, making it suitable for exploring multiple causal relationships in complex systems. IPMA helps identify important influencing factors through visual means and quantitatively analyzes the actual performance of these factors. Through training on vast amounts of data, ANNs are capable of capturing non-linear relationships between variables, thereby enabling the prediction of future trends and outcomes.This combination not only enhances the depth of causal analysis, but also improves prediction accuracy, making it suitable for multidimensional research on complex systems.
The results of the study show that (1)a single condition cannot fully explain the driving mechanisms of knowledge management model innovation;(2) six antecedent conditions from the levels of technological actors, human actors, and coordinating actors have a synergistic impact on knowledge management model innovation,the paths leading to complementary high-level knowledge management model innovations can be categorized into three types: technology-coordination dual-core, human-machine synergistic interaction, and technology-led self-driven;(3) AI + knowledge application and the digital capabilities of organizational members can substitute for each other under certain conditions;(4) the key preconditions for knowledge management model innovation mainly lie at the technological actor level, with AI + knowledge transformation being the core condition for achieving high-level knowledge management model innovation.
The contributions of this study are as follows: First, compared to the previous method of linear path analysis, this study identifies multiple implementation paths for enterprises to achieve high-level knowledge management model innovation under different configuration conditions from a configuration perspective, revealing the driving mechanism of multi factor interaction coupling on knowledge management model innovation. Second, different from the TOE framework commonly used in configuration research, the introduction of ANT integrates technical actors, human actors, and collaborative actors into the same analysis network, with a network logic proposed for innovative construction of knowledge management models under human-machine collaboration, providing a new exploration path for the localized application of ANT in management contexts. Finally, this study proposes that technology actors, human actors, and collaborative actors jointly constitute the core driving force for knowledge management model innovation, and further analyzes the substitutability of technology actors and human actors in specific configurations, responding to the lack of in-depth exploration of human-machine collaboration in existing research.

In the era of "big science", the scientific research paradigm is undergoing a profound transformation from decentralization to concentration, making scientific teams the core unit for addressing complex challenges and achieving major innovations. In response to the significant demand of China's "Organized Research Paradigm" strategy for building stable and highly efficient teams, systematically reviewing and integrating the vast body of global research on scientific team structure and innovation effectiveness, and clarifying its developmental trajectory and core themes, it is not only theoretically urgent but also a crucial prerequisite for providing a solid academic foundation for national strategic practice.
To this end, this study utilizes bibliometric tools to analyze 4 541 key publications from both domestic and international sources between 2000 and 2024, systematically delineating the development process of global research on scientific teams. The analysis indicates that this field has formed a global research landscape led and dominated by European and American scholars. International research started early, possesses a mature system, and continues to deepen its thematic focus, showing a clear evolution from surface-level characteristics to underlying mechanisms. In contrast, although China has experienced exponential growth since 2015 and become a significant research force, Chinese literature still shows noticeable gaps compared to the international frontier in terms of the systematic nature and depth of topics, as well as journal support. Overall, Chinese research exhibits certain characteristics of lag and fragmentation.
Further focusing on the analytical framework of "team identification - structural elements - innovation effectiveness", the study sorts out and reveals research hotspots and findings. In team identification research, the approach has evolved from relying on entity organization lists to using co-authorship networks for identification, forming two main identification paths: entity teams and virtual teams. The former offers reliable information but is difficult to obtain, while the latter is scalable and efficient but susceptible to interference from temporary collaborations. Regarding structural elements, research has expanded from explicit structural elements to implicit structural elements and gradually developed complex model relationships. For instance, implicit elements often act as key mediating variables, regulating the complex mechanism through which diversity affects innovation effectiveness. In terms of innovation effectiveness, research has constructed differentiated evaluation systems: academic teams focus on academic impact, while corporate teams prioritize market value. However, both types of evaluation face the common challenge that traditional quantitative indicators struggle to comprehensively measure the comprehensive contribution of teams in major scientific and technological breakthroughs. This systematic review reveals the development trajectory and core bottlenecks in research methods and theories within this field.
Finally, responding to the demands of China's "organized research paradigm" for stable and highly efficient scientific teams, future research should focus on the following areas: Regarding the focus of research objects, attention should shift from temporary virtual teams to teams with stronger entity attributes, such as those driven by academic leaders or research platforms, to align with the requirement for long-term, stable collaboration in major scientific and technological challenges. Concerning the optimization of structural elements, it is necessary to break through the superficial analysis of existing explicit structures by introducing topological and complex network methods, while simultaneously overcoming the quantitative bottlenecks of implicit structural elements to achieve the transformation from abstract concepts to operable parameters. Pertaining to the evaluation of innovation effectiveness, a multi-dimensional evaluation system aligned with strategic orientations should be constructed, incorporating strategic outputs like national science and technology awards and major engineering breakthroughs into core indicators, and establishing models linking resource input to innovation effectiveness. This will provide solid theoretical support and practical guidance for forging scientific teams capable of undertaking major national missions.
The significance of this study lies in its systematic review of global research hotspots and evolving trends in scientific teams. Building on this foundation, it addresses the new requirements for team building under China's " Organized Research Paradigm" strategy, integrating cutting-edge international topics with the national strategic context. This research pinpoints critical directions for future investigation,including research subjects, structural elements, and innovation effectiveness,thereby offering targeted insights for constructing a theoretical framework to support scientific teams tasked with major national S&T missions.

Artificial intelligence (AI) has emerged as a pivotal driving force for scientific and technological breakthroughs. However, its application effects in corporate practices show significant differences:some enterprises have successfully integrated AI deeply into their core business, while others have made huge investments but failed to achieve substantial breakthroughs.
Existing studies mostly regard artificial intelligence as a passive technical resource or tool, ignoring its evolutionary role as a subject that participates in management activities through autonomous decision-making. From the perspective of resource orchestration theory, AI has shown the potential of a resource manager, being able to identify opportunities in real time, allocate elements, and collaborate or conflict with human decisions, which directly affects the pace, direction, and risk tolerance of innovation. Current research has three shortcomings: first, it fails to fully recognize the attribute of AI as a management subject; second, it lacks a systematic analysis of the path through which AI is embedded in the resource management process; third, it ignores the conflict mechanism between humans and intelligent agents in terms of resource dominance.
Drawing on the resource orchestration theory, this paper employs panel data from 3 131 Chinese firms over the period from 2014 to 2023 to investigate the dual roles of AI in shaping radical innovation—both as an innovation resource and as a resource orchestrator. Specifically, it addresses three critical questions: (1) Does AI as an innovation resource have a nonlinear effect on firms′ radical innovation? (2) How does AI, acting as a resource orchestrator, impact radical innovation through the resource management process? (3) How do human researchers moderate these relationships through interaction with AI?
The empirical findings reveal several important conclusions. First, AI as an innovation resource demonstrates an inverted U-shaped relationship with radical innovation. Specifically, the adoption and initial integration of AI resources significantly promote radical innovation by facilitating rapid knowledge recombination, reducing innovation costs, and enhancing the precision of decision-making processes. However, beyond a certain threshold, excessive reliance on AI tends to lock firms into existing technological trajectories, leading to diminishing returns and even inhibiting breakthrough innovation.Second, As a resource orchestrator, AI exerts its influence primarily through resource structuring and capability enhancement. Through automated data integration and analysis, AI strengthens firms′ resource structuring by accelerating and refining the identification of innovation opportunities and facilitating efficient knowledge recombination. However, the impact of AI on resource capability enhancement is nuanced. Initially, the introduction of AI resources may trigger internal resource allocation conflicts, temporarily undermining operational efficiency. Nevertheless, as firms progressively increase AI utilization, operational efficiency improves significantly, enhancing their capability to undertake radical innovation projects. Conversely,the study finds limited evidence of AI′s effectiveness in resource leveraging, suggesting current AI technologies remain inadequate for handling complex managerial tasks requiring sophisticated judgment, political negotiation, or contextual insight.Third, the study explores the interactive dynamics between AI and R&D personnel. The results highlight a clear substitution relationship: AI′s beneficial impact on radical innovation is stronger when the number of human researchers is limited, indicating that AI effectively compensates for insufficient human expertise in generating innovative ideas and strategic decisions. As the number of R&D personnel grows, however, this marginal benefit declines due to increased organizational complexity and potential resistance arising from perceived job threats, limiting AI′s incremental innovation benefits. Additionally, the strength of this substitution effect varies across firm lifecycle stages, being most pronounced during the growth phase and diminishing during maturity and decline phases.
This study contributes to the existing literature in several significant ways. Theoretically, it enriches the resource orchestration framework by explicitly considering AI transition from a passive, orchestrated resource to an active resource orchestrator. The study also expands current understandings of human-AI interactions by introducing the concept of "positional competition", suggesting that AI and human researchers may compete rather than collaborate in certain innovation processes. Empirically, this approach is novel in systematically analyzing
the complex nonlinear impacts of AI, considering the dual roles of AI and moderating factors such as human researchers and firm lifecycle stages.

The deep integration of artificial intelligence (AI) with the real economy is critically important for cultivating endogenous innovation capabilities within Chinese enterprises, which is a prerequisite for achieving China's national goal of high-level self-reliance and strength in science and technology. This study provides a rigorous empirical evaluation of China's National Artificial Intelligence Innovation and Development Pilot Zone, a major industrial policy initiative. It specifically investigates the policy's dual impact on corporate innovation, meticulously examining its efficacy in promoting substantive technological advancements while also exploring its potential to inadvertently encourage strategic "innovation bubbles". The analysis uncovers the underlying causal mechanisms, heterogeneous responses across different types of firms, and any unintended negative consequences. The findings are intended to offer a comprehensive evidence base for policymakers to refine AI industrial strategies and reduce associated risks.
To achieve these objectives, this study employs a robust quasi-experimental research design, leveraging the phased, quasi-random rollout of the pilot zones across China from 2007 to 2023 as a series of exogenous policy shocks. This temporal and geographical variation allows for a credible identification strategy. The policy data are meticulously combined with a detailed longitudinal micro-dataset covering all A-share listed companies in China over the same period. The constructed dataset is highly comprehensive, integrating firm-level financial statements, patent application and grant records from the State Intellectual Property Office, R&D expenditure disclosures, and key regional indicators such as ICT infrastructure and university density. The primary econometric methodology is a multi-period difference-in-differences model, which effectively isolates the causal impact of the pilot zone policy by comparing the evolution of innovation outcomes between treated firms and a carefully selected control group of firms before and after the policy's implementation. To ensure the robustness and credibility of the causal inferences, the study employs a battery of stringent tests. These include parallel pre-trend analyses, placebo tests using fictitious policy announcement dates, and alternative variable constructions to mitigate measurement error concerns.
The empirical findings yield several nuanced and critical insights. First, the analysis confirms that the pilot zone policy has led to a statistically significant and economically meaningful increase in the output of high-quality AI-related patents among affected enterprises. This result validates the policy's primary objective of stimulating corporate engagement in cutting-edge AI research and development. Second, the analysis reveals significant heterogeneity in the policy's effects. The innovation-enhancing impact is markedly more substantial for enterprises that initially had lower levels of digital transformation, weaker pre-existing technological endowments, and those located in urban clusters with less developed intelligent infrastructure. This pattern suggests that the policy functions effectively as a form of compensatory support, providing a crucial "catch-up" mechanism for less-advanced firms and helping to reduce regional innovation disparities. A detailed mediation analysis further elucidates the dual channels through which the policy operates: it directly stimulates an increase in internal R&D investment by mitigating financial constraints and signaling strategic importance, and it indirectly fosters deeper and more formalized collaborative ties between industry, universities, and research institutions. This latter channel facilitates critical knowledge spillovers, access to specialized talent, and the shared use of advanced research infrastructure.
Furthermore, while the policy demonstrates clear success in boosting measured innovation, a more granular investigation uncovers the emergence of "innovation bubbles" in certain contexts. This phenomenon manifests as strategic corporate behaviors aimed at signaling compliance with policy goals rather than pursuing genuine technological breakthroughs. It includes both "execution-side" bubbles, where firms invest in superficial AI projects disconnected from their core business to secure subsidies, and "acquisition-side" bubbles, characterized by a focus on patent quantity over quality. This critical finding highlights the complex interplay between strong policy incentives and corporate governance, underscoring the need for sophisticated policy design that rewards substantive innovation rather than mere metric inflation. The study concludes by discussing the broader implications of these findings for innovation policy theory and practice, offering concrete recommendations for designing more effective and resilient AI innovation ecosystems that can truly empower the real economy.

As a vital component of data resources, public data is increasingly recognized for its strategic role in driving innovation. Governments at various levels in China have actively advanced platform development, initially establishing a two-tier public data openness system at the provincial and municipal levels. However, objective disparities exist between these two levels in terms of data coverage, data quality, and policy enforcement. Compared to independent innovation, enterprise joint innovation relies more heavily on external data resources and can enhance the efficiency of cross-organizational resource integration, thereby fostering the growth of new quality productive forces.
Existing literature primarily focuses on the overall effects of public data openness on enterprise innovation and digital transformation, with limited attention paid to its role in promoting joint innovation. In particular, there is a lack of systematic analysis regarding the differences in policy effects between provincial and municipal platforms. Therefore, this study extends the analytical perspective to explore how two-level public data openness facilitates enterprise joint innovation, with a particular focus on the variation in policy effectiveness across different administrative levels.
To this end, this paper employs a staggered difference-in-differences(DID) model to examine three core relationships: the impact of provincial and municipal public data openness on enterprise joint innovation, the influencing factors behind policy effect differences between the two levels and the mediating role of enterprise digital technology application. Using panel data of A-share listed companies in China from 2010 to 2023, the study ultimately obtains 26 564 firm-year observations, covering 2 185 listed firms after data preprocessing. The conclusions are validated through a series of robustness tests, including parallel trend tests, endogeneity checks, and placebo tests. Additionally, the study further analyzes heterogeneity in municipal-level platforms in terms of openness, R&D investment, and government digital attention, and conducts subgroup analyses based on region, industry, and firm characteristics.
The empirical findings show that provincial-level public data openness significantly promotes enterprise joint innovation, whereas municipal-level openness, constrained by limited platform capabilities, does not exhibit significant enabling effects overall. However, increased investment in scientific research and heightened government digital attention can effectively mitigate this limitation. Further analysis reveals that provincial-level public data openness facilitates joint innovation mainly by enhancing the breadth and depth of enterprise digital technology application. Heterogeneity analysis shows that the effect is more pronounced in innovative cities, digital economy industries, technology-intensive enterprises, and growth-stage firms.
On the basis of these findings, this paper proposes(1) strengthening institutional design and scenario-based orientation of provincial platforms;(2) increasing resource support for municipal platform construction, operation, and maintenance;(3) enhancing fiscal support for science and technology;(4) improving municipal institutional frameworks and data governance capacity;(5) implementing differentiated data support mechanisms by firm type—providing technology standards and research outputs to technology-intensive firms while offering market trends and strategic financing data to growth-stage firms—to enhance joint R&D efficiency and multi-stakeholder resource synergy; and (6) promoting the orderly development of enterprise digital technology application capabilities to strengthen data-to-innovation conversion.
The marginal contributions of this paper are as follows:Firstly, it expands the theoretical perspective on enterprise joint innovation. By positioning public data as a key production factor, this study analyzes its value in enterprise joint innovation, enriching the resource-based theoretical framework of joint innovation.Secondly, it reveals policy effect differences between two levels of public data openness. While prior studies mainly evaluate the overall effects of public data openness from a macro perspective, this study emphasizes the differentiated impact of provincial and municipal platforms, thereby deepening the theoretical foundation for refined, regionally tailored policy implementation.Thirdly, it clarifies the mechanism of public data openness and highlights the mediating role of digital technology application. From the perspective of internal firm capabilities, this study examines how the breadth and depth of digital technology application mediate the relationship between public data openness and joint innovation, thus advancing the understanding of how data elements generate value.

Computing infrastructure serves as the foundation supporting the development of the digital economy and artificial intelligence. The Chinese government has elevated the construction of computing infrastructure to a national strategic priority, implementing a series of policies to promote its high-quality development. Key initiatives include the "Eastern Data, Western Computing" project and the release of the Action Plan for High-Quality Development of Computing Infrastructure, aimed at fully unleashing the driving effect of computing power on the digital economy. However, influenced by disparities in energy endowments, industrial foundations, and digitalization levels, provincial computing power infrastructure development exhibits significant regional imbalances. Therefore, in order to foster high-quality computing infrastructure and empower advances in AI and the digital economy, it is critical to scientifically measure the level of provincial computing infrastructure construction, and analyze its evolutionary patterns and regional differences.
While the existing literature has laid a foundation, there remains room for expansion. Firstly, most current studies focus on measuring general digital infrastructure, with their indicator systems heavily emphasizing information and communication technology (ICT) access indicators, such as internet penetration rates and fiber-optic cable density. In contrast, the assessment of computing power infrastructure construction levels often relies on single or limited indicators. A comprehensive evaluation system that systematically reflects its functional positioning and technical characteristics has yet to be established. Secondly, existing research lacks empirical analysis regarding the dynamic evolution, sources of regional disparities, and spatial correlation of computing power infrastructure construction levels. In light of this, this paper constructs a measurement indicator system for computing power infrastructure construction levels, comprehensively evaluates the level of provincial computing power infrastructure development in China, and reveals its regional disparities and spatial linkages.
Following the logic of "core elements - external support - high-quality development goals",this study takes 30 provincial-level regions in China from 2016 to 2022 as research samples, and constructs an evaluation index system for China's computing infrastructure construction level from four dimensions: scale of computing facilities, network carrying capacity, computing support environment, and green development of computing power. The entropy method is employed to measure China's computing infrastructure construction level. Then, the Dagum Gini coefficient is employed to reveal regional differences in China's computing infrastructure construction level and their sources, while Moran's I index is applied to explore its spatial correlation.
The findings are as follows: First, China's computing infrastructure construction level has steadily improved, showing a gradient pattern, where Eastern China leads, followed by Central and then Western China, with provinces and cities such as Guangdong and Beijing taking a significant lead. Second, the overall regional differences in computing infrastructure construction level have shown a fluctuating upward trend, with inter-regional differences being the primary source. An analysis of intra-regional variation reveals that Eastern China exhibits the greatest internal disparity, ahead of Western and then Central China. Regarding inter-regional disparity, the East-West divide is the most significant, exceeding both the East-Central and Central-West gaps. Third, the construction level of computing infrastructure across provinces exhibits positive spatial correlation, with the intensity of spatial association characterized by an initial increase followed by a decrease.
Compared with existing studies, this paper makes several key contributions. First, it constructs a more comprehensive evaluation index system for computing infrastructure construction level from four dimensions: scale of computing facilities, network carrying capacity, computing support environment, and green development of computing power. Second, through the Dagum Gini coefficient, it analyzes regional differences and their sources, providing decision support for formulating differentiated regional development policies and promoting coordinated regional development of computing infrastructure. Third, it applies Moran's I index to analyze the spatial correlation of computing infrastructure construction level, revealing its spatial agglomeration phenomenon. This study deepens the understanding of China's computing infrastructure construction level and provides a theoretical framework for optimizing the national layout of computing infrastructure and advancing its high-quality development.

The integration capability of complex product systems constitutes the core competence of system integrators, playing a pivotal role in breakthroughs in critical core technologies and the development of complex products. Unlike the technology catch-up patterns observed in mass-produced goods, complex product systems exhibit extended technology lifecycles and greater learning difficulties. Although significant research has examined paradigm shifts in systems thinking, technological adaptation, innovation ecosystems, resource orchestration, and organizational models within this context, less attention has been paid to the formation process and mechanisms of integrators′ system integration capabilities—an area that warrants further theoretical investigation.
This study adopts complex product system theory to conduct longitudinal exploratory case study on Cell-Bridge Biotech, examining the enterprise's technological substitution pathway in high-end equipment manufacturing for CGT upstream devices from 2011 to 2023, thereby extracting the enhancement mechanisms of complex product system integration capabilities during domestic substitution. Specifically, sustained technological breakthrough provides foundational support for domestic substitution. Through upgrading three dimensions of integration capabilities—technological integration, product integration, and system integration—the case enterprise achieved progressive technological substitution, critical module substitution, and ultimately full-process substitution, demonstrating an evolutionary trajectory characterized by gradualism. Technological integration capability refers to the ability to translate underlying technologies into tangible products by organically combining distinct technical components, systems, or software to achieve specific technological objectives. This capability emphasizes the process of technology development and transformation, typically resulting in the conversion of technological principles into laboratory products (product prototypes). At this stage, products are mostly in the internal testing phase, meeting basic requirements for consistency and stability. Product integration capability refers to the ability to rapidly and efficiently develop new products and bring them to market. This capability emphasizes swift responsiveness to market demands, iterative product refinement, and accelerated delivery to meet differentiated customer needs and maintain competitive advantage in fiercely contested markets. This process involves creating rapid prototypes, enabling user participation in evaluation to identify issues early and make adjustments, followed by iterative product development based on feedback. System integration capability refers to the ability to consolidate different information technology systems, hardware, and software components into a coordinated system or network, delivering comprehensive solutions to customers. This typically involves applying integrated technologies to unify disparate sub-product systems, applications, and data repositories, supporting business processes and requirements.Furthermore, the research reveals that the construction of complex product system integration capabilities emerges from the interplay between technological logic and market logic. The technological logic encompasses decomposition and linkage mechanisms, describing the interconnections among components, modules, or systems, while the market logic includes mechanisms of current market demand alignment and future demand traction, reflecting the market responsiveness at different stages of complex product development.
This study constructs a process model for enhancing the integration capability of complex product systems in domestic substitution. Theoretically, it constructs a dynamic process model for enhancing complex product system integration capabilities during domestic substitution, unveiling the mechanisms and pathways of critical core technology substitution from the complex product system perspective, thereby extending existing theories on both technology domestic substitution and complex product systems. Additionally, it proposes dual interaction logics—technological and market logics—for capability enhancement during substitution, enriching theoretical frameworks concerning complex product system development. Practically, the findings provide guidance for latecomer enterprises in critical technology domains to achieve technological leapfrogging. Continuous improvement of autonomous innovation capabilities through long-term fundamental technology breakthroughs is essential to reduce external dependence on core technologies, thereby establishing the technological foundation for enhancing system integration capabilities and realizing domestic substitution. Simultaneously, complex product system development must consider not only technological dimensions but also ensure alignment between technological and market dimensions.

Facilitating the transfer of tacit knowledge from users to artificial intelligence (AI) (referred to as "human-machine tacit knowledge transfer") is crucial for enterprises to address persistent challenges in AI implementation,including data scarcity,algorithmic constraints,and limited computational resources.This process is thus essential for unlocking the strategic value of AI and gaining a competitive edge.Consequently,how to promote human-machine tacit knowledge transfer has become an important issue that enterprises urgently need to address.
Although research on human-machine tacit knowledge transfer has been increasing,it is still scattered and lacks an effective framework.In addition,existing research emphasizes technological rationality,focusing on the technical implementation of human-machine tacit knowledge transfer from the perspective of AI as the knowledge transfer object,while largely overlooking the role of users as active knowledge contributors.This not only leads to an imbalance in research perspectives,but also results in a dual rupture of human-machine tacit knowledge transfer in terms of subject (users) and object (AI),as well as between behavior and technology,thereby hindering practical guidance for organizations seeking to facilitate effective knowledge transfer.
Therefore,this study investigates how human-machine tacit knowledge transfer unfolds from the user's perspective.Drawing on the theory of AI socialization,the study specifically examines how AI capabilities influence human-machine tacit knowledge transfer.It develops a research model in which AI capability serves as the independent variable,human-machine tacit knowledge transfer as the dependent variable,new professional role identity as a mediator,and perceived value and perceived threat as moderators.On this basis,this study collected data from 321 enterprises through a questionnaire-based survey and conducted empirical analysis and testing of research hypotheses using methods such as structural equation modeling and hierarchical regression analysis.
The results show that AI capabilities positively affect human-machine tacit knowledge transfer; furthermore,AI capabilities positively influence users' new professional role identity,through which AI capabilities positively influence human-machine tacit knowledge transfer.This also indicates that AI capabilities can not only directly affect human-machine tacit knowledge transfer,but also indirectly affect it through the mediating role of users' new professional role identity.The results also show that perceived value positively moderates the influence of AI capabilities on users' new professional role identity; whereas perceived threats do not hinder the influence of AI capabilities on users' new professional role identity.This also indicates that perceived value can indirectly enhance the impact of AI capabilities on human-machine tacit knowledge transfer.
This study contributes to the theory of human-machine tacit knowledge transfer:first,it promotes the diversification of research perspectives on human-machine tacit knowledge transfer by introducing a user behavior perspective; second,it provides a behavioral mechanism for the theory of human-machine tacit knowledge transfer,promoting the connectivity of research content in this field; third,it constructs an analytical framework of "capabilities (AI capabilities) - cognition (users' new professional role identity)-behavior (human-machine tacit knowledge transfer)" to study the behavioral mechanism of human-machine tacit knowledge transfer,providing effective theoretical support and a logical framework for subsequent research.
The findings offer actionable insights for organizations deploying AI on how to obtain the expected value of AI.Specifically, companies should prioritize building users' AI capabilities to facilitate tacit knowledge transfer between humans and machines. Additionally, aligning users' AI competencies with their evolving professional identities can further enhance this knowledge exchange. Organizations should also actively communicate the value of AI to encourage users to fully leverage these capabilities.The results also suggest that enterprises should not be fixated on whether to eliminate users' perceived threat of AI.Instead,they can guide users to turn perceived threats into motivation,thereby encouraging them to use their AI capabilities to carry out new work and transfer tacit knowledge to AI.

Digitalization and green transition are increasingly critical trends in the global economic and social development. In this context, the concept of “twin transition” has emerged, emphasizing the integration of digital and green development. Industrial internet platforms, with their inherent strengths in data integration, resource optimization, and cross-industry collaboration, are closely aligned with the objectives of coordinated digital and green development, positioning them as critical enablers for enterprises undergoing twin transitions. The extent to which enterprises engage with platforms directly impacts the effectiveness of their twin transition process. However, current research only focuses on isolated effects of digital or green transition and relies on static indicators, leaving gaps in understanding the dynamic evolution of digital-green coordination and how platform-participatory manufacturing enterprises utilize industrial internet platforms to drive twin transition.
This study, rooted in digital ecosystem theory, employs a longitudinal multi-case study approach, selecting Weihai Baowei New Materials on the Uniorange platform, Guangshan Cement on the Tianxin platform, and Yuyao Lingke on the Geega platform as case examples. These platforms differ in terms of their types and technical characteristics, while the case companies vary in size and belong to different industry sectors, each exhibiting distinct advancements. As a result, these cases provide a comparative and dynamic framework for analyzing the internal processes through which industrial internet platforms empower participating manufacturing enterprises to achieve twin transitions. By integrating time-series analysis and case-based coding, this study outlines the stage-specific characteristics of industrial internet platforms in supporting platform-participatory manufacturing enterprises through their twin transition. It further identifies the action paths and process models associated with the digital-green coordination in each stage of platform use.
The findings suggest that the twin transition of platform-participatory manufacturing enterprises occurs in three distinct stages: (1) digital-green technology application, (2) digital-green collaborative advancement, and (3) digital-green symbiotic innovation. The level of digital-green synergy facilitated by industrial internet platforms plays a critical role in determining whether enterprises can progress to the next stage, creating a progressive relationship that evolves from initial, surface-level implementation to deeper, more integrated practices. Two key evolutionary paths are identified in the transition from Stage 1 to Stage 2: enhancing big data processing capabilities and strengthening supply chain cooperation. In the progression from Stage 2 to Stage 3, the paths include improving the platform's innovation and reconstruction capabilities, as well as enhancing its integration and innovation capabilities. At different stages, the digital transition manifests as a sequence of changes: from digital renovation to online networking, and ultimately to comprehensive intelligentization. Meanwhile, the green transition advances through a process of high-efficiency resource utilization, green process optimization, and organizational greening. Internally, the technological features of industrial internet platforms provide a foundation for enabling these transitions. Enterprises exhibit differential characteristics in their platform usage at various stages, leading to the formation of unique chaining capabilities that contribute to the development of a collaborative ecosystem.
This paper introduces a three-stage process model for the role of industrial internet platforms in empowering manufacturing enterprises for twin transition. It offers a new theoretical framework for understanding the dynamic nature of platform empowerment, shifting research from a one-size-fits-all approach to a more nuanced, stage-specific qualitative and quantitative evaluation. Additionally, it bridges the gap between technical structure perspectives and process-oriented views in digital platform ecosystem studies, highlighting the influence of platform application stages on enterprises’ chaining capabilities and the role of these capabilities in fostering ecological synergies. In addition, it proposes practical pathways for guiding the phased evolution of twin transitions, enriching the application of value co-creation theory in this field. Finally, the study suggests that platform-participatory enterprises should build their capabilities in phases, tailored to the specific stage of digital and green twin transition they are in; and platform companies should provide differentiated empowerment services based on the needs of enterprises at different stages. While the government should establish dedicated funds to support digital and green twin transitions, promote the implementation of industrial internet platforms across industries and regions, support the development of industrial data standards, and advocate for the open and unified protocols of industrial equipment and industrial software interfaces.

In the current era, the world is experiencing an unprecedented transformation in a century, and the instability of the international situation and the continuous escalation of international trade barriers continue to aggravate the operational risks of the global supply chain, and enterprises around the world are facing the potential risks of blocking, jamming, and breaking the chain due to the interruption of the supply of core raw materials, the blockade of key technologies, and the restriction on the import and export of products. In this context, how to effectively enhance the resilience and stability of the industrial chain supply chain has become a major theoretical and practical problem to be solved. Existing research shows that the diversified layout of supply chain can effectively deal with the risk of enterprise supply chain disruption. With the increasing level of digital industry agglomeration, its impact on supply chain configuration diversification will be more and more prominent.
This study takes the data of A-share listed companies in Shanghai and Shenzhen from 2011 to 2023 as research samples, and to examine how the agglomeration of the digital industry affects the supply chain configuration of enterprises, this study first constructs a panel two-way fixed effects model. Subsequently, based on the results of the baseline regression, a mediating effect model is further introduced to reveal the underlying mechanism of this impact. In order to effectively mitigate endogeneity arising from omitted variables and bidirectional causality, this study further employs the two-stage least squares (2SLS) method to re-evaluate the impact of digital industry agglomeration on the diversification of corporate supply chain configuration.
It is found that digital industrial agglomeration can significantly promote the diversification of corporate supply chain configuration, and this promotion is mainly achieved by enhancing the technological innovation ability of enterprises and improving the quality of corporate information disclosure. Further analysis reveals that there are significant differences in the impact of digital industrial agglomeration on firms' supply chain configuration, and that such a facilitating effect is more pronounced among non-state-owned firms, firms with relatively low levels of AI application, firms in regions with relatively high degrees of marketization, and firms in coastal regions. Therefore, this study proposes three policy recommendations from spatial layout of the digital industry, agglomeration-efficiency release, and differentiated governance to enable the digital sector to optimize supply chains and broaden the sharing of agglomeration dividends.
Compared to existing research, this paper's contributions lie in three aspects. First, it expands the research scope of digital industrial agglomeration's economic effects. Existing literature mainly explores its economic spillovers from perspectives like regional green tech innovation, manufacturing export tech complexity, and corporate resilience, but lacks systematic research on its impact on micro-level enterprises. This paper uses a firm supply chain configuration perspective, builds a theoretical framework for digital industrial agglomeration's impact on firm supply chain arrangements, and examines these effects systematically. Second, it enriches research on the determinants of firm supply chain configuration. Existing studies focus on micro-firm (e.g., client CEO changes, AI innovation) and macro-policy (e.g., relaxed market access) perspectives, but no research has examined digital industrial agglomeration's impact,despite its growing prominence. This paper adopts a digital industrial agglomeration perspective to study its impact on firm supply chain configuration, offering a new viewpoint for this field. Third, it reveals the operational mechanisms and differentiated effects of digital industrial agglomeration on firm supply chain configuration, providing new insights for subsequent studies. It examines the mechanisms through the dual lens of technological innovation and information transmission, and explores differential impacts across internal corporate characteristics and external environmental factors. This not only provides theoretical support for understanding supply chain configuration evolution in the digital economy, but also offers decision-making references for improving supply chain efficiency and resilience via optimized digital industry spatial arrangements during the 15th Five-Year Plan period.

Leveraging artificial intelligence (AI) technologies to advance digital and intelligent transformation has become an important pathway for Chinese manufacturing enterprises to achieve high-quality development. At the same time, the rapid expansion of AI applications has sparked widespread societal concern regarding whether AI adoption inevitably substitutes for human labor. Accordingly, two interrelated questions have attracted growing attention from both academia and practice: which factors or combinations of factors shape AI adoption intensity in manufacturing enterprises, and how does AI adoption intensity affect the scale of labor demand? Existing research, however, tends to examine either the determinants of AI adoption behavior at the individual level or the effects of AI adoption intensity on the scale of enterprise labor demand. Systematic evidence on the enterprise-level antecedents of AI adoption intensity and their implications for labor demand remains limited. In particular, prior studies have rarely investigated the interactive effects of technological, organizational, and environmental (TOE) factors on AI adoption intensity from a configurational perspective, nor have they examined the mediating role of adoption intensity in linking TOE-based configurations to the scale of labor demand.
Against this backdrop, this study draws on the TOE framework to examine the configurational pathways through which TOE factor combinations drive AI adoption intensity in manufacturing enterprises and to assess how such adoption intensity subsequently influences the scale of labor demand. Methodologically, the study adopts a mixed-methods approach that combines fuzzy-set qualitative comparative analysis (fsQCA) and regression analysis, using data from 148 A-share listed manufacturing enterprises in 2023. Specifically, fsQCA is employed to identify multiple TOE configuration patterns associated with high AI adoption intensity, while regression analysis is used to test the mechanisms through which TOE configurations and AI adoption intensity jointly influence the scale of labor demand.
The fsQCA results reveal three distinct pathways to high AI adoption intensity. The first is an “Organizational Dual-Core Driven under Limited Technological and Environmental Constraints” pathway, in which enterprises compensate for limited technological and environmental conditions (e.g., low R&D investment, weak policy support, and limited innovation capacity) by strengthening internal organizational transformation, knowledge intensity, and risk-taking capacity. The second is a “Technology-Organization Collaborative Innovation with Policy Nudges under Limited R&D Investment” pathway, whereby enterprises with constrained R&D resources enhance AI adoption by building internal-external collaboration mechanisms and leveraging government policy support to bolster technological innovation, organizational knowledge intensity, and risk-taking capacity. The third is a pathway dominated by R&D investment and risk-taking, indicating that even when organizational capabilities and external support are weak, enterprises can still achieve high AI adoption intensity by substantially increasing R&D investment and organizational risk-taking.
Regression analyses further demonstrate the mediating role of AI adoption intensity between TOE configuration patterns and the scale of labor demand. The results show that high AI adoption intensity generally has a negative impact on the scale of labor demand. Specifically, the first two configuration patterns—while promoting AI adoption intensity—are associated with a reduction in the scale of labor demand. By contrast, the “R&D investment and risk-taking dominant” configuration increases AI adoption intensity without exerting a significant negative effect on labor demand.
This study offers both theoretical and practical contributions. Theoretically, it (1) advances understanding of AI adoption intensity by examining the synergistic effects of TOE dimensions from a configurational perspective, thereby addressing the limitations of linear, single-factor approaches; (2) extends research on the mechanisms through which AI adoption intensity influences the scale of labor demand in manufacturing enterprises; and (3) advances beyond prior paradigms that focused solely on antecedents of AI adoption intensity or solely on its scale of labor demand consequences, by integrating antecedent configurations and outcome effects into a unified causal system, thereby forming a logical loop more consistent with real-world scenarios. Practically, the findings offer guidance for manufacturing enterprises in selecting appropriate AI adoption pathways and optimizing labor management strategies, while also providing policy insights for governments seeking to balance AI-driven transformation with employment stability.

In an era marked by escalating geopolitical tensions, recurrent global disruptions, and rapid technological transformation, strengthening industrial chain resilience is a strategic imperative. It is crucial not only for advancing high-quality economic development and building a modern industrial system, but also for adapting to the reconfiguration of global supply chains and safeguarding national security. China has established a large-scale, competitive, and relatively independent modern industrial system, with industrial chains generally stable and resilience gradually improving. However, amid the dual pressures of domestic transformation and external uncertainties, structural vulnerabilities remain, particularly in foundational capabilities and critical technological bottlenecks in sectors including semiconductors, advanced materials, and core software. These weaknesses increasingly expose the fragility of key industrial chains, making resilience enhancement an urgent strategic priority.
The academic community has explored pathways to enhance industrial chain resilience from various perspectives, primarily focusing on industrial agglomeration, the digital economy and finance, digital-real economy integration, innovation-driven policies, and the chain leader system. However, research on the impact of artificial intelligence(AI) on industrial chain resilience remains relatively under-explored and requires further deepening. This gap mainly manifests in two aspects: First, the mechanism pathways through which AI drives resilience need further refinement. While existing studies identify the mediating role of technological innovation, they predominantly emphasize supply-side transmission mechanisms and largely neglect the mediating role of demand-side factors. Second, AI's application and development often facilitate cross-regional flows of production factors, potentially generating spatial spillover effects that transcend geographical boundaries, yet such effects are frequently overlooked in current research.
On the basis of this theoretical foundation, this study employs panel data from 30 Chinese provincial-level administrative regions(excluding Hong Kong, Macao, Taiwan, and Xizang) from 2011 to 2023. It takes industrial chain resilience as the dependent variable which is measured through a five-dimensional indicator system(foundation, resistance, recovery, sustainability, leadership) using the entropy method; and the study selects artificial intelligence as the core independent variable that assessed via input-output indicators(infrastructure, capital, talent, industrial application, economic output) . Technological innovation and residential consumption serve as mediating variables, while five indicators including the degree of openness,government participation, population density, environmental regulation intensity, and employment density are employed as control variables. It constructs mediating effect models and a spatial Durbin model to examine how artificial intelligence affects local industrial chain resilience through dual supply-demand mechanisms while investigating its spatial spillover effects on neighboring regions. Mediation analysis employs both two-step and three-step methods with bootstrap tests for robustness, while spatial analysis uses bidirectional fixed effects and multiple weight matrices, revealing that indirect effects are smaller than direct effects.
The main findings are as follows: First, AI enhances industrial chain resilience through its data-driven nature, systemic intelligence, and pervasive penetration, strengthening foundational stability, resistance, recovery capacity, sustainability, and leadership across dimensions. Second, as resilience is influenced by both supply and demand where technological progress and consumption play crucial roles, AI improves resilience by promoting technological innovation and stimulating household consumption. Third, AI demonstrates significant spatial spillover effects in enhancing industrial chain resilience, with positive spillovers dominating and direct local effects exceeding indirect cross-regional impacts.
This paper makes two primary contributions. Theoretically, it advances the literature by integrating supply-side(technological innovation) and demand-side(household consumption) mechanisms into a unified analytical framework, addressing the prevailing bias toward supply-side explanations and offering a more comprehensive understanding of AI's role in shaping industrial resilience. Methodologically, it enriches the spatial dimension of industrial resilience research by examining cross-regional spillovers, revealing geographical interdependence and broadening theoretical implications of AI in regional economic dynamics. In summary, these findings contribute to a more holistic, spatially informed understanding of how artificial intelligence influences industrial chain resilience in a complex, interconnected economy.

Innovation is the key to sustained organizational growth, and artificial intelligence (AI) is an important innovation enabler. With the release of Big Language Models by DeepSeek, a deep-search company, a new round of discussions in the global AI community has been triggered. More and more companies are privatizing their deployments through AI and restructuring their organizations to create new business models and enhance sustainability. In this context, AI has significant potential for business model innovation due to its deep learning and human-computer collaboration characteristics. However, despite the consensus on the positive impacts of AI technological advances, companies face confusion in practical applications, especially the lack of expertise and capabilities in transforming business models and successfully implement AI, and thus the impact mechanisms of AI-adoption intensity on business model innovation remain unclear.
Thus, the study constructs a theoretical framework of “Resource Acquisition-Capability Enhancement-Value Reconstruction” based on the resource-based view and resource dependence theory. The study empirically examines 337 sets of data from Chinese manufacturing firms using hierarchical regression analysis. It then systematically explores the transmission mechanisms through which AI-adoption intensity affects business model innovation through explorative AI capabilities and exploitative AI capabilities, as well as the moderating roles of environmental hostility and environmental dynamism.
This study mainly draws the following conclusions: (1) AI-adoption intensity has a significant positive impact on business model innovation, consistent with current research on technology-driven business model innovation. It suggests that organizational AI-adoption intensity provides foundational technical resource support for organizations in the context of digital intelligence transformation, and becomes a key path to drive business model innovation. (2) AI-adoption intensity has a positive effect on business model innovation via explorative AI capabilities and exploitative AI capabilities. Specifically, explorative AI capabilities focus on the opportunity creation dimension, while exploitative AI capabilities are dedicated to efficiency enhancement, and the two synergistically promote business model innovation. This finding not only echoes the core viewpoint of organizational duality theory but also reveals the micro-mechanism of the transformation of technological resources to innovation capabilities in the AI context. (3) Environmental hostility and environmental dynamism play a moderating role in the relationship between explorative AI capabilities and exploitative AI capabilities and business model innovation, and at the same time have a significant enhancing effect on the mediating role of dual AI capabilities. This finding suggests that external environmental pressures motivate firms to unleash dual AI capabilities, whereas a stable environment may trigger AI capability rigidity.
The research makes theoretical contributions to advance the intersection of AI and innovation. First, it expands the explanatory boundaries of AI-adoption intensity in the context of intelligence, pointing out that the value realization path of AI technology is different from that of traditional technology due to its cognitive complexity and strong generative characteristics. Therefore, the study explores the impact of organizational AI-adoption intensity on business model innovation based on the resource-based view, which not only echoes the core proposition of the resource-based view that “heterogeneity of technological resources creates competitive advantage”, but is also a useful supplement to the study of technology adoption in the context of intelligent transformation. Second, previous research on AI technology in organizations mainly focuses on the advantages it brings at the technical level, however, the use of technology alone is not enough to bring innovation benefits directly to organizations . The study confirms the importance of explorative AI capabilities and exploitative AI capabilities for business model innovation, emphasizes that technology must be transformed into irreplaceable and hard-to-imitate capabilities to create value, and contributes to the theoretical development of organizational capability research. Finally, by parsing the moderating effects of environmental hostility and environmental dynamism, the study systematically reveals the boundary conditions of explorative AI capabilities and exploitative AI capabilities to drive business model innovation, enriches the discussion of environmental factors in organizational innovation, and enhances the contextualized study of business model innovation.

Universities constitute a fundamental pillar of China's innovation ecosystem, holding approximately one-quarter of the national invention patents. However, their patent commercialization rate remains below 10%, creating a critical disconnect between knowledge generation and productive application. To bridge this "last mile" gap in technology transfer, China introduced the patent open licensing system in its 2020 Patent Law revision—a standardized mechanism enabling patent holders to publicly declare licensing terms and fees. By December 2023, universities had contributed 86% of the 16 065 patents in the national open licensing pool, emerging as the primary drivers of this institutional innovation. Despite significant participation, the micro-level decision mechanisms underlying universities' choices between open licensing and exclusive protection remain poorly understood. Thus,this study investigates how multidimensional patent quality characteristics influence university patent open licensing decisions, addressing critical gaps in understanding technology transfer strategies. Unlike existing literature focusing on macro-institutional analysis or treating patent quality as unidimensional, this study develops a comprehensive framework examining trade-offs between "monopoly premiums" from exclusive licensing and broader benefits of open licensing. It explores how patents' technical quality (innovation sophistication and knowledge complexity), legal quality (protection scope and enforcement strength), and economic quality (market validation and commercialization maturity) differentially affect licensing decisions.
The theoretical foundation integrates cost-benefit analysis with multidimensional patent quality assessment. Universities face complex optimization problems: exclusive licensing promises monopoly rents but entails substantial transaction costs including partner search, negotiation, and monitoring; conversely, open licensing reduces transaction costs through standardization while potentially sacrificing monopolistic returns. This trade-off becomes nuanced when considering that patent quality manifests across multiple dimensions, each potentially exerting differential effects on licensing decisions.
The empirical analysis employs large-scale data comprising 473 621 invention patents from 312 Chinese universities filed between 2004 and 2024, with 13 809 patents declared for open licensing during June 2021 to December 2023. Patent quality indicators were constructed using entropy weighting methods based on eleven sub-indicators: technical quality (IPC classifications, inventor teams, citations), legal quality (claim counts, patent families, international filings), and economic quality (licensing history, transfer records, pledge activities). Binary logistic regression models with multilevel moderation effects test hypothesized relationships, supplemented by robustness checks using propensity score matching and complementary log-log models addressing rare event bias.
Findings reveal heterogeneity in how different quality dimensions influence licensing decisions. Patents with higher technical and legal quality significantly reduce open licensing probability, as these patents embody greater potential for monopoly premiums through exclusive arrangements. Their technological sophistication and robust legal protection create substantial competitive barriers. Conversely, patents with higher economic quality demonstrate positive association with open licensing, suggesting that once a technology has been market-validated universities are willing to trade monopoly rent for the transaction-cost savings and network effects of standardized licensing so as to maximize social impact.
The study uncovers important contextual dependencies. Regional technology market maturity weakens the positive relationship between economic quality and open licensing, as mature markets provide alternative commercialization channels. Market competitiveness strengthens this relationship, indicating competitive pressure drives universities toward open licensing to establish user bases rapidly. Organizational characteristics reveal another complexity layer: universities with greater public resource support show weakened negative effects of technical and legal quality on open licensing, suggesting well-resourced institutions prioritize social impact over monopoly revenues. Institutional pressure for technology transfer similarly moderates these relationships.
This research makes several theoretical contributions. First, it introduces the monopoly premium concept to systematically explain core trade-offs in patent licensing decisions. Second, it resolves theoretical ambiguity by demonstrating dimension-specific effects of patent quality, challenging unidimensional approaches. Third, it identifies multilevel contextual mechanisms: market environments primarily influence economically valuable patents through efficiency considerations, while organizational characteristics affect technically and legally sophisticated patents through social responsibility channels. The findings also offer practical implications for patent management and policy design.

The report of the 20th National Congress of the Communist Party of China underscores the imperative to promote the development of “specialized, sophisticated, differentiated, and innovative”(SSDI) small and medium-sized enterprises (SMEs). As a key part of the innovation-driven development strategy, SSDI SMEs have gained new attention for their innovation capabilities. Yet, innovation requires heavy upfront investment and brings uncertain results. For SSDI enterprises—typically constrained by limited scale—technological innovation faces compounded challenges. They often face serious R&D funding shortages, structural talent deficits, and a common leaning toward imitative innovation. Consequently, many SMEs are compelled to adopt low-value-added innovation models that prioritize quantity over quality, creating a critical bottleneck in their innovative advancement.Thus, addressing financing constraints, optimizing talent allocation, and refining innovation mechanisms are urgent tasks for promoting the innovative growth of SSDI SMEs.
Against the backdrop of digital-real economy integration, the disclosure of data assets can effectively address the aforementioned challenges. As a form of voluntary corporate disclosure, data asset disclosure serves as a critical channel for the public to understand value creation from data assets, big data applications, digital transformation, and related corporate activities. It also holds significant implications for firms' own innovation development. Given the exponential growth of data as an “asset class” and the pressing need to enhance the innovation capabilities of SSDI SMEs, academia urgently needs to investigate the innovation effects of data asset disclosure. Such research would provide valuable insights for China to seize new opportunities in digital development and expand new frontiers for economic growth.
Existing literature has empirically demonstrated the positive effects of data asset disclosure, including lowering external financing costs, boosting market reactions, enhancing analyst forecasts, and increasing firm value.Yet, its impact on SME innovation quality is still unclear. This research gap takes on heightened significance when examining SSDI SMEs. Given their advanced digital maturity, richer data asset portfolios, and more comprehensive disclosure practices compared to conventional SMEs, SSDI enterprises present an ideal context to investigate whether data asset disclosure exerts a disproportionately stronger effect on innovation quality enhancement. Therefore exploring this is crucial for innovation policies and corporate digital strategies.
Adopting a micro-level perspective, this study examines the impact of data asset disclosure on innovation quality in SSDI SMEs, using a sample of China's “Little Giant” listed enterprises from 2011 to 2023. The empirical results demonstrate that data asset disclosure significantly enhances innovation quality among SSDI SMEs, with findings remaining robust across a series of sensitivity tests. Mechanism analysis reveals that data asset disclosure primarily improves innovation quality through three channels: alleviating financing constraints, optimizing human capital allocation, and reducing innovation conformity behavior. Furthermore, the study identifies significant moderating effects, showing that stronger intellectual property protection, intensified industry competition, and increased analyst coverage can amplify the positive influence of data asset disclosure on SMEs’ innovation quality.
Compared with existing literature, this study makes three key marginal contributions: First, it enriches micro-level research on data assets. Prior studies focused on macro-level aspects, neglecting micro-level analyses of how data asset disclosure enhances innovation quality. This work bridges the gap between information and innovation economics. Second, this study elucidates the mechanisms through which data asset disclosure improves innovation quality. Previous research has largely treated the relationship between data disclosure and SME innovation as a “black box”. Departing from conventional approaches, this study adopts an information transparency perspective to demonstrate how SSDI SMEs leverage data asset disclosure to enhance innovation quality along three dimensions: (1) improving capital accessibility, (2) optimizing labor structure, and (3) strengthening governance capacity. Third, existing studies predominantly rely on annual reports as the primary source of data asset information. Recognizing that corporate news coverage serves as an equally critical channel for information dissemination,and is indispensable for a comprehensive evaluation of data asset disclosure,this study systematically analyzes data-related disclosures in both annual reports and news articles. Methodologically, it employs Python-based text mining techniques and constructs a specialized data asset lexicon using Word2Vec neural network modeling. This dual-source approach not only enhances measurement accuracy but also pioneers a novel methodological pathway for future data asset research.

The Chinese economy is now subject to recurrent disruptions arising from geopolitical tensions, environmental shocks, resource constraints, and rapid iterations of digital technology. This highlights an urgent need to enhance economic resilience, so as to ensure the long-term stability and development of the economy. In this context, digitalization and greening act as dual engines for boosting economic resilience, where digital technology empowers green transformation and green development guides digital innovation, collaboratively driving the economic system toward structural upgrading and high-quality development.
Nevertheless, existing literature predominantly focuses on economic resilience significance, the impact of digitalization on green development, digital-green synergy mechanisms and spatiotemporal patterns, and independent contributions of each dimension to resilience enhancement. However, theoretical analysis and empirical exploration of how their synergy drives economic resilience remain relatively underdeveloped. The use of digital devices relies on energy consumption represented by electricity, which may bring environmental burdens such as electronic waste. While digital technologies accelerate green development, they may induce resource crowding and energy rebound effects. Greening improves environmental quality, yet without digital support, its carbon-reduction efficiency and industrial transformation potential remain constrained. Furthermore, regional heterogeneity in resource endowments, technological foundations, and geographical advantages produces divergent digital-green synergy patterns. The coexistence of digital divides and green development gaps creates structural imbalances that weaken systemic shock absorption and adaptation capacity, ultimately hampering overall economic resilience improvement. Such complexity necessitates comprehensive investigation into how digitalization and greening interact synergistically to enhance economic systems' adaptive capacity and long-term sustainability.
Against this backdrop, this study constructs an analytical framework for understanding how digital-green synergy drives improvements in economic resilience, drawing upon Complex Adaptive Systems (CAS) theory to capture nonlinear interactions, feedback loops, and emergent dynamics. Methodologically, it employs dynamic qualitative comparative analysis (QCA) to identify the configurations associated with both high and low economic resilience, thus revealing the empirical manifestations of regional heterogeneity.It further applies a random forest regression model to rank the importance of conditioning variables and generate partial dependence plots, allowing for a more granular examination of interaction effects and threshold characteristics. This mixed-method approach supplements the limitations of traditional QCA and enriches the literature on economic resilience by integrating configurational and machine-learning perspectives.
The empirical results yield several key findings. First, neither digitalization nor greening alone constitutes a necessary condition for enhancing economic resilience, although both exhibit pronounced regional effects driven by differences in resource endowments, industrial structures, and technological bases. Second, five high-resilience configurations are identified and can be distilled into three archetypal models:a “digitalization and resource–environment optimization synergy” model,a “digitalization-led, low-carbon breakthrough” model, and a “green-led, digital-facilitated” model. These diverse pathways share a “multiple routes to the same outcome” characteristic, underscoring the equifinality of resilience formation. Third, in the temporal dimension, the consistency of high-resilience configurations has generally increased over time, reflecting a strengthening synergy effect. Spatially, while the five configurations display broad applicability, notable regional differences remain: the eastern and central regions show stronger alignment, whereas the western and northeastern regions face challenges due to weaker digital foundations and delayed green transitions. Fourth, both the QCA and random forest analyses indicate that improvements in economic resilience are not determined by digitalization or greening in isolation but depend on their complementary and mutually reinforcing interaction. The two dimensions operate through distinct yet synergistic mechanisms, producing a stronger integrated effect.
In light of these findings, the study proposes that regions should adopt differentiated strategies in light of their resource endowments, technological capabilities, and policy environments. They should select the most suitable synergy pathway to meet region-specific needs for resilience enhancement. Moreover, integrated governance should be strengthened: efforts should be made to optimize digital-green coordination models, establish robust monitoring and evaluation systems, and build institutional safeguards to ensure effective implementation. Such measures will comprehensively enhance the economy’s capacity to absorb shocks, adapt to disruptions, and undergo structural transformation—thereby sustaining and deepening economic resilience over the long term.