The rapid development of Artificial Intelligence (AI) technology creates opportunities while simultaneously intensifying concerns regarding its safety risks and alignment problems. Existing research often discusses AI functionality (the technical capability for automated decision-making) and AI safety (the responsible attribute for risk prevention) in isolation, overlooking their inherent paradoxical relationship. This study aims to explore how to coordinate the tension between functionality and safety through strategic design to achieve sustainable competitive advantage. We propose an AI "Functionality-Safety" paradox framework and identify four core scenarios of interaction between functionality and safety. Meanwhile, by integrating perspectives from paradox theory and dynamic capability theory, we formulate the core viewpoints on the AI "Functionality-Safety" paradox. On this basis, the study puts forward dynamically matched strategic paths for different scenarios, explains the non-linear relationships and balance mechanisms among AI application timing, internal and external capabilities, and corporate performance under different strategies. The main contributions of this study lie in constructing an AI technology analysis framework that integrates the functionality and safety dimensions, forming dynamic synergy perspectives on the AI "Functionality-Safety" paradox, and establishing the dynamic matching between AI scenarios and collaborative strategies. This study provides theoretical guidance and practical implications for firms' paradox collaborative management and AI strategies.
Currently, artificial intelligence (AI) has become a core focus of strategic competition among major nations. China's AI sector is undergoing profound structural transformations. Growing evidence suggests that the AI industry may be at a critical transitional window-evolving from a fluid state characterized by technological diversity toward a transitional phase where the emergence of a dominant design is becoming apparent. However, existing literature lacks a clear definition and examination of this phase. Such ambiguity constitutes a primary “puzzle” hindering the effective advancement of the AI industry and the precise formulation of relevant policies.
The Abernathy-Utterback(A-U) model, a classical theoretical framework for analyzing industrial innovation dynamics, divides the evolution of innovation into three stages: the fluid phase, the transitional phase, and the specific phase. It thus provides a powerful theoretical tool for understanding industrial transition dynamics. Based on this model, this study systematically constructs a theoretical analytical framework by examining the multidimensional characteristics of the industrial innovation process, clarifying the internal mechanism behind the evolution of the innovation process and the formation of a dominant design, and revealing the dynamic adaptation principles that influence strategy design.
Using this theoretical framework, this study first systematically examines the development stage of China’s AI industry. It finds that the industry is transitioning from the fluid phase to the transitional phase-a conclusion highly consistent with the evolutionary patterns of industrial innovation described in the A-U model. From a technological perspective, the industry exhibits accelerated convergence of technical pathways, a systematic shift of innovation focus toward performance optimization, and the crossing of critical thresholds in technological maturity. In terms of industrial structure, embryonic forms of a dominant design are emerging, process innovation is becoming the core of value creation, and the division of labor within the industry is being profoundly reconfigured. Regarding the competitive landscape, a notable coexistence of open-source and closed-source models is observed, characterized by complementarity and symbiosis.
Secondly, focusing on historical cases where dominant designs were successfully established during similar transitional windows-such as the Wintel alliance, the Android system, and the European GSM standard-this study summarizes general winning principles that can inform the current development of AI in China. These include: technological feasibility as a prerequisite for the formation of a dominant design; the critical importance of dynamically balancing market acceptance and production economics for its widespread adoption; the shift in competitive paradigm from zero-sum game to ecological collaboration as a hallmark of the transitional phase; and the pivotal role of rule-setting power in transforming tri-dimensional coupling outcomes into enduring competitive advantages.
Finally, addressing core challenges such as high computing costs, lagging standardization, and immature ecosystem governance mechanisms in the development of China's AI industry, this study proposes an adaptive strategy framework encompassing technological breakthroughs, ecosystem development, and rule-making engagement. Specifically, in technological advancement, efforts should focus on overcoming computational bottlenecks and pursuing forward-looking development of key technologies. In ecosystem building, the emphasis lies on creating an open yet controllable industrial ecosystem. Regarding rule-making, it is essential to strengthen standard setting and enhance international governance cooperation. These recommendations aim to overcome critical bottlenecks during the transition phase, foster leapfrog development of China’s AI industry, and build an independent, controllable, and globally influential AI ecosystem.
The increasing rigidity of costs and technological leaps have led to the growing prominence of "rat race" competition in the manufacturing sector. The gradual improvement of technical standards has put forward new requirements for the development of the manufacturing industry and has raised the question of what impact they have on "rat race" competition, which is the focus of this study. Based on the data of Chinese A-share listed manufacturing enterprises from 2011 to 2023, this study empirically analyzed the impact of technical standards on "rat race" competition and its underlying mechanism. The empirical results show that technical standards can alleviate "rat race" competition, and the effect is more significant in alleviating product homogeneity-driven "rat race" competition. In terms of the mechanism of action, technical standards mainly alleviate "rat race" competition by promoting technological innovation and enhancing human capital. In terms of the moderating mechanism, the fair competition review system, by breaking down administrative monopolies and local protection, enhances the inhibitory effect of technical standards on "rat race" competition. Intellectual property protection, by reducing the property rights risks in the implementation of technical standards, encourages enterprises to transform common technologies into competitive advantages, thereby alleviating "rat race" competition. In terms of heterogeneity tests, the impact of technical standards on "intra-industry" competition varies asymmetrically due to industry nature and regional characteristics. On the one hand, the improvement of technical standards can only slightly alleviate the overcapacity-driven "rat race" competition of enterprises in the eastern region and non-high-tech enterprises. On the other hand, technical standards can alleviate the product homogeneity-driven "rat race" competition of high-tech enterprises and enterprises located in both the eastern and western regions. The marginal contributions of this study mainly involve three aspects. First, it focuses on the "rat race" behavior at the enterprise level and, based on the manifestations and formation logic of "rat race" competition, innovatively divides it into overcapacity-driven and product homogeneity-driven types, providing a clear analytical framework for understanding the composition and connotation of enterprise "rat race" competition. Second, from the perspective of "rat race" competition, it is the first to verify the micro-governance effect of the macro technical standard system, providing direct empirical evidence for the government to use regulatory policies to solve the problems of overcapacity and product homogeneity. Third, it incorporates technological innovation and human capital into the research framework of technical standards and "rat race" competition, and introduces the fair competition review system and intellectual property protection as moderating variables to systematically analyze the impact mechanism and boundary conditions of technical standards on "rat race" competition. This not only fills the gap in existing research but also provides a reference for enterprises to find effective paths to break the "intra-industry" competition dilemma. The research results provide new solutions for enterprises to alleviate "rat race" competition and for government and other relevant departments to optimize policy design. For enterprises, they should not only actively embrace the standard strategy, treating the adoption, participation, and even the leading formulation of technical standards as an important competitive strategy, but also simultaneously increase R&D and innovation, focus on core technologies, and attach importance to continuously improving the knowledge structure and skill levels of employees through recruitment, training, and incentive mechanisms, making human capital a key force in supporting the implementation and surpassing of standards. For the government, it is necessary to vigorously build an institutional environment that supports technological development, taking technological standards as the main handle of industrial and innovation policies. It should proactively plan and promote advanced technological standards in key national areas, guide the effective allocation of resources, accelerate industry integration and upgrading, and avoid disorderly internal competition. It is also essential to strengthen the rigid constraints of the fair competition review system to prevent technological standards from being improperly used to cause market monopolies or regional blockades, and ensure the openness, non-discrimination and fairness of standards. At the same time, it is necessary to improve and strictly enforce intellectual property protection, increase the cost of infringement, and effectively safeguard the legitimate rights and interests of innovators and essential patent holders.
The third distribution is an important means to promote common prosperity for all people and an effective pathway to realizing the essential requirements of socialism with Chinese characteristics. How to encourage Chinese residents to participate more in social donations and thereby advance the development of the third distribution is a question of both urgent practical significance and theoretical importance. The rise of short videos may offer a new, technology-driven approach to addressing this issue. Based on the novel features of short-video technology applications, this paper constructs a general theoretical analytical framework to demonstrate the promoting effect of short videos on household social donations, and it empirically tests the existence of this effect using panel survey data of Chinese households.
The study produces three main findings. First, baseline regression results show that short videos use significantly increases household participation in social donations and contributes to growth in donation amounts. Second, mechanism analysis indicates that short videos has a more pronounced promoting effect on donations by households with lower initial levels of trust, and also a stronger effect on households that perceive livelihood problems as more severe — that is, there exist two channels of influence: trust deepening and opinion reinforcement. Third, extension analyses find that short videos use more strongly stimulates donations by households that perceive higher severity in livelihood issues such as housing, social security, environmental protection, education, and healthcare. These results suggest that employing emerging digital technologies like short videos to promote the development of the third distribution has practical value.
Based on the findings, this paper proposes three policy implications. First, relevant government departments are advised to encourage nonprofit organizations to actively use short videos platforms for outreach and promotion, producing and disseminating emotionally engaging and mobilizing short videos to attract more households to follow and participate in social donations. Second, the government should strengthen review and guidance of charitable short-video content to ensure information is truthful and reliable, avoiding false publicity or inappropriate inducement, and thereby preserving a healthy charitable environment and donation atmosphere. Third, the government can further integrate livelihood-related platforms with philanthropic resources for precise matching: by leveraging short-video platform data and households’ perceived severity of different livelihood issues, it can intelligently match households to appropriate charitable projects and improve the efficiency of philanthropic resource allocation.
This study contributes three innovations. First, it enriches research on the role of digital technology in the development of the third distribution. While academic work on the economic effects of short videos is still emerging, this paper is among the earliest to address— from an economic perspective— the impact of short videos on household social donations and to conduct empirical analysis using nationally representative panel survey data. Second, it explicates the mechanisms through which short videos affects household charitable giving. The paper builds a general theoretical framework and empirically demonstrates that short videos promote household donations via two pathways: trust deepening and opinion reinforcement. Third, it uncovers the positive role of short videos in stimulating household charitable donations. The findings provide new evidence that emerging digital media technologies represented by short videos can further activate residents’ enthusiasm for donating, offering an actionable, technology-based idea to help construct a tertiary redistribution framework.
As an important part of scientific and technological innovation talents, academicians of aerospace are top professionals in the aerospace field, typical representatives of strategic scientists, and the core force driving breakthrough innovations in China's aerospace technology. Under the great changes unseen in a century, the aerospace field has become an important strategic high ground for countries to showcase their comprehensive national strength and scientific and technological capabilities. At present, talents in China's aerospace field share the achievements and experiences of aerospace development through international academic exchanges and cooperation, spread China's aerospace concepts, and actively learn the latest international knowledge. This promotes better alignment of China's aerospace science and technology with the international community, enhances the country's say in the global aerospace industry, and enables the world to see the strength of China's aerospace. It also creates favorable conditions for China to carry out more extensive and in-depth international cooperation in the aerospace field, continuously injects new vitality into China's aerospace science and technology, and achieves a leap from following, keeping pace with, to partially leading.
This study focuses on the integration and influencing factors of the core innovation capabilities of academicians in the aerospace field. Based on 38 biographies of academicians in the aerospace field, it uses grounded theory to construct a growth influencing factor model of "Motivation - Capability - Support - Behavior". Research findings show that the advancement of core capabilities of aerospace academicians is a dynamic and complex process, with different capabilities at different stages. During the growth stage, one possesses the ability driven by curiosity, the desire for knowledge and the ability to learn. During the stage of studying and choosing a major, one should possess the ability to seek knowledge and explore, the ability to plan one's career, and the ability to perceive interdisciplinary knowledge. At the initial stage of one's career, one should possess the ability to innovate and break through, the ability to collaborate in a team, and the ability to communicate and coordinate. During the glorious stage of one's career, one should possess strategic leadership capabilities, scientific research management capabilities, the ability to transfer technology across generations, and the ability to integrate across disciplines. In the vigorous stage of an old horse, one should possess the ability to pass on and cultivate, guide with experience, and promote the industry. This model reveals the spiral upward law of "motivation guidance - ability breakthrough - support and guarantee - behavioral practice" of academicians in the aerospace field.
Based on the above research content, in order to cultivate more professional talents in the aerospace field, this study puts forward the following suggestions: Cultivate a sense of patriotism and belief, and strengthen the sense of mission and responsibility of talents; Promote the development of innovative practices and enhance the innovation capabilities of talents; Strengthen professional education support and cultivate fertile ground for talent growth; Create an environment that values and honors talents, and build a highland for scientific research and development. In subsequent research, the sample types can be further expanded to conduct quantitative analysis, thereby exploring the growth mechanism of professional talents and providing more forward-looking theoretical support for the precise cultivation of talents in the aerospace field.
The global digital transformation, while reshaping the social power structure and resource distribution pattern, has also given rise to the practical paradox of technological equity. This paradox of technological equity reveals the complex mechanism where technological empowerment and technological alienation coexist, and equal opportunities and structural imbalance are intertwined. It is a significant obstacle to achieving the goal of technology for good. Resolving this paradox not only helps to clarify the value tensions in the application of digital technology but also provides possible pathways for transforming the value demands of technology for good into operational institutional arrangements. In light of this, this study focuses on the core issue of the paradox of technological equity. Based on a systematic review of the evolution of the concept of technological equity, the theoretical framework, and the inherent contradictions, an integrated analytical framework of “paradox manifestations→underlying causes→governance pathways” was constructed. The study deeply analyzed the fourfold structural tensions that led to the paradox of technological equity, and accordingly proposed a resilient governance approach centered on the three-dimensional synergy of “institution—technology—society”, aiming to provide systematic theoretical support and governance solutions to address the structural dilemmas in the practice of technological equity.
The research findings are as follows: (1) Technological equity is not a simple issue of technological diffusion; rather, it is a progressive composite proposition encompassing accessibility of access, empowerment of capabilities, and justice of outcomes. Its ideal goal is to achieve equal access to resources, fair distribution of benefits, and reasonable sharing of risks through institutional arrangements and technological design. However, in practice, due to the interweaving of multiple tensions, paradoxical characteristic that the coexistence of empowerment effects and alienation risks is presented. (2) The root cause of the paradox of technological equity lies in the interweaving and superposition of four tensions: the intersection of equity ideal and technological practice, market logic and social goal, social structure and technological system, and governance system and technological evolution. (3) The governance approach needs to complete the paradigm shift from fragmented regulation to systematic resilience-based collaborative governance. Through multiple dimensions such as institutional reconfiguration, technology domestication, entity collaboration, and global cooperation, a dynamic coupling effect can be formed between institutional adaptability supply, technology manageable design, and social initiative activation. This will provide a resilient governance practical approach for systematically resolving the paradox of technological equity.
This study reveals the deep generative mechanism of the technology equity paradox, providing theoretical support and practical approaches for its systematic governance. However, this study also has some limitation. Methodologically, it mainly relies on normative analysis and case induction, and lacks empirical verification using large sample data to test the effectiveness of the governance approach for the technology equity paradox. Future research can deepen in two aspects: Firstly, at the methodological level, cross-national panel data can be adopted to empirically test the synergy effect of the three?dimensional governance elements of institutions, technology and society; Secondly, at the contextual level, attention should be focused on emerging technology iteration scenarios to explore the dynamic adaptability of governance tools.
Amid growing global technological competition, understanding the reasoning behind U.S. artificial intelligence (AI) policy is crucial for evaluating its strategic impact and potential risks for China. Building on the theoretical framework of technological nationalism, this study introduces a four-dimensional analytical model covering national technological strategic positioning, the establishment of technological sovereignty, the connection between technology and national security, and the development of global technological discourse power.
Using the Dynamic Topic Model (DTM), we analyze 306 U.S. AI policy documents issued from 2015 to 2025 to identify key thematic trends and map the evolving policy emphasis. Complemented by keyword co-occurrence analysis, the study examines the interdependencies among policy themes to reveal how technological nationalism has become structurally embedded in the United States’ AI policymaking process and how this shapes its external technological stance toward China. The findings demonstrate that U.S. AI policy is characterized by a strong technological nationalist orientation. Domestically, the United States tightly links AI development with national security goals, creating a legitimacy narrative around “national security threats” to justify restrictive technological measures targeting China. Internationally, it promotes governance frameworks infused with American values and fosters alliances with like-minded partners, thereby institutionalizing a bloc-based regime of technological containment.
In response, the study proposes three strategic pathways for China—foundational breakthroughs, standard elevation, and partnership restructuring—aimed at mitigating external technological vulnerabilities and enhancing its strategic autonomy and competitiveness within the global AI landscape. These findings contribute to a deeper understanding of the geopolitical dimensions of AI governance and offer actionable insights for national technology strategy in an era of renewed techno-political rivalry.
The rapid spread of generative artificial intelligence is reshaping how content is produced and disseminated on social media platforms, intensifying challenges such as cross-platform cascading diffusion of misinformation, rising moderation costs, and blurred accountability chains. Current practice largely relies on linear “human + algorithm” governance and minimal compliance. Addressing gaps in existing research—namely the lack of mapping between technical architectures and governance tiers, overly static collaboration mechanisms, and the difficulty of parameterizing and transparently assessing the safety-efficiency-fairness trade-offs—this paper proposes an integrated, four-tier collaborative governance framework based on the co-construction of “technical architecture-governance level.” At its core, the framework establishes a clear correspondence of “who is responsible—what is governed—how it is governed—how it is evaluated,” linking technical layers (data, models, content distribution) with governance tiers (institutional/regulatory, industry, platform, societal). It is supplemented by a bidirectional governance loop (top-down rules and bottom-up risk signals) and cross-platform coordination mechanisms, translating abstract governance principles and values into measurable, adjustable, and auditable metrics and weighting schemes. This enables multi-stakeholder collaboration to be operationalized as executable and assessable meso-level designs, bridging gaps between regulations, standards, and platform operations; supporting agile transitions between routine and crisis contexts; shortening response latency; curbing cascading spread; and improving exposure fairness and external accountability at a manageable efficiency cost. The framework provides a theoretical foundation and a systematic analytical toolkit for platform governance in the generative AI era.
Big data algorithms are not valueless. Social goals are inevitably embedded in the process of algorithm development.And it brings new problems such as algorithm bias and even algorithm discrimination at the same time. Algorithm bias is a very hidden intelligent copy of social bias, including the bias in training data, the bias of algorithm designers, and the bias of the algorithm itself. It is necessary to further distinguish algorithm bias and algorithm discrimination, prevent political embedding and capital intervention in algorithm design, and avoid big data algorithms becoming tools for interest groups and power departments to achieve economic, social and even political goals. At the same time, effective algorithmic supervision and review mechanisms should be employed to ensure and promote future social justice through algorithmic justice.
Different from existing literature that primarily focuses on the direct impact of government-guided funds (GGFs) on invested enterprises, this paper innovatively examines how such funds influence innovation interactions among multiple invested enterprises. A limited body of literature has found that GGFs can significantly enhance corporate innovation performance. However, these investigations have largely concentrated on direct effects (such as capital provision and legitimacy empowerment), with limited in-depth analysis of positive externalities, particularly regarding innovation spillovers and industrial ecosystem development. Drawing on ecosystem theory, this study posits that GGFs can serve as ecosystem architects, constructing an innovation ecosystem centered on the GGFs through common shareholder connections, encompassing fund managers, co-investors, and numerous portfolio companies. Within this system, GGFs perform core organizational and coordination functions, enhancing entrepreneurial firms' embeddedness in the ecosystem by coordinating innovation interactions among ecosystem members. Specifically, GGFs promote innovation spillovers among portfolio companies through three mechanisms: information transmission, resource integration, and relational governance.
This study employs GGFs investment event data from 2003-2019, patent application data and patent citation data from 2000-2022, and utilizes parametric and non-parametric event study methods to systematically examine the effects and mechanisms of GGFs on intra-portfolio innovation spillovers. Empirical results demonstrate that after firms join GGFs portfolios, their innovation interactions with other portfolio companies are significantly enhanced. Mechanism tests reveal that GGFs facilitate inter-firm innovation collaboration through information transmission, resource integration, and relational governance channels, with these promotional effects being more pronounced when portfolio companies exhibit higher geographical dispersion, GGFs operate at higher administrative levels, and trust levels between entrepreneurial firms and portfolio companies are lower. Heterogeneity analysis results indicate that intra-portfolio innovation interactions are significantly enhanced when funds possess extensive investment experience, participate as lead investors, and have managers with state-owned backgrounds, or when entrepreneurial firms are high-tech enterprises, are in non-early development stages, represent cross-regional investment projects, and operate in regions with well-developed intellectual property protection systems. Further research reveals that intra-portfolio innovation spillovers effectively promote corporate innovation output, particularly demonstrating superior performance in core technology breakthroughs. These findings suggest that policymakers should deepen their understanding of the ecosystem-enabling functions of GGFs and refine institutional designs to fully unleash their positive externalities; fund managers should focus on optimizing operational mechanisms and strengthening innovation leadership and resource allocation capabilities; and enterprises should proactively integrate into innovation ecosystems and improve their innovation interaction strategies with portfolio companies.
The theoretical contributions of this study manifest in three dimensions: First, from an innovation ecosystem perspective, it deepens theoretical understanding of the mechanisms through which GGFs influence corporate innovation performance, extending the research perspective from individual firms to the portfolio level and providing ecosystem-level empirical evidence for how GGFs promote corporate innovation activities. Second, it reveals the unique functions of GGFs—as special common shareholders with public policy orientation—in shaping internal ecosystem structures and driving innovation spillovers, elucidating how equity-based coordination mechanisms uniquely facilitate innovation interactions within ecosystems and ultimately advance key core technologies. Third, by thoroughly examining the moderating effects of fund characteristics, firm attributes, and regional governance environments on the heterogeneous effects of intra-portfolio knowledge flows, it provides important insights for optimizing GGFs operational models and advancing innovation ecosystem evolution.
The integration of innovation chains and industrial chains in emerging industries is essential for aligning technological and industrial innovation. Digital platforms can effectively streamline this integration by serving as connectors and orchestrators of resources. Drawing on resource orchestration theory, this study examines the mechanisms of dual-chain integration in emerging industries empowered by the Qin Chuangyuan digital platform, with empirical cases from the hydrogen energy, intelligent connected vehicles, photonics, and new materials sectors. The findings reveal that: (1) dual-chain integration evolves toward technology breakthrough-oriented, industry cultivation-oriented, or chain-ecosystem integration-oriented pathways, shaped by each industry’s distinct resource endowments and contextual conditions; (2) under different integration pathways, digital platforms orchestrate resources through organizational, structural, and field-level empowerment, thereby generating three types of integration capabilities—linkage, interaction, and coordination; and (3) these capabilities lead to three integration patterns: single-point linkage, full-chain transformation, and chain-ecosystem integration. This study provides theoretical insights and decision-making implications for nurturing emerging industries, deepening dual-chain integration, and promoting sustainable development.
Collaborative R&D by agribusinesses to trigger knowledge recombination creation, thereby enhancing innovative entrepre- neurial opportunities, is a complex systems management issue. Accordingly, this study constructs a “motivation-capability-trigger” element framework for collaborative R&D in agribusinesses based on Fogg’s Behavior Model. Furthermore, drawing on absorptive capacity theory and incorporating a configurational perspective and a complex mediation model, it analyzes how different elements of collaborative R&D jointly configure to drive knowledge recombination creation within agribusiness. This study finds that high knowledge recombination creation can be achieved through three types of collaborative R&D configurations: orientation-driven, market-driven, and institution-driven. These configurations not only directly facilitate the realization of innovative entrepreneurial opportunities but also indirectly promote such realization through knowledge recombination creation; that is, knowledge recombination creation plays a complex mediating role between the three configurations and the realization of innovative entrepreneurial opportunities. The findings have important theoretical and practical implications for how agribusinesses can promote the realization of innovative entrepreneurial opportunities through collaborative R&D strategies.
As a minority and marginalized group long overlooked in mainstream entrepreneurship research, underdog entrepreneurs face distinctive and often difficult-to-change structural challenges in their entrepreneurial journey, that arise from their persistent disadvantaged positions. Investigating the entrepreneurial underdog status is not only crucial for providing targeted support to this group but also offers new theoretical perspectives for advancing inclusive and context-sensitive entrepreneurship research. Despite growing academic interest, the conceptualization and measurement of entrepreneurial underdog status remain underdeveloped, with most existing studies relying on fragmented theoretical discussions or secondary data proxies. To address this gap, this study adopts a mixed-methods approach, combining qualitative inquiry based on grounded theory with quantitative validation across three rounds of surveys, to systematically explore the content, structure, and predictive validity of entrepreneurial underdog status.
Drawing on semi-structured in-depth interviews with 27 representative underdog entrepreneurs from diverse backgrounds across China, including individuals with disabilities, migrant workers, female entrepreneurs, and veterans, the study identifies four core categories of challenges that collectively define entrepreneurial underdog status: resource endowment deficits, physiological constraints, cognitive limitations, and external environmental barriers. Through open, axial, and selective coding procedures, these dimensions are further refined into a coherent theoretical framework. An initial 22-item scale was developed and subsequently administered to two independent samples (N?=191; N?=425). Exploratory and confirmatory factor analyses yield a robust four-factor, 14-item scale with satisfactory internal consistency, convergent validity, and discriminant validity. The four dimensions are respectively conforming to the classical theoretical perspectives, such as resource-based view, human capital theory, social cognition theory, and institutional theory, thereby ensuring the conceptual grounding of the measurement tool.
To further evaluate criterion-related validity, the study draws on self-regulation theory and examines the nonlinear relationship between entrepreneurial underdog status and entrepreneurial exit intention using two-wave data from 317 entrepreneurs. Results reveal a significant U-shaped relationship: at low levels of underdog status, individuals tend to adopt approach-oriented self-regulation strategies, such as persistence and resilience, which reduce exit intentions; at high levels, however, defensive mechanisms dominate, leading to increased withdrawal intentions. This curvilinear pattern remains robust after controlling for demographic characteristics, firm characteristics, and fear of failure. Further dimensional analyses show that the U-shaped effect is primarily driven by resource endowment and cognitive challenges, whereas physiological constraints and external environmental barriers exhibit linear positive effects on exit intentions, reflecting their distinct proximal versus distal roles in the entrepreneurial process.
Taken together, this study makes three key contributions. First, it clarifies the multidimensional structure of entrepreneurial underdog status, providing a systematic analytical framework for future theoretical development. Second, it develops and validates a reliable measurement scale, thereby enabling empirical research to move beyond conceptual discussions. Third, it uncovers a nuanced U-shaped mechanism linking entrepreneurial underdog status to exit intentions, challenging conventional linear assumptions and extending the application of self-regulation theory in entrepreneurship research. Practically, the findings offer actionable insights for underdog entrepreneurs to reframe their disadvantages and for policymakers to design targeted, inclusive support systems. The study also acknowledges limitations regarding self-reported data, cultural specificity, and cross-sectional design, while outlining promising directions for future research, including dynamic, cross-cultural, and longitudinal investigations.
The paradigm of enterprise technological innovation is undergoing a significant transition from a traditional "value chain" framework to a more complex, multi-stakeholder "value network". This transition is driven by the rise of global trade protectionism, accelerated by the urgency for domestic technological self-sufficiency, particularly in critical sectors like the semiconductor industry, where "chokepoint" technologies pose severe national security and economic stability risks. The semiconductor industry’s innovation ecosystem is shifting toward dense regional networks, demanding collaborative and non-linear approaches to value creation, contrasting sharply with the sequential and linear nature of the traditional value chain model. This paradigm shift exposes enterprises to new challenges, including cognitive limitations, technological uncertainty, high coordination costs, and capability obsolescence, which exacerbate the risk of falling into the "valleys of death" across the innovation lifecycle. These valleys manifest as critical breaks in the network nodes from basic research through application development to industrialization and scaling.
This study systematically investigates how enterprises can effectively navigate these challenges, focusing on the six distinct types of innovation valleys of death that emerge within this value network context. Adopting the "Awareness-Motivation-Capability" (AMC) perspective, which is highly suited for analyzing enterprise responses to external changes and strategic challenges, we conduct a multi-case study on eight semiconductor subsidiaries of the Changguang Group, an enterprise group integral to China's high-tech development. We track their archival records, patent portfolios, and financial data between 2014 and 2024 to provide robust longitudinal evidence.
Our empirical findings reveal two distinct strategic paths for overcoming innovation valleys, contingent on the type of core technology being pursued.
First, for "chokepoint" technologies (those significantly lagging behind international leaders and facing external restrictions), the most effective strategy follows the path of "social-expectation-based awareness—developing intrinsic motivation—cultivating core capabilities". Social expectation awareness, driven by comparison to industry benchmarks, compels the enterprise to adopt a differentiation strategy to break free from "chain"-based path dependency. Intrinsic motivation, rooted in a sense of mission and long-term commitment, facilitates continuous learning and the tacit knowledge accumulation necessary to overcome network monopolies and master crucial know-how. Finally, cultivating core capabilities allows the firm to achieve rapid product transformation, scale production efficiently, and realize sufficient network benefits, thereby securing the autonomous control and industrialization of critical core technologies.
Second, for "game-changer" technologies (asymmetric technologies originating from internal, primary innovation that are difficult to imitate), the successful approach is "self-expectation-based awareness—developing extrinsic motivation—cultivating dynamic capabilities". Self-expectation awareness, based on the enterprise’s past performance, provides a stable internal anchor for sustained investment in basic research, mitigating the risk of divergent goals among multiple stakeholders (such as government, research institutes, and market-driven firms). Extrinsic motivation, often triggered by external pressures (e.g., policy support, competitive attacks), facilitates the coordination of diverse actors and the infusion of external resources, thereby resolving the challenges of adoption network delays in the transition from application to product. Crucially, dynamic capabilities (sensing, seizing, and reconfiguring) enable the enterprise to quickly adapt to market changes, integrate resources, and reconstruct business models, effectively addressing diffusion network delays and ensuring the rapid dissemination of asymmetric technologies and successful business model innovation.
This study contributes to the literature on innovation management by extending the AMC framework into the context of innovation valleys of death in value networks and providing a differentiated strategic response mechanism for "chokepoint" and "game-changer" technologies. Practically, the findings offer vital guidance for enterprises in critical industries to proactively respond to paradigm shifts, successfully commercialize R&D outcomes, and enhance national competitiveness.
Since its introduction, the theory of National Innovation System (NIS) has become an important theoretical basis for the Chinese government to formulate innovation policies, and has had a wide impact on China's scientific and technological economic development. However, NIS theory typically treats government as an actor alongside firms and universities. This perspective is insufficient for explaining the coordinating and organizational functions that the Chinese government has played in building the NIS. To address this limitation, this article argues that the Chinese government occupies two structural positions within the NIS. First, as architectural role (Type I), it operates outside the system to design rules and define the division of responsibilities. Second, as interventionist role (Type II), it enters the system to mobilize resources and correct operational deficiencies. By examining five stages in the development of China’s NIS, this study finds that the relative importance of these two roles has shifted dynamically in response to changing national objectives, and the two roles have evolved into a governance pattern characterized by a clear distinction between external design and internal intervention. Based on this experience, the article offers three contributions to the NIS literature. First, the role of government should transcend its conventional positioning as merely an institutional provider; second, the government’s role is closely intertwined with national objectives; and third, the governance of latecomer NIS requires a stronger emphasis on dynamic adaptability.
In the context of fiscal constraints faced by local governments, it is an imperative issue to improve the allocation of science and technology (S&T) expenditure and enhance the efficiency of regional innovation systems. Research and development activities can be divided into basic research, applied research, and experimental development. Based on the resource-based view, extant studies suggest that the structure of local S&T expenditure as a critical resource endowment, directly shapes innovation performance. Beyond the resource-based view, local governments must weigh central mandates and peer competition—rather than efficiency alone—when setting S&T budgets, owing to multidimensional performance metrics.
Extant literature shows that local governments exhibit two distinctive traits when allocating S&T expenditure under the institutional pressure. First, local government complies the S&T strategies of central government. Within China’s centralized political system, subnational governments consistently align their fiscal priorities with those of the central government. When a national S&T initiative is announced, local governments rapidly mobilize resources to establish similar projects—such as provincial-level science parks, key laboratories, and engineering research centers—that closely reflect the central policy blueprint. Second, there is a convergence in S&T expenditure across local governments. As fiscal S&T expenditure has become a key performance metric in governmental evaluation systems, competition among local governments in the S&T domain has significantly intensified. In pursuit of competition advantage, local officials routinely benchmark their S&T expenditures against peer regions and recalibrate their budgetary allocations in response. This strategic behavior results in a noticeable homogenization of S&T funding patterns across local governments.
Although existing studies indicate that local governments align their fiscal S&T expenditures with both central and peer governments, few studies have examined the impact of such conformity on regional innovation performance. Furthermore, while prior research has mainly examined convergence in the scale of S&T expenditure, little attention has been given to structural conformity in fiscal S&T expenditure. Under the current fiscal constraints, how local governments choose the structure of S&T expenditures to improve regional innovation performance is a critical issue.Based on organizational legitimacy theory, this study examines the relationship between compliance, convergence of local government of S&T expenditure structure and regional innovation, as well as the moderating role of local official turnover. Using an unbalanced panel dataset from 31 provinces (excluding Hong Kong, Macao, and Taiwan) from 2013 to 2022, we employ a fixed-effects ordinary least squares (OLS) model.
The findings reveal that compliance with central government in fiscal S&T expenditure structure is positively associated with regional innovation. That is, when local governments align their S&T expenditure structure with that of the central government, they gain access to central S&T resources and policy support, thereby fostering regional innovation. Conversely, there is an inverted U-shaped relationship between convergence with peer governments and regional innovation. The moderate level of conformity in S&T expenditure structure among local governments helps accumulate R&D resources and facilitates innovation; however, beyond a certain threshold, the costs of homogenization outweigh the legitimacy benefits, ultimately hindering regional innovation. Furthermore, the local official turnover weakens the positive relationship between compliance and regional innovation but does not significantly affect the relationship between convergence and regional innovation.
Against the backdrop of global economic integration and rapid technological advancement, small and medium-sized industrial enterprises (SMEIs) have emerged as a crucial force in the national economy, with technological innovation serving as the core driving force for their development. Exploring the driving effects between the two is of great significance for policy formulation and enterprise development. This study constructs an indicator system for evaluating the technological innovation and development level of SMEIs. It employs classical cycle theory and the growth rate cycle composite index to quantitatively assess their capabilities and levels, and utilizes impulse response function analysis and variance decomposition to dissect the driving mechanisms. The findings are as follows: (1) In the classical cycle, both the technological innovation input and output indices exhibit an overall upward trend, with the output index growing at a faster pace. After a brief decline in the enterprise scale index from 2017 to 2018, growth resumed due to policy support. (2) The growth rate cycle reveals that there is a lag effect in the fluctuations of the growth rate cycle indices for input and output. The fluctuation trends of enterprise scale and innovation capability are consistent, but the scale fluctuations are more pronounced. (3) Impulse response analysis indicates that the initial impact of technological innovation input is moderate. An initial increase in the output index leads to a temporary fluctuation in enterprise development followed by an upward trend and can promote the formation of a virtuous cycle in the input index. (4) Variance decomposition shows that both the input and output indices have a four-period lag effect on enterprise development. The variance contribution rate of the output index (approximately 28%) is higher than that of the input index (6%), indicating a more significant driving role of output. Both governments and enterprises should attach great importance to technological innovation. While increasing investment, they should strengthen the monitoring of technological achievement transformation, promote technological innovation output, and drive the steady development of SMEIs.
Manufacturing single champion firms (MSCFs), as industrial “bellwethers,” are pivotal in fostering the virtuous development of upstream and downstream industrial chains through their coopetition behaviors. An increasing number of MSCFs are shifting from the traditional “zero-sum game” competitive mindset to a “positive-sum game” coopetition mindset. However, translating this coopetition mindset into sustained practice remains challenging. A key difficulty lies in identifying and sequencing the driving factors of coopetition within differentiated scenarios. Given the limited prior research on this issue, it is of significant theoretical and practical value to explore the key drivers and their action paths for MSCFs’ coopetition behaviors across different scenarios, using simulation methods to uncover the underlying logic of how these driving factors transform as scenarios change.
This study draws on 7931 unbalanced panel observations from 103 MSCFs spanning the years 2017 to 2024. Employing bibliometric analysis and keyword co-occurrence networks, we identify the driving factors of coopetition behaviors and construct a 2×2 “technology-market” scenario matrix. Using the entropy weight method and DEMATEL-NK techniques, we systematically examine scenario-specific driving factors and their implementation paths. The findings reveal that: (1) Innovation Advantage Empowerment (IAE), Strategic Orientation Change (SOC), Network Relationship Construction (NRC), Organizational Risk Control (ORC), and Key Resources Development (KRD) constitute the fundamental drivers of coopetition behaviors in MSCFs. (2) Through 100000 simulations conducted in MATLAB 2016b across four scenarios, the realization paths for coopetition behaviors are identified based on the highest frequency of element occurrence. The paths vary by scenario: under the conservative-maintenance scenario, the path is ORC→IAE→SOC; under the technology-driven scenario, it is NRC→SOC→IAE→KRD; under the market-driven scenario, it is NRC→SOC→ORC; and under the radical synergy scenario, it is NRC→ORC→IAE→SOC. Case analyses further validate these differentiated paths. (3) The study identifies two transition logics for key driving factors when scenarios change: SOC acts as a “transformation hub” when shifting from a conservative-maintenance scenario to an adjacent (market- or technology-driven) scenario, while ORC serves as the “transformation hub” in the transition from a market-driven to a radical synergy scenario.
The study contributes to coopetition theory and offers practical guidance for MSCFs to effectively prioritize and coordinate key drivers in sustaining coopetition relationships. Theoretically, by integrating a randomly generated fitness landscape with functional computation, this approach mitigates limitations inherent in the traditional NK model, enhancing the objectivity and robustness of the results. The multi-method research design enriches methodological diversity in coopetition research and expands the conceptual framework for understanding how MSCFs’ coopetition drivers adapt to differentiated scenarios. Practically, managers should activate key driving factors and optimize their allocation sequence according to the specific scenario in which the firm operates. Moreover, MSCFs need to dynamically adjust to scenario changes, reallocating resources promptly and transitioning to new implementation paths to ensure the effectiveness and sustainability of coopetition. From a policy perspective, targeted support should be provided to facilitate coopetition in different scenarios. Governments should also focus on establishing a fair, open, and institutionally supportive environment that encourages coopetition, helping MSCFs adapt key drivers in response to scenario shifts and thereby promoting the sustainability of coopetition relationships.
The new wave of technological revolution is driving a paradigm shift in scientific research toward organized research models. The key to translating organized research from theoretical exploration to practical efficacy lies in the allocation and synergy of innovation elements. The Regional Innovation and Development Joint Fund as a typical practice of scientific funds under organized research, serves as an important benchmark for evaluating the transition from theoretical exploration to practical application in organized research. This study takes the Shandong Regional Joint Fund as an example and uses Necessary Condition Analysis (NCA) and fuzzy-set Qualitative Comparative Analysis (fsQCA) to investigate how different configurations of factors influence project effectiveness from a configurational perspective. The study aims to explore how scientific funds drive the improvement of their effectiveness through the configuration matching of multiple elements under organized scientific research. We try to open the 'black box' of improving the efficiency of organized scientific research. The study finds that: (1) The improvement of scientific fund efficiency depends on the coordinated matching of multiple elements, and a single element does not constitute a necessary condition for high efficiency. We found that among elements such as the project leader's (PI) ability, PI age, team heterogeneity, team collaboration, disciplinary foundation, and institutional reputation cannot individually serve as necessary conditions for the high efficiency of regional joint funds. The effective way to improve the efficiency of scientific funds is the complex combination of multiple conditions. (2) Disciplinary foundation is a key factor affecting project effectiveness and play a rigid supporting role. With the advancement of Double First-Class Initiative in China, first-class disciplines have strengthened their core position in resource aggregation and innovation ecosystems. The finding that disciplinary foundation serves as a critical support for project effectiveness not only empirically validates the "Matthew Effect" in scientific research but also demonstrates the rationality of the Double First-Class Initiative in China being discipline-based from the perspective of performance. (3) There are multiple configurational pathways to enhance the effectiveness of regional joint scientific funds: Resource-driven young leader type, Discipline-supported team growth type, Government-Industry-academic collaboration type. Among these, the Resource-driven young leader type is the most common. We also found that there is a certain substitution effect between different elements. For example, under a strong discipline, elastic factors such as PI capabilities, team collaboration, and institutional reputation can be substituted through combination and replacement. This finding provides an important basis for the allocation of scientific research resources. These elements can be reconfigured through combinatorial optimization to enhance project efficacy. This paper breaks through the limitations of traditional single-factor analysis of project efficiency. It reveals the impact of the combination and matching of scientific research elements in organized scientific research from a configurational perspective, providing a new paradigm for scientific research management under the complex system perspective. The conclusion that "disciplinary foundation is a rigid support for enhancing efficiency, and elastic elements exhibit conditional substitution effects" demonstrates that organized scientific research is the result of the synergistic matching between rigid resource support and elastic elements. This not only provides an important basis for optimizing the allocation of scientific research resources in practice, but also solves the "paradigm code" of organized scientific research from theoretical exploration to practical verification.