Digital technologies have brought profound transformations to the entrepreneurial activities, particularly by enabling new forms of firm-user interactions. Interaction orientation—a firm's strategic tendency to interact with and learn from users—has been increasingly recognized as a key enabler of innovation, its effectiveness remains context-dependent and underexplored in digital entrepreneurial settings. Prior research often highlights its linear benefits while overlooking potential drawbacks and the mechanisms through which it operates. This gap is especially salient in the context of digital entrepreneurial, where uncertainty and limited resources make decisions more complex. This study draws upon the Lean Startup Theory and investigates the “double-edged sword” effect of interaction orientation on innovation performance in digital new ventures, examines experimentation as a key mediating mechanism and incorporates latent need identification capability as a boundary condition.

Based on a multi-stage survey of digital new ventures, this study provides empirical evidence of an inverted U-shaped relationship between interaction orientation and innovation performance. When interaction becomes excessive, it generates diminishing returns—leading to information overload, resource strain, and decision-making fatigue due to redundant or conflicting user input. In addition, this study confirms the mediating role of experimentation, thus unpacking the black box of how user interaction drives innovation success. In dynamic and uncertain digital markets, experimentation enables ventures to iteratively test assumptions, validate MVP, and optimize product-market fit. Interaction orientation supplies the raw input—user insights and behavioral signals—while experimentation transforms this input into actionable innovation outcomes. Moreover, this study highlights the moderating effect of latent need identification capability, offering new insights into the conditions under which interaction orientation is most effective. Ventures with strong capabilities to detect unspoken, future-oriented user needs are more adept at filtering valuable signals from user interactions and translating them into focused experiments. The moderation analysis reveals that such ventures reach the optimal point of experimentation at lower levels of interaction orientation, suggesting that latent need identification sharpens the strategic value of firm-user interaction.

Theoretically, this study contributes in three key ways. First, this study clarifies the "double-edged sword" effect and underlying mechanism of interaction orientation on innovation performance in digital entrepreneurship contexts, offering a contextualized explanation to reconcile conflicting findings in existing literature. By focusing on the characteristics of digital entrepreneurship, this study provides both theoretical justification and empirical evidence for the inverted U-shaped effect of interaction orientation on innovation performance of digital new ventures. Second, this study uncovers the inverted U-shaped relationship between interaction orientation and experimentation, addressing the research gap on the antecedents of experimentation from the demand-side perspective. While prior studies have primarily explored the drivers of entrepreneurial experimentation from the supply side, the highly dynamic nature of user needs in digital contexts necessitates timely sensing and response to guide effective experimentation. This study demonstrates the positive effect of interaction orientation on experimentation in digital new ventures, thereby enriching the antecedent research of entrepreneurial experimentation in digital settings. Third, this study highlights the critical contingency role of latent need identification capability in digital entrepreneurship, expanding the boundary condition literature on interaction orientation. Enabled by digital technologies, digital new ventures are increasingly capable of capturing and interpreting user needs. In this context, the capability to identify latent needs serves as a crucial catalyst for realizing the benefits of interaction orientation, yet it remains underexplored in existing research.

Practically, this study highlights key innovation management strategies for digital new ventures. They should leverage digital technologies to interact deeply with users, personalize products and services, and predict latent needs through data analysis. They must remain agile by adjusting innovation directions based on market changes and iterating MVPs rapidly. At the same time, these new ventures should avoid excessive interaction that leads to information overload and resource dispersion. Prioritizing relevant user feedback, integrating information effectively, and allocating resources strategically are crucial for enhancing innovation performance in fast-changing digital environments.

Against the backdrop of the deep integration between global technological innovation waves and institutional reforms, patent thicket has emerged as a quintessential phenomenon in the evolution of intellectual property systems, characterized by property right fragmentation and unbalanced patent allocation. Despite its growing relevance in modern industrial landscapes, existing literature has not fully explored the specific mechanisms through which its structural contradictions influence the high-quality development of enterprises. This research aims to bridge this gap by adopting a multi-theoretical framework integrating resource-based theory, transaction cost theory, and dynamic capability theory, seeking to systematically dissect the complex interplay between patent thicket dynamics and enterprise development trajectories.

The study constructs a two-dimensional analytical model explicitly incorporating two critical dimensions: density-induced fragmentation costs and quality-driven synergistic benefits. Departing from traditional single-dimensional approaches, this model recognizes patent thicket as an interactive system where the quantitative accumulation of patents and their qualitative attributes co-evolve to shape organizational outcomes. Leveraging a comprehensive dataset of A-share listed companies from 2010 to 2023, the research employs panel data fixed-effect models to establish causal relationships and moderating effect tests to explore contextual contingencies. This empirical strategy enables a nuanced examination of how patent thicket operates across diverse organizational and environmental contexts.

Key findings reveal a dualistic influence of patent thicket on enterprise high-quality development. Specifically, patent thicket density demonstrates a significant inhibitory effect, primarily attributed to escalating transaction costs, innovation bottlenecks, and strategic lock-ins arising from fragmented intellectual property rights. Conversely, patent thicket quality—operationalized through indicators including patent originality, technological centrality, and commercial value—exerts a robust promoting effect by facilitating knowledge spillovers, enhancing competitive positioning, and enabling dynamic capability building.

The study further uncovers that environmental dynamism, characterized by rapid technological changes and market fluctuations, and environmental complexity, reflecting stakeholder diversity and regulatory landscapes, act as positive moderators. These factors amplify both the inhibitory effect of density and the promotive effect of quality, indicating that in volatile environments, challenges posed by patent density intensify while advantages of high-quality patents become more pronounced.

Additionally, the research highlights heterogeneous effects across firm characteristics. Small-scale enterprises exhibit heightened sensitivity to patent density, likely due to their limited resource buffers and bargaining power in patent licensing negotiations. In contrast, high-quality patents demonstrate a more significant innovation-driven effect for enterprises in highly competitive industries, where technological superiority serves as a critical differentiator. These findings illuminate how resource endowment, such as firm size, and market structure, such as industry competitiveness, moderate the patent thicket effect, providing insights into context-specific strategic responses.

By reconstructing patent thicket as an interactive system of density and quality, this study transcends conventional paradigms and offers a more nuanced understanding of its operational laws in dynamic environments. Theoretical contributions include integrating fragmented literatures, proposing a novel analytical framework, and identifying contingency mechanisms. Practically, the findings provide strategic guidance for Chinese enterprises to navigate patent thicket challenges, particularly in formulating balanced patent acquisition strategies, enhancing patent quality management, and adapting to environmental uncertainties. This research thus serves as a vital theoretical anchor for promoting high-quality development within China's evolving innovation ecosystem.

New quality productive forces represent a pioneering pathway of advanced high-quality development. Driven by technological innovation, the productive forces embody the productive capacity arising from critical and disruptive technological breakthroughs and serve as a vital means for achieving high-quality development. Recently, artificial intelligence (AI) has emerged as a transformative technology reshaping economic and social operations, as well as driving productivity leaps, assuming an increasingly prominent role in international competition and national development. As an institutional framework, AI policy holds substantial potential to guide and promote AI-enabled development of new quality productive forces. However, the mechanisms through which AI policy empowers new quality productive forces, as well as the contingent conditions governing this empowerment effect, remain under-researched through systematic empirical analysis. Existing evidence on this relationship has relied predominantly on theoretical discussions, while the limited quantitative literature focuses primarily on the firm level, leaving a gap in city-level empirical analysis. Given that cities serve as critical hubs for innovation clustering and industrial transformation, such analysis carries greater policy relevance. Furthermore, prior studies have emphasized the technology itself rather than policy effects. Drawing on an analytical framework encompassing technological innovation, production-factor allocation, and industrial upgrading, this study exploits the gradual implementation of China’s National New Generation Artificial Intelligence Innovation and Development Pilot Zones since 2019 as a quasi-experiment to quantitatively examine the impact of AI policy on multiple dimensions of new quality productive forces, including new workers, new means of production, new objects of labor, new technologies, new production organization, and new data factors. Results from a staggered difference-in-differences (DiD) estimation indicate that AI policy has increased new quality productive forces by at least 5.84%. Mechanism analysis reveals that the policy enhances new quality productive forces through stimulating technological innovation, optimizing production-factor allocation, and catalyzing industrial upgrading. Heterogeneity tests further indicate that the policy’s effect on new quality productive forces is more pronounced in cities with higher marketization levels and stronger technological capacity. This study offers insights into how AI policy can enhance high-quality development and foster the advancement of new quality productive forces. Sustained, AI policy-driven progress in new quality productive forces requires prioritizing technological innovation as the core driver, optimizing production-factor allocation as a supporting pillar, and advancing industrial upgrading as the implementation pathway, while concurrently improving cities’ marketization levels and technological capacity. Policy implementation should first establish a multi-tiered innovation ecosystem, such as creating an integrated support mechanism spanning basic research, applied research and commercialization. Second, efforts should continue toward building a unified national market for data factors, promoting institutional innovation in data ownership confirmation, circulation, trading, and regulation. Third, policies should foster coordinated development between traditional and emerging industries—on one hand, advancing the intelligent transformation of traditional industries, and on the other, cultivating strategic emerging industries and promoting the development of new AI-driven models. Finally, AI policy should enhance targeting precision and equity through guiding market entry cities with lower levels of marketization while reducing algorithm adoption barriers in technologically disadvantaged areas via national big data hub clusters and cross-regional scheduling networks.

From the economic development experience of other countries in the world, it can be seen that the service-oriented stage is a differentiation stage for whether growth can be sustained and whether economic catch-up can be successful. To successfully overcome the middle-income trap and ensure the sustainability of economic growth, it is necessary to apply the theory of new quality productivity to practice, and then promote the development of China's service industry towards high-quality direction, that is, the new quality of service industry. The new quality of the service industry is a high-quality service economic form, characterized by endogenous knowledge production, which reshapes the efficiency model through the optimization of the service industry structure and the dynamic balance mechanism between supply and demand. With the transformation of the new generation of information technology, digital technology innovation has undoubtedly become the most important new quality productivity during the Fourth Industrial Revolution, and is a strategic highland for major economies around the world to compete for layout. So, can and how can digital technology innovation promote the high-quality development of the service industry? This is the core question that this article intends to answer.

The article elaborates on the theoretical mechanism based on the logic of "elements to products and then to industries". Among them, the element level emphasizes that knowledge becomes the core driving force of economic growth; At the product level, there is a service upgrade from standardization to personalization; At the industrial level, it is manifested as the rise of knowledge intensive service industries. Subsequently, the article constructed an evaluation index system for the new quality development level of the service industry, which includes 5 core dimensions, 14 primary indicators, and 23 secondary indicators. The analysis pointed out that the new quality development of China's service industry presents characteristics of overall low-level slow rise, significant stage heterogeneity, uneven development of each core dimension, and regional development polarization. Finally, the article conducted empirical tests using panel data from 30 provinces in China from 2011 to 2023, and found that digital technology innovation can significantly promote the high-quality development of the service industry. This conclusion still holds true after a series of robustness tests. Mechanism testing shows that upgrading supply side factors, driving demand side momentum, and ensuring institutional environment constitute a triple empowerment path, which can more fully unleash the empowering potential of digital technology innovation for the new quality development of the service industry.

The article studies the application of the theory of new quality productivity to the practice of service economy development, providing theoretical support and policy inspiration for China to seize opportunities for digital technology innovation and promote high-quality economic development. Firstly, it is necessary to establish a "dimensional regional" dual balanced governance system to solve the dilemma of uneven development of core dimensions of service quality and significant regional differences. Secondly, both supply and demand sides should work together to build a collaborative innovation ecosystem for digital and service industries, and strengthen the empowerment efficiency of the digital economy. Thirdly, innovate the linkage mechanism between social security and market incentives, and unleash the multiplier effect of institutional environment.

With the continuous development of the economy, China has entered a new era of high-quality development. In the new stage of Chinese development, the main direction is to vigorously develop new quality productive forces driven by innovation based on local conditions. For a large economy like China, the real economy has always been the foundation of the entire economic framework, and industrial production is the main component in the real economy. Therefore, supporting and incentivizing industrial innovation has become a major gain for policy makers in the process of building an innovative country. The government has two methods to incentivize industrial innovation, including explicit and implicit fiscal expenditure of science and technology. Firstly, the government can support industrial innovation by directly injecting fiscal funds into industrial enterprises, which is known as explicit fiscal expenditure of science and technology. Secondly, in the process of market-oriented reforms, enterprises are the main body of innovation, so the expenditure methods for supporting industrial innovation more prefer to the ways of market. Therefore, implicit fiscal expenditures of science and technology in the form of R&D expense additional deduction and high-tech enterprise tax reduction are more applicable in more scenarios. Currently, in China’s Public Finance Model, fiscal expenditure of science and technology for promoting industrial innovation has presented a new feature as follows: the implicit fiscal expenditures of science and technology in the form of tax reductions have greatly exceeded the explicit fiscal expenditures of science and technology.

Direct effect of science and technology expenditure by the public sector in supporting industrial innovation activities is increasing industrial R&D innovation input because enterprises are the main body of innovation as described above. However, its actual economic value is bringing about a significant increase in innovation output. Using data from China’s industrial sectors and taking the number of invention patent applications as an indicator of innovation output, this paper estimates the impact of explicit fiscal expenditure in the form of direct government funding and implicit tax expenditure in the form of R&D expense additional deduction and high-tech enterprise tax reduction on industrial innovation output through various methods. This study finds strong and robust evidence that the two forms of implicit fiscal science and technology expenditure have no significant impact on industrial innovation output, and their combined effect is also not significant. Contrastively, R&D expense additional deduction and high-tech enterprise tax reduction have a significantly positive impact on internal R&D expenditure of industrial sectors, with elasticities of 22.38% and 7.9%, respectively. Therefore, industrial enterprises have a strong tendency of “strategic catering” innovation when facing tax reduction incentives. There is evidence that explicit fiscal science and technology expenditure has a significantly positive impact on industrial innovation output, but it is not robust. Since industrial enterprises do not significantly increase internal R&D expenditure due to government inputs fiscal science and technology funds, there is no incentive for industrial enterprises to engage in “strategic catering” innovative behaviors due to explicit fiscal science and technology expenditure.

Based on the theoretical analyses and empirical tests, to increase the value of the market in the allocation of innovation resources, the form of government support for industrial innovation should prefer to implicit tax subsidies rather than explicit fiscal expenditures. However, the “strategic catering” innovation behavior of market entities for tax reduction is not conducive to the effective utilization of innovation resources. This dilemma is a major obstacle to support industrial innovation by using the fiscal funds. Eliminate this bottleneck requires the integration and coordination among technology policies, industrial policies, and fiscal and tax policies in the future.

Key technology enterprises serve as the carriers, creators, and pioneering users of technologies for elderly care services. They provide the underlying technical framework for the innovation system of elderly care services and drive the evolution of this system. Most of the existing studies examine the innovation system of elderly care services from the perspectives of the macro-architecture and operational mechanisms, and there is insufficient attention paid to the important roles and diverse influences of key technology enterprises.

Based on an investigation of Tencent Silver Hair Technology Laboratory, we derive the evolutionary process of the elderly care service innovation system and the key technology enterprise effects. It reveals that the elderly care service innovation system consists of a technology niche system, a technology competition-cooperation system, and interaction flows between them. Within the technology niche system, key technology enterprises exert demand-driven effects and technology-driven effects. Internal and external linkage coordination effects are performed in the technology competition-cooperation system. During the interaction process, key technology enterprises exert a catalytic effect of connecting and consolidating the social and technological systems through demand linkage, resource integration, and rule improvement.

This study clarifies the evolutionary framework and key technology enterprise effects of the elderly care service innovation system from the perspective of actor interactions, deepens and develops socio-technical systems theory, and provides new research perspectives and practical implications for elderly care service innovation.

The “Double First-Class” initiative represents a major strategic endeavor in the reform of higher education in China. Central and local governments have provided substantial resources to support the selected universities, making it as a policy shock that is exogenous to universities. Using a Difference-in-Differences (DID) model, this study exams the “Double First-Class” impact on the quantity and quality of scientific research output of universities. The results indicate that the “Double First-Class” policy has achieved significant outcomes in both increasing the quantity and enhancing the quality of research output. Specifically, the “First-Class Discipline” policy during the first phase of the initiative demonstrated notable spatial spillover effects. In particular, regarding quality dimensions such as highly cited papers, top 1% leading papers, and Category Normalized Citation Impact (CNCI), non-selected universities surpassed the construction universities, thereby driving an overall improvement in the quality of scientific research output across the higher education system. Based on these findings, the study discusses issues such as the balance between construction-oriented and development-oriented thinking in the “Double First-Class” policy, the focus on institutions versus disciplines, and the tensions between homogenization and diversification. Looking ahead, it is recommended to better leverage the systemic driving role of “Double First-Class” in national higher education, and to vigorously advance the strategy of building a strong education system through mechanisms such as dynamic adjustment and categorized development.

This study aims to establish a framework for identifying the key security risks of generative AI in China, particularly in the context of the ongoing technological decoupling between China and the United States. It seeks to provide theoretical foundations and practical guidance for navigating the increasingly complex global technological environment. Through a layered analysis of foundational, technical, and application levels, this study constructs an analytical framework for evaluating the factors influencing the development of generative AI technology in China. Key technological characteristics such as computing power, data, and algorithms are considered, and expert knowledge is utilized to analyze and assess the identified risk factors. The identified security risks include limitations in computing power, data, and development tools at the foundational level; restrictions on model algorithms, computing architectures, and open-source platforms at the technical level; and challenges in market expansion and security thresholds at the application level. The study recommends establishing a coordinated national strategy to develop a response that ensures foundational autonomy, technical integration, and application-driven needs, thereby securing China's strategic position in global AI competition.

In an increasingly fractious geopolitical landscape, the utilization of tariffs has emerged as a pivotal instrument within the arsenal of international economic conflicts. This strategy, characterized by the uncertainty it injects into the policy framework, exerts a substantial impact on the resilience of corporate innovation. The present research delves into this complex interplay by employing a comprehensive dataset comprising A-share listed manufacturing firms in China. Employing a meticulous approach, the study employs textual analysis to construct key explanatory variables and harnesses the robustness of the difference-in-differences methodology to discern the causal efficacy of tariff shocks on the innovation resilience of firms under scrutiny. The empirical findings elucidate a compelling relationship: tariff shocks, often characterized by the imposition or withdrawal of protective tariffs, have been demonstrated to significantly enhance the innovative capacity of Chinese manufacturing firms. This outcome is in line with the widely accepted "adversity breeds innovation" hypothesis, which posits that exposure to adverse economic conditions can serve as a catalyst for increased innovation. Furthermore, the analysis delves deeper into the underlying mechanisms responsible for this phenomenon, revealing a dual-channel transmission effect. The first channel is rooted in macroeconomic policy uncertainty, which can spur firms to adapt and innovate as they navigate through a volatile regulatory environment. The second channel is grounded in internal corporate governance practices, indicating that effective governance structures can bolster resilience by facilitating informed decision-making and fostering a culture of innovation. An intriguing aspect of this analysis pertains to the role of directorate networks and managerial hubs within the firm. The study discerns that these elements serve as reinforcing agents, amplifying the resilience against the disruptions caused by tariff shocks. Conversely, it identifies that executive team faultlines can detract from this resilience, suggesting that the cohesive nature of leadership within the firm is a critical determinant of its ability to innovate in the face of such shocks. In conclusion, this investigation offers novel empirical insights into the intricate relationship between tariff shocks and innovation resilience within the context of Chinese manufacturing firms. These findings not only provide theoretical underpinnings for understanding the mechanisms through which economic stressors impact innovation but also offer practical guidance for policymakers navigating the treacherous waters of trade disputes. By elucidating the factors that can enhance or diminish resilience, the study contributes to the formulation of policy frameworks that are conducive to the cultivation of an innovative ecosystem within the broader context of international economic relations.

As a new infrastructure for data circulation and utilization, trustworthy data space is an important carrier for building a national integrated data market. The construction of trustworthy data space may become a key path to solve the contradiction between security and circulation in the market-oriented configuration of data elements, providing dual support of “security base” and “circulation hub” for digital transformation. The construction of a trustworthy data space can reconstruct data production relationships through technological integration, providing institutional guarantees and technological infrastructure for efficient allocation of data elements across organizations and industries, thereby achieving innovation in the paradigm of data sovereignty governance, and ultimately realizing the secure release of data value and high-quality development of the digital economy.

Trusted data space is a value exchange network with endogenous security attributes built on the basis of data space technology. By integrating privacy computing, blockchain, and smart contract technology, it forms a full cycle trusted guarantee system for the circulation of data elements. Its five foundations include the construction of key standard systems, breakthroughs in core technology innovation, optimization of infrastructure services, improvement of governance systems and mechanisms, and expansion of international cooperation networks. Its eight functions include identity authentication function, certificate traceability function, resource management function, access control function, privacy computing function, security sandbox function, contract execution function, and value evaluation function.

Based on the concept, foundation, function, and related theories of trusted data space, a three-dimensional model of “system technology ecology” for the construction of trusted data space has been constructed. By breaking the “trust security value” triple paradox in the flow of data elements, a systematic theoretical tool has been provided for the “trust controllability availability” of data elements in complex business environments. Research has found that: (1) The construction of trustworthy data space needs to follow the coordinated development logic of institutional, technological, and governance. Institutions can constrain the ethical boundaries of technological applications, technology can support the efficiency improvement of ecological collaboration, and ecology can practice feedback on the iterative optimization of institutional rules. The organic integration of the three has broken through the limitations of traditional single dimensional research and provided a systematic theoretical tool for the “trustworthiness controllability availability” of data elements in complex business environments; (2) The core technology focuses on the dual breakthrough of privacy security and trusted verification. The collaborative innovation of these two types of technologies can not only solve the “privacy utility” paradox in data sharing, but also establish a technical closed loop for ownership traceability through an undeniable proof mechanism; (3) At the level of governance logic, a hybrid framework of multi subject collaborative governance and risk immunity system needs to be established, and a three-dimensional governance model of “government regulation industry autonomy enterprise self-discipline” should be constructed to achieve the organic integration of technical barriers and institutional safeguards; (4) The realization of the value of a trustworthy data space depends on the iterative adaptation of technical tools and institutional rules.

The high-quality development of the trustworthy data space requires the construction of a system that promotes institutional innovation, technological breakthroughs, and governance experiments in a coordinated manner, in order to solve the triple problems of ownership ambiguity, technological constraints, and cross-border barriers in the circulation of data elements. Firstly, we need to accelerate the implementation of the “Measures for the Administration of Data Property Rights Registration” and clarify the rules and registration process for data ownership separation through legislation; Secondly, it is necessary to establish a national level data space technology laboratory to address the bottleneck problem of core technologies in trustworthy data space, integrate resources from universities, research institutes, and leading enterprises, and establish a national level laboratory; Thirdly, we need to explore the “regulatory sandbox” mechanism to promote cross-border data pilot projects. In response to the conflict of rules and regulatory uncertainty in cross-border data flow, it is recommended to carry out “regulatory sandbox” pilot projects in free trade zones.