The deployment of artificial intelligence (AI) in the public sector raises concerns about accountability gaps, yet its impact on bureaucratic felt accountability remains unclear. This study investigates how AI assistance influences felt accountability (including expected accountability, forum legitimacy, and forum expertise) among street-level bureaucrats, with a focus on gender representation and discretion in citizen interactions. Through a 2 & times; 2 vignette experiment with 399 respondents, we find that AI assistance reduces felt accountability, especially expected accountability, with reduced discretion serving as a key mechanism. Unexpectedly, gender representation worsens this decline. This study bridges theoretical insights to provide causal explanations underlying accountability in the AI era.

This study investigates nuanced public perceptions of artificial intelligence (AI) in public sector decision-making, extending existing scholarship by examining human-AI collaboration and its various configurations across diverse scenarios. Using a data-centric categorization of decision scenarios and drawing on two experiments in China: a vignette experiment (Study 1; n = 611) and a conjoint experiment (Study 2; n = 894), this study yields two main findings. First, while human decision-making is perceived as more acceptable than AI, AI is seen as more efficient, and human decision-making is viewed as fairer. Notably, human-AI collaboration tends to achieve high public acceptance by leveraging the complementary strengths of both humans and AI across different decision scenarios. Second, the specific configuration of human-AI collaboration significantly influences public perceptions, with collaborative approach, outcome explanation, responsibility attribution, and stakeholder involvement playing critical roles. This study not only provides a data-centric perspective on decision scenarios but also advances our understanding of human-AI collaboration in decision-making, offering practical insights for optimizing its configurations.

Although transparency is often proposed as a remedy for legitimacy challenges in AI-assisted decision-making, empirical evidence remains limited, particularly given the complexity of transparency types and long-standing debates about their effects. This study distinguishes between transparency in rationale and process, and examines their influence on citizens' perceived legitimacy. Drawing on survey experiments on public recruitment conducted in China (N = 934) and the United States (N = 947), the results show that AI disclosure consistently reduces perceived legitimacy, whereas AI explanation has no significant effect in either country. Bureaucratic rejection of AI recommendations reliably undermines perceived legitimacy in both contexts, while bureaucratic adoption decreases perceived legitimacy in the United States but has no significant effect in China. We also observe a tentative pattern suggesting that AI explanation may shape legitimacy judgments regarding bureaucrats' adoption or rejection of AI recommendations. This study develops a two-dimensional conceptual framework of transparency in the AI context and provides cross-national evidence on its relationship with perceivedlegitimacy. The study calls for moving beyond generalized assumptions about a singular transparency-legitimacy relationship and toward more nuanced, context sensitive transparency strategies.

This seminar series is jointly organized by [***World Salon***](https://www.world-salon.com/) and [***Risk Science***](https://www.keaipublishing.com/en/journals/risk-sciences/). Professor Runhuan Feng is a Chair Professor in the School of Economics and Management at Tsinghua University. He is a Fellow of the Society of Actuaries and a Chartered Enterprise Risk Analyst. Prior to joining Tsinghua, he was a tenured Professor, the State Farm Companies Foundation Endowed Professor at the University of Illinois at Urbana-Champaign, the Faculty Lead for the Finance and Insurance sector at the Discovery Partnership Institute of the University of Illinois System. He is currently serving as the Executive Editor-in-Chief for Risk Sciences. Professor Volker H. Schmidt's current research, conceptualized within the framework of "Global Modernity", focuses on global social change and its practical as well as normative implications. His latest work, a book reconstructing the semantic evolution of the concept of society and entitled "From Societas to World Society. Genealogy of a Concept", appeared in May 2025. Robert D. Atkinson. As founder and president of the Information Technology and Innovation Foundation (ITIF), recognized as the world’s top think tank for science and technology policy, Robert D. Atkinson leads a prolific team of policy analysts and fellows that is successfully shaping the debate and setting the agenda on a host of critical issues at the intersection of technological innovation and public policy. Professor Zheng LIANG now serves as the professor of the School of Public Policy and Management, Tsinghua University, as well as the research fellow and deputy director of China Institute for Science Technology Policy at Tsinghua University (CISTP), which is jointly established by Ministry of Science and Technology of China and Tsinghua University, mainly focusing on the studies of S T policy and the national strategy of S T development.

The adoption of artificial intelligence (AI) by governments promises to enhance public governance, yet its implementation pace varies significantly across the globe. While existing research has identified various influencing factors, the deep-seated impact of national culture has not been fully explored. To address this gap, our study quantitatively investigates the influence of national culture on government AI adoption, examining government effectiveness as a mediating mechanism. Leveraging a panel dataset of 244 country-year observations from 2020 to 2023, we employ a random effects model analyzing Hofstede’s National Culture Index against the Government AI Readiness Index from Oxford Insights. The results reveal that masculinity and uncertainty avoidance are negatively associated with government AI adoption, whereas long-term orientation shows a positive association. Mediation analysis confirms that government effectiveness is a key pathway through which these cultural dimensions exert their influence. Moreover, heterogeneity analysis indicates that this relationship is contingent on national contexts, including GDP per capita, unemployment rates, and internet penetration. This study contributes by empirically demonstrating that national culture is a foundational determinant of government AI adoption, offering nuanced insights for developing culturally-adapted AI strategies.

Privacy is a critical concern in the AI era, yet its impact remains underexplored in public service delivery. This study develops a moderated mediation model, where privacy utility mediates the relationship between decision-makers (AI, human-AI collaboration, humans) and perceived legitimacy, with AI knowledge nonlinearly moderating this relationship. A vignette experiment was used to investigate causality, with 512 participants recruited from China. It finds that AI's privacy utility is lower than human services due to higher privacy risks without additional privacy benefits, while human-AI collaboration mitigates these risks. The effect of AI on legitimacy is fully mediated by privacy utility, with subjective AI knowledge moderating the relationship in a U-shaped pattern. These findings provide a nuanced understanding of privacy in the AI context, linking it to legitimacy and underscoring the importance of AI knowledge.

This paper demonstrates how platform service solutions are solidified into enterprise standards through proprietary algorithms provided by software development enterprises. These algorithm standards permeate social life but remain black boxes. By examining three cases in China, the authors show that enterprises’ algorithm standards coordinate and control various elements of service processes at the micro level. Their control extends beyond the enterprise into a wide range of fields of society where their software is applied, exerting control over the human social order. Thereby, proprietary algorithm standards of enterprises blur the boundaries between technical standards and regulations, as well as between technical standards and laws, producing a series of social problems. The authors argue that standardization work by standard developing organizations is an important channel for addressing problems arising from corporate algorithm standards and, when necessary, needs to be combined with legislation enacted by the state.

Although chatbots are widely adopted in public sectors worldwide, citizen approval remains suboptimal. This study adopts a public service design approach to develop a detailed service blueprint for chatbot, focusing on how specific design components influence public preferences. Through a conjoint experiment with 859 respondents in China, the study finds that citizens prefer online chatbots using human-like styles and formal official language, as well as those that provide extended content, recovery strategies, and feedback channels. Moreover, Public preference varies across different service contexts and individual characteristics. This study contributes a comprehensive service design framework for chatbots and address several controversies on design components, offering actionable insights for improving chatbot design and realizing their potential in public service.

Artificial intelligence (AI) is increasingly employed to support decision-making in the public sector, yet it exacerbates the “problem of many hands and eyes” in public accountability. Using a vignette experiment with a 3 × 3 factorial design, we investigated how the configuration of account givers and holders in AI accountability affects perceived legitimacy among citizens. Our findings indicate that human agents are perceived as more legitimate account holders than AI systems, where bureaucratic operators outperform technological developers. Additionally, citizens are perceived as more legitimate account holders than political leaders and experts. We also explored the interactive effects between account givers and holders. This study responds to calls for empirical evidence on AI accountability, expands the conceptual framework of accountability in the AI era, and highlights the crucial roles of bureaucrats and citizens.

This study establishes a theoretical framework linking organized R&D (ORD) and mission-oriented innovation (MOI) through a collective action lens. MOI performance is evaluated using three key indicators: academic publications, Science and Technology Awards (STA), and granted patents. ORD dimensions are operationalized through research teams, human resources, academic milieu, and public funding. Leveraging survey data and archival records from 23 Chinese universities, we employ baseline regressions and structural equation modeling (SEM) to elucidate ORD’s influence pathways on MOI performance. Results indicate that research teams serve as significant mediators linking public funding, academic milieu, and human resources to MOI outcomes as well as the heterogeneous roles of ORD determinants in MOI performance. This study specifically highlights how the scale and allocation mechanisms of public funding more actively facilitate MOI performance outcomes through ORD. By integrating macro-micro connections between MOI and ORD, this research provides policymakers with targeted and actionable recommendations for enhancing MOI in higher education institutions.

As artificial intelligence (AI) becomes increasingly influential, governments worldwide are developing policies to manage its multifaceted impact across sectors. This study employs the structural topic model (STM) to analyze 139 AI policies from China, the European Union (EU), and the United States (US), three key actors in global AI governance. The analysis identifies 13 primary topics within AI policy frameworks, which are categorized into “research and application” (e.g., talent education, industrial application), “social impact” (e.g., technological risk, human rights), and “government role” (e.g., government responsibility, management agency). Notably, “government role” receives the most attention, while “social impact” is the least emphasized. The findings reveal that China prioritizes “research and application,” the EU emphasizes “social impact,” and the US focuses on “government role,” while all three demonstrate a growing emphasis on institutional systems, human rights, and scientific research. This study provides a comprehensive policy framework for AI governance, highlights the strategic priorities of China, the EU, and the US, and introduces an innovative method for policy text analysis. Moreover, it underscores the need for AI governance to balance industry development with ethical imperatives, foster comprehensive technological ecosystems, and prioritize public participation and international cooperation.

Public participation is crucial for the governance of artificial intelligence (AI). However, the public is typically portrayed as a passive recipient in practice, and little is known about their expectations, assumptions and knowledge about AI. Based on the theoretical lens of technological frames, and using Latent Dirichlet Allocation (LDA) and content coding on 114,393 relevant comments, the article explores public perceptions of the nature of AI, AI strategy and AI in use, and further compares the discourse before and after the launch of ChatGPT. The findings show that ChatGPT amplifies public enthusiasm and expectations of AI, as well as concerns and fears, and draws attention to national competition. However, public discourses are generally conflicting and preliminary, characterized by abstract and grandiose narratives, posing challenges in reaching consensus and generating practical insights. Therefore, this analysis raises serious concerns over public participation in AI governance, as the public not only faces limited channels for substantive involvement but also struggles to articulate effective discourses in the public sphere.

Despite the recognized potential of artificial intelligence (AI) to improve governance, a significant divide in AI adoption exists among governments globally. However, little is known about the underlying causes behind the divide, hindering effective strategies to bridge it. Drawing on the AI capability concept and the Technology-Organization-Environment (TOE) framework, this study employs Explainable Artificial Intelligence (XAI) models to analyze the multifaceted factors influencing AI adoption by governments worldwide. The results underscore the critical roles of internet security and internet usage within the technological dimension, regulatory quality, government effectiveness, government expenditure, rule of law, and corruption control within the organizational dimension, and globalization, median age and GDP per capita within the environmental dimension. Notably, our analysis explores the intricate effects of these variables on government AI adoption, identifying inflection points where their impacts undergo significant shifts in magnitude and direction. This nuanced exploration provides a comprehensive understanding of government AI adoption globally and illustrates targeted strategies for governments to bridge the AI adoption divide, making theoretical, methodological and practical implications.

While many studies have investigated the impact of artificial intelligence (AI) deployment in the public sector on government-citizen interactions, findings remain controversial due to the technical complexity and contextual diversity. This study distinguishes between rule-driven and learning-driven AI and explores their impact as automated respondents on citizen-initiated contact, an important scenario for public participation with initiative. Based on a conjoint experiment with 763 participations (4578 observations), this study suggests that AI deployments enormously reduce the likelihood of citizen-initiated contact compared to human response, with learning-driven AI having a higher negative effect than rule-driven AI. In addition, the causal effects of respondent image, contact channel, contact purpose, and matter attributes on citizen-initiated contact, as well as their moderating effects, are explored. These findings make theoretical implications and calls for public participation in the roaring AI deployment in the public sector.

In the field of numerical control machining, tool alignment technology is a key link to ensure machining accuracy and quality. Tool alignment refers to determining the correct position of the tool relative to the workpiece, and its accuracy directly affects the precision of part machining. With the development of precision machining technology, the research and application of cutting technology are increasingly valued. Tool alignment methods are mainly divided into two categories: contact and non-contact. The contact type tool alignment method relies on direct contact between the tool and the workpiece or tool alignment instrument to measure the position. Among them, the trial cutting method is a traditional contact type tool alignment method that determines the tool position through actual cutting, which is intuitive but inefficient. The contact type tool presetter uses specialized equipment to improve the accuracy and efficiency of tool presetting through contact measurement. The non-contact tool alignment method does not rely on physical contact, while the image method uses image recognition technology to determine the tool position, making it suitable for high-precision applications. The laser diffraction method and the laser direct method use laser technology for non-contact measurement. The laser diffraction method determines the position of the tool by analyzing the diffraction mode of the laser beam, while the laser direct method directly measures the distance between the laser and the tool. This article mainly introduces the classification of tool alignment, commonly used knife alignment methods and common tool alignment devices, as well as the development status of international tool alignment instrument products.

Technological empowerment has facilitated the development of cities, which have progressed from pre-industrial to industrial to information-based and are currently transitioning towards the advanced stage of smart cities. The evolution and transformation of cities are fuelled by technology, which serves as a key driver. Disruptive technologies are radically scientific innovations that dramatically change the way consumers, businesses, and industries operate by destroying the value of existing technical competencies, thereby providing organisations with the capability or technical foundation to alter their business environments. To ensure that a city has a clear understanding of its smart city development direction, it is crucial to establish a scientifically valid and reliable evaluation index and method to analyse and recognise the disruptive technologies closely related to industrial development, transformation, and competitiveness in smart cities. However, there is a paucity of study on this topic. This paper addresses this research gap by developing a framework for disruptive technology identification and evaluation for smart cities using an entropy weight method and analytic hierarchy process. The evaluation index system contains 5 primary indicators and 11 secondary indicators according to the connotation of disruptive technologies in smart cities. The feasibility and effectiveness of the proposed framework are verified in the field of information science. This study provides technical knowledge and theoretical support for the evaluation and construction of smart cities.

0引言 自工业革命以来,如何建构完整的技术现代性认知体系、评价体系与约束体系成为了人们永恒追求的时代"主题",技术的"风险分配逻辑"便逐渐代替"福利分配逻辑"成为了技术治理关注的焦点.每一项划时代意义的技术创新都会带来技术治理体系的变迁,甚至影响整个社会治理形态的演进,其中,以人工智能最为典型.不同于其他新兴技术,人工智能具有技术内核的隐秘性、技术形式的拟人性、应用场景的跨域性、利益主体的交织性、技术风险的多维性、社会影响的复杂性等属性,使得人工智能治理成为了一项复杂工程.在经验世界中,不同主体基于不同的价值、利益和专业视角试图清晰刻画人工智能治理的轮廓,而认知逻辑中的框架前提差异则直接决定了人工智能治理的强度、维度和形态的异质性.

针对当前元宇宙的发展态势和理论研究,本文重新界定了元宇宙的概念内涵.元宇宙的样态本源可追溯至数字化数据,其技术源头则是各类数字化数据生成技术.根据数字样态的作用场域和虚实世界互动过程的主体跨域性等两个拓扑维度,可将元宇宙划分为四类阶段性形态:纯数字化世界、数字孪生世界、虚实互构世界和虚实协同世界.将元宇宙置于数字样态的整体演进历程中发现其四种发展形态之间的演化机理,即,纯数字化世界与传统网络空间紧密相关而成为元宇宙创新的首选之地,数字孪生世界的元宇宙侧重于工业实在世界,虚实互构世界综合了纯数字化世界和数字孪生世界的结构与功能,虚实协同世界则是人类数字化观念革新的驱使下虚实互构世界的高级形态.本文澄清了元宇宙的概念内涵、形态发展与演变机理,为后续的产业政策制定和治理体系建构提供了学理基础.

Many countries have enacted AI policies in various fields as artificial intelligence (AI) has emerged one of the most prominent double-edged swords among destructive technologies. Applying the novel technique of Structural Topic Model (STM), this study analyses 139 AI policies of China, the European Union (EU) and the United States (US), three major players in global AI governance. The analysis identifies 13 major topics in AI policies, which can be grouped into 3 categories based on their relevance: “research and application”, “social impact” and “government role”, with the “government role” receiving the most mentions and “social impact” the least. Further, it demonstrates the comparability of topic prevalence in China, the EU, and the US over time, with China prioritizing “research and application”, the EU stressing “social impact”, and the US focusing on “government role”. A consistent trend, however, is the growing emphasis on institutions, human rights and scientific research. This analysis therefore provides an overview of the policy frameworks regarding AI, making theoretical and practical contributions.