生成式人工智能赋能公共图书馆未成年人服务创新研究

This paper presents a pilot study on integrating the use of Generative AI (GenAI) into maker‐based STEM learning through playful activities for children (aged 6‐12) and their families in libraries. Our preliminary analysis of log data from an LLM‐enabled chat system and video recordings shows how children and their caregivers used GenAI for various purposes, such as seeking assistance when encountering challenges, through which they could also learn about how to formulate prompts for GenAI. We also found some limitations of the LLM‐enabled chat system in providing situated feedback during maker‐based learning. Our study speaks to the conference theme, the Role of IS in the age of Human‐Centered AI, by shedding light on the potential of hands‐on activities based on designing and making for children's AI literacy as well as problem‐solving, creative thinking, and prompt engineering skills.

No abstract available

: If Generative Artificial Intelligence (GAI) is to define the future of education, it is essential that children are at the center of its design and development. Equity, sustainability and social justice must be essential operating principles at every stage of the AI lifecycle, not mere aspirations. Children should not be treated as invisible users, but recognized as rights holders whose protection must be reflected in every line of code, every design decision and every public policy of the digital ecosystem. Achieving this goal requires ethical and practical commitment from educators, technology developers and governments. Educational equity, child protection and social justice must be translated into concrete, achievable and measurable goals, ensuring that AI contributes to a safe, inclusive and child-center educational environment.

Generative AI (GenAI) platforms, such as ChatGPT, have gained popularity among the public due to their ease of access, use, and support of educational and creative activities. Despite these benefits, GenAI poses unique risks for families, such as lacking sufficient safeguards tailored to protect children under 13 years of age and not offering parental control features. This study explores how parents mediate their children's use of GenAI and the factors/processes that shape this mediation. Through analyzing semi-structured interviews with 12 families, we identified ways in which families used and mediated GenAI and factors that influenced parents'GenAI mediation strategies. We contextualize our findings with a modified model of family mediation strategies, drawing from previous family media and mediation frameworks. We provide insights for future research on Family-GenAI interactions and highlight the need for more robust protective measures on GenAI platforms for families.

Hate speech among children in primary and secondary schools represents a widespread and concerning phenomenon. It is unrealistic to expect young individuals to refrain from discriminatory behavior if they are regularly exposed to hateful rhetoric in their environment. This research focuses on the design and development of a web-based application for the detection and classification of hate speech in the Serbian language, with the broader goal of supporting efforts to reduce hate speech among school-aged children. Serbian, as a South Slavic language with limited natural language processing resources, poses additional challenges for computational analysis compared to globally dominant languages due to the scarcity of publicly available datasets. The implemented software tool leverages generative artificial intelligence, specifically state-of-the-art large language models including GPT-4o, Claude 4, Gemini 2.5-Pro, Llama 3, and DeepSeek-R1. The tool is intended for use in educational contexts in Serbia, providing functionality for the analysis of textual content prior to its publication on internet portals and social media platforms. Key features include the automatic detection of hate speech and the generation of revised, nonhateful versions of the text while preserving its original intent.

Publicly available Generative Artificial Intelligence (GenAI) tools are said to liberate students from the instrumental use of English and empower them to write creative texts to communicate with different communities. This paper reports on an undergraduate language-related service-learning subject in a Hong Kong tertiary institution. In the subject, students co-created digital stories with asylum-seeking children, in written and podcast formats, with the help of GenAI. The qualitative content analysis of semi-structured interviews with the students found that this experience expanded the students’ creative potential. Meanwhile, GenAI played a peripheral role in the story creation processes, in that the students exercised agency to use the tools and remained critical of the AI-generated content. This study argues that digital storytelling with GenAI, when used critically, promotes linguistic, digital and cultural awareness among ESL learners, offering them a third space to interact with culturally diverse communities in Hong Kong and giving them genuine ownership of English for creative and communicative purposes.

Public libraries in the U.S. are increasingly facing labor shortages, tight budgets, and overworked staff, creating a pressing need for conversational agents to assist patrons. The democratization of generative AI has empowered public service professionals to develop AI agents by leveraging large language models. To understand the needs of non-AI library professionals in creating their own conversational agents, we conducted semi-structured interviews with library professionals (n=11) across the U.S. Insights from these interviews informed the design of AgentBuilder, a prototype tool that enables non-AI experts to create conversational agents without coding skills. We then conducted think-aloud sessions and follow-up interviews to evaluate the prototype experience and identify the key evaluation criteria emphasized by library professionals (n=12) when developing conversational agents. Our findings highlight how these professionals perceive the prototype experience and reveal five essential evaluation criteria: interpreting user intent, faithful paraphrasing, proper alignment with authoritative sources, tailoring the tone of voice, and handling unknown answers effectively. These insights provide valuable guidance for designing AI-supported "end-user creation tools" in public service domains beyond libraries.

This paper explores the integration of generative artificial intelligence (AI) into adaptive learning systems to create customized learning support aids for students with disabilities, as traditional educational aids commonly fail to meet the diversity in the needs of these learners, and generative AI has shown innovative solutions that can offer real-time adaptation to content with personalized learning experiences. This paper introduces ALGA-Ed, a novel adaptive learning system utilizing generative AI that includes (i) a user profile module that captures cognitive, sensory, and behavioral profiles; (ii) generative AI models that create personalized multimodal content in real-time; (iii) an adaptive feedback mechanism using reinforcement learning to adjust content delivery based on real-time engagement metrics dynamically; and (iv) a real-time monitoring system that tracks progress and adapts learning pathways accordingly. The framework leverages heterogeneous datasets, including real and synthetic data, to effectively address diverse disability profiles. Pilot studies demonstrate the effectiveness of the framework in improving participation, retention, and learning outcomes for students with disabilities. This study enhances adaptive learning by encouraging inclusion via AI-driven tailoring and providing a basis for further advancements in AI-powered education catered for children with impairments. The source code for this research will be publicly available at https://github.com/aasimwadood/ALGA-Ed.

ABSTRACT AI is seeping into the fabric of our information environment as generative AI tools are increasingly used to search for and discover information. Despite their promise for improving efficiency, AI systems regularly produce errors (also known as ‘hallucinations’), which demonstrate that uncertainty is a feature rather than a bug of such systems. Despite this problem, we regularly hear stories about people who have mistakenly used false information provided by these tools in their communications and outputs – from lawyers’ reports to government hearings. There is wide agreement about the need for AI literacy to recognise how to use AI effectively and ethically but less consensus on how AI literacy is best achieved. A key component of many AI literacy frameworks is an understanding of how AI works. Using a case study of a critical AI literacy intervention in four Greater Sydney libraries, we argue that instead of learning only about how AI works, AI literacy might involve learning when, how, and why AI doesn’t work. The concept of socio-technical error and uncertainty is a useful heuristic for understanding AI – particularly in the context of information search and discovery, a primary practice in both public and academic libraries.

No abstract available

Abstract Graphic novels occupy an interesting intersection in libraries where they are both popular to read and popular to challenge or censor. Graphic novels are often the subject of great scrutiny, from their ability to channel and reflect inclusivity and empathy to their capacity to be taken out of context due to their highly visual nature. Interestingly, emerging technology developments such as Generative AI also pose surprising challenges for graphic novels, their creators, and the library communities that these works reside in. In the context of queries and challenges levelled against graphic novels in libraries, this article synthesizes a review of relevant literature and creative autoethnography in the form of reflective writing vignettes to explore the experiences of working in libraries and what these subjective stories reveal about graphic novel collections, their uses, and their perceived value. In doing so, graphic novels are critiqued for library engagement potential, while autoethnography is considered for its ability to elevate and analyze lived experience in research.

Employing a mixed postqualitative methods approach, this article examines the concept of reading imaginaries in public library events in three cases where children read with robots, dogs, and drag performers. Using the critical analytical tools of sociotechnical imaginaries and matters of care, we regard our cases as containers for resolving societal problems, and we explore how they contain societal expectations and imaginaries about reading and literacy in postdigital ecologies. The care practices observed in our case studies transcend digital technologies, encompassing broader and politically charged issues and the importance of situating the hype and fear surrounding emerging technologies, such as AI, within a more comprehensive and far-reaching framework. As concerns over privacy, data flows, and security increasingly extend to analog technologies such as books, curriculum content, and the embodiment of educators, mediated public discourse continues to play a significant role in shaping these dynamics. An example of this is how drag reading events become hot-button issues in a politicised and polarised school debate, while dogs and robots do not.

The article examines the prospects for integrating artificial intelligence (AI) technologies into the activities of Ukrainian libraries. The study analyzes contemporary approaches to implementing intelligent systems in library practice and evaluates their impact on the automation of routine processes, improvement of user services, optimization of workflows, and development of personalized information services. Particular attention is paid to successful case studies of Ukrainian libraries that have initiated AI-based projects, as well as to the financial, technical, ethical, and regulatory challenges associated with AI implementation. The paper outlines key areas of AI application in libraries, including automated cataloguing, semantic search, optical character recognition (OCR) for digitization of archival materials, data analytics for collection development, virtual assistants and chatbots for user support, and generative AI tools for creating bibliographies, annotations, multimedia content, and marketing materials. The role of AI in enhancing access to multilingual resources through machine translation and in supporting research and educational processes in higher education institutions is also emphasized. Special focus is placed on the emerging practices of Ukrainian academic, public, and research libraries, which have begun to integrate AI tools such as ChatGPT, Gemini, and other neural network-based systems into training activities, exhibitions, digital services, and methodological seminars. The study identifies перспективні directions for further development, including the creation of national AI-based library platforms, open APIs for integration with educational systems, generative AI for user support, predictive analytics for management decisions, and participation in international grant programs such as Horizon Europe and Erasmus+. The findings demonstrate that AI integration is a strategically necessary step in the digital transformation of libraries. Although financial, infrastructural, and кадрові limitations persist, Ukrainian libraries show openness to innovation and gradual adoption of intelligent technologies. Artificial intelligence does not replace librarians; rather, it expands their professional capabilities and strengthens the role of libraries as intellectual communication hubs within the digital ecosystem

There is a growing discussion on the relevance of libraries in the Generative AI era, where large language models are taking over the information space, and the role of libraries is affected in meeting the users' needs (Khan, 2024). This study explores the public perspective on the relevance of public libraries in the Generative AI era using sentimental analysis (quantitative approach) and thematic analysis (qualitative approach). The data was collected from Reddit, 1282 cleaned, stemmed, and used for the analysis. Sentimental analysis was adopted to understand the public positive, negative, and neutral sentiments about the usefulness of public libraries amidst proliferation of Generative AI technologies. The findings from most comments indicated positive sentiments, attitudes, and opinions, showing that more believe that public libraries are still relevant, and a few stated that it is irrelevant in the Generative AI era. Thematic analysis of the frequency comments revealed that public libraries still provide core services, health and counseling assistance, recreational commons, and financial literacy to users. Besides contributing to artificial intelligence literature on public libraries, the findings of this study will inform the preparation of pre-service librarians interested in public librarianship as they take courses and begin their professional journey as public library practitioners. Universities and colleges that train librarians for public library roles would also be impacted and guided through this study's findings to make informed decisions on the skill set expected of public librarians in the Generative AI era as they train them for public library service roles.

No abstract available

Addressing the need for effective LIS training in handling patron crises, this alternative event showcases the potential of integrating Virtual Reality (VR), Generative AI (GenAI), and serious games. We present a 90‐minute interactive session demonstrating a Unity‐based serious game designed for de‐escalation skill development for LIS students and professionals. Using Meta Quest 3 VR headsets, attendees will navigate a virtual library, learn techniques through gamification, and apply them by interacting with a dynamic GenAI bot (“Arthur”) simulating a patron in distress. The event combines presentation and hands‐on demonstration, exploring topics like serious VR games, GenAI roleplaying, gamification, and VR for high‐stress simulation. Participants will understand the application of these technologies in LIS training, experience a specific use case, and contribute to brainstorming future possibilities, targeting researchers, professionals, and students interested in technology‐enhanced learning.

Abstract:Decisions about the use of generative artificial intelligence (GenAI) within the field of library and information science (LIS) will have serious impacts on our patrons, our colleagues, and ourselves. This paper asserts the importance of LIS practitioners and educators for modeling wisdom, especially regarding the use of GenAI. I amplify perspectives from historically marginalized and equity-deserving groups as a much-needed counterbalance to majority opinions on GenAI and ethics by introducing diverse concepts of "wisdom" and ethical guidelines on the use of AI, then briefly discussing how these perspectives might inform the use of GenAI within librarianship. I next survey and discuss examples in the current literature of how libraries (and librarians) have begun using GenAI. I analyze the extent to which these examples consider either wisdom or ethics in their use of this technology. I then propose the use of character ethics (sometimes referred to as virtue ethics) as an approach to both embodying and modeling the virtue of wisdom within the field of LIS. Finally, I conclude with recommendations for the field on how to use a framework of character ethics so that we hold ourselves and others accountable to upholding wisdom as a virtue of the field.

AI companions are increasingly popular among teenagers, yet current platforms lack safeguards to address developmental risks and harmful normalization. Despite growing concerns, little is known about how parents and developmental psychology experts assess these interactions or what protections they consider necessary. We conducted 26 semi structured interviews with parents and experts, who reviewed real world youth GenAI companion conversation snippets. We found that stakeholders assessed risks contextually, attending to factors such as youth maturity, AI character age, and how AI characters modeled values and norms. We also identified distinct logics of assessment: parents flagged single events, such as a mention of suicide or flirtation, as high risk, whereas experts looked for patterns over time, such as repeated references to self harm or sustained dependence. Both groups proposed interventions, with parents favoring broader oversight and experts preferring cautious, crisis-only escalation paired with youth facing safeguards. These findings provide directions for embedding safety into AI companion design.

As Artificial Intelligence (AI) systems increasingly mediate the digital lives of youth, ensuring their online safety is an urgent priority. While HCI/CSCW scholars have explored participatory and ethical dimensions of AI, a critical gap remains: how to systematically integrate youth online safety experts, advocates, and young users themselves across the AI development lifecycle. This is especially imperative for Generative AI (GenAI), which introduces new risks such as dis/misinformation, deepfakes, and manipulative personalization that are difficult to anticipate or govern due to GenAI’s underlying development paradigm (e.g., reliance on large training datasets designed for general purposes and result in unpredictable outputs). This workshop brings together interdisciplinary researchers, practitioners, and advocates to explore how expertise from child-centered domains can be embedded into GenAI development. We will collaboratively define the role of expertise in the AI lifecycle, explore how to operationalize multistakeholder collaboration, and identify future directions for building participatory-driven tools and methods. While GenAI is a core focus, the workshop welcomes broader AI and youth online safety perspectives. This workshop aims to develop actionable strategies for building AI systems that are not only innovative but also inclusive, ethical, and responsive to the needs and rights of youth.

As Generative Artificial Intelligence (GenAI) becomes an increasingly significant part of young people’s lives, educators worry about its impact on learning and attainment. However, understanding this impact requires more than simply studying young people’s behaviours or soliciting their opinions. It is essential to involve them actively as co-researchers, allowing their unique perspectives to shape the conversation around GenAI in schools. This project moves beyond seeing young people as research subjects, positioning them instead as co-designers, co-researchers, and potential influencers of national policy on GenAI in education. We recruited eight young people (aged 16–18) from three Scottish high schools to serve as Young People Co-Researchers (YPCR). Together, we explored their perspectives on GenAI at school, including their current usage, views on appropriate tasks for AI, and opinions on teachers’ use of AI. The YPCR organised and conducted semi-structured focus groups with 50 peers and collaboratively analysed the findings with adult researchers. Our results show that young people are cautiously optimistic about GenAI’s potential for learning and do not support outright bans in schools. They clearly distinguish between AI use for learning and in assessments, often expressing confusion over current policies and wishing for clearer guidance. The YPCR stated a strong desire to deepen their understanding of AI’s advantages and risks and for schools to teach responsible, effective use. Their insights are valuable for national policy development and for AI literacy initiatives.

No abstract available

: Against the backdrop of the deepening “Double Reduction” policy, public libraries have been entrusted with new after-school educational responsibilities. Yet empirical surveys reveal that primary and secondary school students remain overwhelmingly test-oriented (65 %) and confined to supplementary workbooks (62 %). Sixty percent read intensively for less than 30 minutes a day, a pattern that significantly constrains social-emotional learning (SEL). Integrating Piaget’s cognitive-development theory with the CASEL SEL framework, we employed stratified random sampling to administer questionnaires and rating scales to 500 pupils across six primary and secondary schools in Tai’an. Results show that literary reading, science reading and intrinsic motivation positively predict SEL (β = 0.25, 0.18, 0.31, p < 0.01), whereas test-oriented motivation exerts a negative effect (β = –0.14, p < 0.05). Grounded in these findings, we propose a “cognitive–affective” dual-pathway reading-promotion strategy: (1) a staged and leveled reading system (concrete reading kits for primary students and critical interdisciplinary booklists for junior-high students); (2) SEL-themed bibliographies coupled with emotion-driven experiential activities; (3) a three-way “library–school–family” linkage for the “3:30 p.m. classroom”; and (4) digital empowerment via intelligent recommendation and virtual communities. The study offers an evidence-based framework for targeted public-library interventions under the “Double Reduction” policy and contributes both theoretical insight and practical guidance for fostering the synergistic development of cognitive advancement and social-emotional competence in primary and secondary school students.

: This study investigates the application of Generative AI (GenAI) in human-AI collaborative reading promotion within Chinese university libraries, addressing the paradigm shift from 'resource pushing' to 'demand response.' Utilizing a case study methodology, it analyzes practices from libraries including Harbin Institute of Physical Education, Shijiazhuang University, and Shenyang University of Technology to examine AI's role across the entire activity chain: planning, resource provision, service implementation, and effect evaluation. The research finds that mature collaboration models exhibit a three-dimensional structure of 'technology empowerment, librarian leadership, and reader co-creation,' where the organic integration of intelligent tools and professional expertise enhances service coverage and engagement. However, the study also identifies persistent challenges, including technical immaturity and data security risks. It concludes by proposing strategic countermeasures—such as optimized technology selection, specialized talent cultivation, and robust system construction—to provide a practical reference for advancing intelligent reading promotion in the academic library sector.

Immersive learning has been recognized as a promising paradigm for enhancing educational experiences through the integration of VR. We propose an architecture for intelligent tutoring in immersive VR environments that employs LLM-based non-playable characters. Key system capabilities are identified, including natural language understanding, real-time adaptive dialogue, and multimodal interaction through hand tracking, gaze detection, and haptic feedback. The system synchronizes speech output with NPC animations, enhancing both interactional realism and cognitive immersion. This design demonstrates that AI-driven VR interactions can significantly improve learner engagement. System performance was generally stable; however, minor latency was observed during speech processing, indicating areas for technical refinement. Overall, this research highlights the transformative potential of VR in education and emphasizes the importance of ongoing optimization to maximize its effectiveness in immersive learning contexts.

Agentic artificial intelligence (AI) systems that act as autonomous agents are rapidly evolving due to the explosion of large language models (LLMs) [1]. This paper presents a comprehensive analysis of major approaches in the field of Agentic AI, including the large visual language model (LVLM), the React and Plan-and-Execute agent architectures, the smolagents library with a “coding-first” orientation, tool invocation techniques that extend the capabilities of LLMs, the visual Agentic AI model with multi-agent coordination capabilities, as well as scientific agent systems such as AI Scientist and the AgentRxiv collaboration platform. We analyze the characteristics of each approach, including representation models, advantages, limitations, and integration capabilities, for building intelligent agent systems that aim for AGI. The paper proposes an integrated scheme that leverages achievements from multimodal capabilities, multistep reasoning and planning, multi-agent coordination, and research automation, laying the foundation for a new generation of autonomous AI agents. Finally, we discuss the potential applications of Agentic AI in the context of Vietnam, especially in education, scientific research, and technology development, and provide recommendations for domestic developers and researchers.

This article is based on the data from the UNESCO 2024 Global Education Monitoring Report, which indicates that 73% of multi-ethnic countries face issues of insufficient cultural representation in curricula. It systematically elaborates on the theoretical framework of Culturally Responsive Educational Technology (CRET). By analyzing the SBAC assessment standard system developed by the Smarter Balanced Assessment Consortium in the United States, and examining typical cases such as the bilingual teaching platform in China's "Three Regions and Three Prefectures" (Tibet, Tibetan areas in four provinces, the southern Xinjiang four prefectures, Sichuan's Liangshan Prefecture, Yunnan's Nujiang Prefecture, Gansu's Linxia Prefecture, and the production and construction corps located in the southern Xinjiang four prefectures), as well as conducting a comparative study of educational technology policies across 28 countries, this paper proposes a CRET three-dimensional model that includes cultural adaptation algorithms, multimodal resource libraries, and intelligent diagnostic systems. Research shows that classrooms adopting CRET can increase minority students' participation by 41% and improve knowledge retention by 33%. These groundbreaking advancements not only validate the feasibility of technology-enabled culturally responsive teaching but also establish a technological pathway to address the phenomenon of "cultural poverty in education", providing a replicable solution for global digital education governance. The future educational ecosystem will be a collaborative intelligent system between humans and machines, preserving the advancement of technology while guarding the unique creativity and empathy inherent to humanity.

Generative artificial intelligence (AI) heralds a transformative era for libraries, redefining their role as knowledge hubs through innovative content creation. This conceptual paper explores how generative AI can revolutionize library services by enhancing personalization, efficiency, and innovation. Focusing on theoretical insights without empirical data, it examines opportunities, challenges, and future trends, drawing from library science and AI ethics literature. Key opportunities include AI-driven personalized recommendations, enabling tailored reading lists, and virtual storytelling, fostering inclusive user engagement (Cox 425). Operational efficiencies, such as automated metadata generation, streamline cataloging, enhancing resource accessibility (Hadi et al. 5). Challenges encompass ethical dilemmas like algorithmic bias, which risks marginalizing voices, and privacy concerns tied to data-intensive models (Bender et al. 612). Implementation barriers, including integration costs and staff training needs, further complicate adoption (Massaro 45). Future directions propose multimodal AI and federated learning to create privacy-conscious, immersive services, alongside policy frameworks for ethical integration (UNESCO 12). The paper advocates for cautious adoption, emphasizing transparency and equity to align AI with libraries’ democratic mission. Results highlight generative AI’s potential to transform libraries into dynamic, co-creative ecosystems while underscoring the need for ethical oversight to mitigate societal risks like misinformation. This framework offers actionable insights for librarians, policymakers, and researchers, positioning libraries as leaders in the AI-driven information landscape.