The contemporary library landscape is undergoing a fundamental transformation, shifting from a traditional custodial role—characterized by the passive preservation and distribution of information—to a dynamic model of proactive knowledge facilitation. This evolution is driven primarily by the integration of Artificial Intelligence (AI) platforms, encompassing machine learning (ML), natural language processing (NLP), and generative AI (GenAI). Unlike conventional search tools, which rely on rigid keyword matching and Boolean logic, generative AI, powered by large-scale language models and advanced NLP, enables dynamic, context-sensitive interactions between users and information systems.[1] This technological leap necessitates deep shifts in organizational paradigms, moving libraries toward active information delivery models that anticipate user needs and provide tailored information before it is explicitly requested.[1, 2]
The integration of these technologies into the library ecosystem is emerging as a revolutionary trend, changing traditional library services and functions as they adapt to the evolving digital world.[2] This transformation is not merely a change in toolsets but a complete reimagining of the library as a cognitive infrastructure—a proactive guide through complex information ecosystems that contextualizes resources to individual academic needs and co-creates meaningful learning experiences.[1] By utilizing artificially intelligent algorithms and other forms of innovative technology to automate procedures, enhance decision-making, and offer customized assistance, libraries are maintaining their relevance in a rapidly evolving digital world.[2, 3]
Technological Foundations and the Platformization of Library Services
The architecture of the modern library is increasingly defined by the integration of AI into integrated library systems (ILS) and library services platforms (LSPs).[4] These platforms have evolved from simple database managers into complex AI ecosystems that facilitate every stage of the information lifecycle, from resource acquisition and cataloging to discovery and long-term preservation.[5]
One of the most prominent examples of this platformization is the Ex Libris Alma library management system, which has unified disparate processes into a single cloud-based platform.[5] By leveraging artificial intelligence, Alma allows libraries to remove unnecessary silos and streamline processes through a unified library experience.[5] The platform’s capability to manage all resources—print, electronic, and unique digital materials—within a single environment provides the foundation for data-driven insights through tools like Alma Analytics.[5, 6] This unification is critical as the volume of scholarly content explodes, placing growing demands on library staff who must manage vast amounts of data quickly and adapt to changing standards and user expectations.[6]
Beyond the core LSP, a suite of specialized AI-powered vendors has emerged to address specific library functions. Third Iron, for instance, employs AI in the form of an Expert System to build direct links to articles, streamlining the content connection between patrons and resources.[7] LIBNOVA assists organizations in digital preservation by applying AI techniques for the automation of ingesting processes, content analysis of digital objects, and automated classification of files.[7] In the clinical domain, DynaMed, a subsidiary of EBSCO Information Services, has partnered with IBM’s Watson Assistant to offer AI-powered natural language queries for healthcare providers and pharmacists.[7] Similarly, Evidence Partners offers DistillerSR, which uses AI to automatically classify and deduplicate references and literature reviews, significantly enhancing the efficiency of systematic reviews.[7]
| Platform / Vendor | Core AI Technology | Primary Library Application |
|---|---|---|
| Ex Libris Alma | LLM (AI Metadata Assistant) | Automated cataloging and metadata enrichment [5, 6] |
| Third Iron | Expert System | Streamlined linking between patrons and resources [7] |
| LIBNOVA | ML (Automated Classification) | Digital preservation and automated file ingest [7] |
| DynaMed (with Watson) | NLP (Watson Assistant) | Natural language queries for clinical topics [7] |
| Evidence Partners | ML (Classification Manager) | Systematic literature review screening and deduplication [7] |
| Clarifai | Computer Vision | Automated tagging and categorization of digital images [8] |
Automated Metadata and the Efficiency Frontier in Technical Services
The technical services of a library, specifically cataloging and metadata creation, have traditionally been manual efforts that required significant expertise and time. However, as scholarly output has scaled, manual cataloging of thousands of records has become unfeasible.[6] AI-driven cataloging tools are now expediting this task through the automated generation of metadata when new documents arrive.[9, 10] These tools employ NLP and machine learning algorithms to analyze content, identify appropriate subject headings, and create consistent metadata, thereby improving the discoverability of library materials.[9, 10]
A significant development in this area is the AI Metadata Assistant, part of the Alma features released in early 2025.[5] This tool uses a Large Language Model to process information about a library resource and suggest relevant metadata—such as subject headings, abstracts, and language identification—to the cataloger.[5, 6] This approach does not replace the human expert but rather complements them; the “human-in-the-loop” model ensures that librarians evaluate the results, phrase the prompts, and add depth and accuracy where the AI might lack nuanced judgment.[6] This automation of repetitive tasks allows library professionals to reallocate their expertise to higher-order tasks, such as supporting research and teaching.[1, 6]
The impact of these tools on operational efficiency (OE) is profound. Quantitative surveys of Library and Information Science (LIS) professionals have revealed a strong positive correlation between AI integration and operational efficiency, with one study reporting a correlation coefficient of r=0.902.[3] By automating routine inquiries and technical services, libraries can serve a larger user base without a proportional increase in staff workload.[11] This is particularly crucial for academic libraries that are expected to do more with limited resources.[11]
Impact on Discovery and Knowledge Retrieval
The transformation of metadata workflows has a direct ripple effect on resource discovery. AI-powered search engines enhance the discoverability of library resources by understanding user queries better than traditional keyword-based systems.[8] By guiding users to relevant materials they might not have discovered otherwise, these systems optimize the use of library collections and enhance the overall quality of information retrieval.[2]
For instance, the Primo Research Assistant and Natural Language Search (“Ask Anything”) are tools designed to lower technical barriers for patrons, allowing them to interact with the catalog using plain English.[5, 6] These systems analyze user search patterns and behavioral histories to offer individually tailored resource recommendations, sometimes even introducing users to emerging scholarly domains that align with their academic trajectories.[1] This proactive curation makes it easier for users to find suitable books, articles, and multimedia content without requiring manual intervention from staff.[12]
Predictive Analytics: Transitioning to Proactive Collection Management
One of the most strategic applications of AI in the modern library is predictive analytics, which is revolutionizing collection management.[9, 10] Predictive analytics involves the use of historical data, machine learning algorithms, and statistical models to forecast future trends, resource demands, and user behaviors.[13, 14] This allows libraries to shift from a reactive purchasing model to a proactive stance, anticipating the needs of their communities before they are explicitly expressed.[13, 15]
Acquisition and Resource Allocation
By analyzing usage trends, borrowing patterns, and external factors like academic curricula or research output, predictive analytics systems help the acquisition section make data-driven decisions.[9, 10, 15] For example, regression analysis can establish relationships between multiple variables, such as user demographics and academic calendars, to provide accurate predictions of resource requirements.[14] Time series forecasting methods, such as ARIMA (Autoregressive Integrated Moving Average), are particularly effective for understanding seasonal variations and cyclic borrowing patterns, ensuring that high-demand resources are readily available when needed.[14]
This predictive capability ensures cost-effective budget planning by identifying “deserts and overages” in the collection.[14] Librarians can prioritize acquisitions that are likely to see significant use while eliminating unnecessary financial drain on resources that are not in high demand.[14, 15] Furthermore, predictive analytics can be applied to digital resource management, helping libraries optimize licenses and subscriptions to avoid paying for access levels that exceed actual demand.[14]
| Predictive Analytic Task | Machine Learning Algorithm / Technique | Library Application |
|---|---|---|
| Demand Forecasting | ARIMA / Exponential Smoothing | Predicting seasonal peaks in resource usage [14] |
| User Segmentation | K-Means / DBSCAN Clustering | Identifying distinct user groups for targeted promotions [14, 16] |
| Resource Optimization | Regression Analysis | Correlating demographics and curricula with book demand [14] |
| Decision Support | Text-to-SQL Pipelines | Enabling natural language queries of circulation data [17] |
Weeding and Preservation Strategies
The process of de-selection, or weeding, is equally transformed by data-driven strategies.[13] AI-enabled systems categorize books and identify underutilized materials that are candidates for weeding or off-site storage.[16] Research has shown a strong correlation between well-designed collection strategies—which include aggressive, data-informed weeding—and user satisfaction.[14] Relevant, up-to-date materials encourage usage, while the presence of “old world” or irrelevant materials often discourages it.[14] By supporting proactive weeding and retention strategies, AI-driven analytics ensures that library space is maximized and that the collection remains vibrant and relevant to the community’s current needs.[16]
Enhancing the User Experience: Personalization and Accessibility
The integration of AI is significantly reshaping the user experience (UX), with a study indicating a strong correlation of r=0.791 between AI integration and positive user perception.[3] This improvement is largely driven by personalized recommendations, virtual assistance, and enhanced accessibility.[3]
Virtual Assistants and Chatbots for 24/7 Support
Academic and public libraries are witnessing a surge in the use of AI-based chatbots and virtual assistants to provide round-the-clock support.[9, 18] These tools handle high volumes of repetitive inquiries—such as questions about library hours, account management, and basic resource location—which can otherwise leave staff stretched thin.[12, 19] Chatbots provide a uniform level of service and quick responses, which improves overall satisfaction for users who prefer digital solutions or require assistance beyond standard operating hours.[18, 19]
While basic chatbots operate on hierarchical decision trees, more advanced hybrid models combine menu-based interactions with keyword recognition and machine learning to converse more naturally with users.[18] However, these systems still face limitations in handling complex or nuanced research queries, often necessitating human intervention for in-depth scholarly assistance.[11, 19] This underscores the reality that AI is a way to extend human expertise, not replace it.[20]
AI-Driven Accessibility and Inclusion
AI technologies are critical for improving accessibility for patrons with disabilities.[8] For instance, AI-powered image description tools use computer vision to generate textual descriptions of visual content, making digital archives accessible to users with vision impairments.[8] Text-to-speech capabilities and navigation systems also enhance the autonomy of patrons in special library settings.[8, 20] In one study, AI-based assistive tools were found to have a significantly positive impact on patron autonomy, though this was balanced by ongoing concerns regarding the privacy and data security of these users.[20]
Strategic Insights and Decision-Making: The Case of ChatGPT in Library Analytics
A landmark study at the University of Toronto Libraries (UTL) showcased the transformative potential of generative AI, specifically ChatGPT, in library data analytics and strategic decision-making.[21] By integrating AI, UTL was able to move beyond efficiency gains to a paradigm where they could extract more robust, informed strategic insights from their data.[21]
UTL applied ChatGPT across seven distinct data analytics tasks, demonstrating practical utility that fundamentally shifted how the institution engaged with its information [21]:
• Data Preparation and Cleaning: The AI was used to automate the reformatting of dates, the standardization of contributor names, and the identification of duplicate values. In one experiment, ChatGPT successfully generated a Python script to normalize call numbers by removing Cutter numbers and publication years.[21]
• Code Generation: Library staff utilized the AI to write sophisticated Python scripts for data analysis without requiring a dedicated programmer. This included scripts that could identify trends in print format usage and automatically visualize them through graphs and charts.[21]
• Exploratory Data Analysis (EDA): ChatGPT analyzed dataset structures, identified outliers, and extracted meaningful patterns from large datasets. It helped identify seasonal usage fluctuations and platform preferences in COUNTER reports, which now inform the library’s collection development strategies.[21]
• Data Visualization: The system transformed raw numbers into actionable visuals, creating line graphs for circulation trends, radar charts for e-book usage by Library of Congress (LC) classification, and heat maps to visualize physical resource usage within stack ranges.[21]
• Data Categorization: UTL used AI to categorize user search queries by assigning disciplines and subjects to raw search strings, achieving a level of efficiency and accuracy that would have traditionally required extensive human domain knowledge.[21]
This technological leap allowed UTL to respond more dynamically to user needs and optimize resource allocation.[21] By automating repetitive tasks—which typically consume the majority of an analyst’s time—professionals were able to focus on the strategic implications of the data, such as identifying emerging research trends or spotting gaps in collection coverage.[21]
Ethical Challenges: Bias, Privacy, and the Black Box Dilemma
While the benefits of AI are significant, its integration into libraries raises serious ethical and governance considerations.[20] Libraries have long been advocates for intellectual freedom and user privacy, principles that often come into tension with the data-intensive nature of AI systems.[22, 23]
Algorithmic Bias and Fairness
One of the most critical ethical issues is algorithmic bias, which can result in societal biases being reinforced and inequalities being perpetuated.[10] If an AI model is trained on biased historical data, it may produce unfair search results or marginalized recommendations, potentially creating “echo chambers” that limit exposure to diverse perspectives.[10, 22, 24] LIS professionals have an ethical obligation to ensure that information is provided without prejudice, which requires regular evaluation, auditing, and the updating of models to address evolving concerns.[2, 24]
The “Black Box” Problem and Transparency
Many AI systems, particularly those using deep learning, function as “black boxes”—their decision-making processes are not easily interpretable by humans.[22, 24] This opacity creates challenges for accountability and transparency; librarians and users may not understand why a certain resource was recommended or why an acquisition tool excluded a particular title from purchase.[22] To build user trust, libraries must prioritize “explainable AI” (XAI) and consider adopting open-source models that allow for greater customization and ethical oversight.[22, 23]
Privacy and Data Profiling
AI systems require access to vast amounts of user data, including borrowing histories and digital interactions, making them susceptible to risks like unauthorized access, data breaches, and algorithmic profiling.[22, 23] This profiling may unintentionally expose individuals to targeted surveillance, especially those researching sensitive topics.[23] There is a fundamental trade-off between the convenience of personalized recommendations and the traditional library value of anonymity.[22]
To mitigate these risks, libraries must adopt robust data protection strategies, including [23]:
• Anonymization and Encryption: Protecting sensitive user information from unauthorized access.
• Data Minimization: Collecting only the data strictly necessary for the AI to function.
• Privacy-Preserving AI: Techniques like differential privacy and federated learning, which enable data processing while keeping individual identities hidden.
• User Consent and Transparency: Establishing clear policies for data collection and ensuring users are informed of how their data is being profiled.
Intellectual Property and the Legal Landscape of AI Training
The legal framework surrounding AI and copyright has evolved rapidly between 2024 and 2025, with several landmark court decisions shaping how libraries and AI developers can use copyrighted collections.
The Fair Use Defense in AI Training
In the 2025 case Bartz v. Anthropic PBC, District Judge William Alsup ruled that using purchased, copyrighted books for training AI models qualifies as fair use.[25, 26] The court found the purpose of using copyrighted works to train Large Language Models (LLMs) to be “transformative—spectacularly so,” likening the AI’s learning process to a human reader aspiring to be a writer.[25, 27] The judge noted that the training was “reasonably necessary” for the LLM to perform its function and that there was no traceable connection in the AI’s output that would constitute an infringing copy.[25, 27]
However, the court also established a critical boundary regarding the provenance of the training data. Judge Alsup ruled that using “pirated” copies to build a central digital library was not fair use, doubting that an infringer could justify downloading copies from pirate sites when they could have been purchased or accessed lawfully.[25, 28] This highlights that while the act of training may be transformative, the source of the data remains a vital legal consideration.[26]
Market Effects and Licensing
The U.S. Copyright Office (USCO) released a report in May 2025 identifying “market effects” as the single most important element of fair use in the generative AI context.[28] Potential harms include lost sales, lost licensing opportunities, and “market dilution,” where the output from an AI is similar enough to copyrighted works to reduce their demand.[26, 28] For AI companies and libraries, the report suggests that implementing effective guardrails to prevent infringing outputs and leveraging existing data licensing frameworks will weigh in favor of a fair use finding.[28]
| Fair Use Factor | Ruling in Bartz v. Anthropic | Key Reasoning |
|---|---|---|
| Purpose and Character | Favors AI Company | Highly transformative; likened to human reading/learning [25, 26] |
| Nature of Work | Favors Authors | Books contain significant creative and expressive elements [25, 27] |
| Amount Used | Favors AI Company | Entire works necessary for the transformative training function [25, 27] |
| Effect on Market | Favors AI Company | Copyright is designed to promote work, not protect against competition [27] |
Workforce Transformation: Reskilling and the Changing Role of Librarians
The rise of AI is fundamentally redefining academic librarianship, shifting the professional role from a “knowledge carrier” or “custodian” to a “competent partner” and “digital educator”.[29] While there is a fear of job displacement, particularly for those engaged in routine technical services, the emerging consensus is that AI complements rather than substitutes for librarians.[10, 29, 30]
The Imperative for AI Literacy
AI adoption necessitates a reevaluation of the skills required for the profession.[11] Proficiency in AI allows library professionals to automate routine tasks, freeing up time for more complex research support, information literacy instruction, and ethical oversight.[29, 31] This requires librarians to gain competencies in digital literacy, AI management, and what is becoming known as “metaliteracy”.[11, 32]
A key case study in this professional evolution is the GPT-4 Exploration Program at the University of New Mexico’s College of University Libraries and Learning Sciences.[31] The program aimed to foster a culture of continuous learning and adaptability through a practical, hands-on approach to advanced AI technology.[31] Participants reported significant improvements in AI literacy and confidence, demonstrating that well-structured reskilling initiatives can empower library professionals to navigate the AI-driven landscape effectively.[31]
Prompt Engineering and Educational Roles
Librarians are increasingly taking on the role of instructors in the use of AI tools.[29] This includes teaching “prompt engineering”—an emergent discipline at the intersection of GAI, library science, and UX design.[33] Frameworks like TCEPFT (Task, Context, Example, Persona, Format, Tone) are being used by information professionals to refine iterative processes and produce consistent, high-quality outputs.[33] This educational shift ensures that libraries remain essential conduits for cutting-edge technologies and critical thinking in an era of information deluge.[29, 30]
Case Study: AI Innovation in a Specialized Educational Setting
Small and specialized libraries are also finding unique ways to innovate with AI. At a veterinary technology school library in Pittsburgh, PA, staff have integrated AI tools into several core areas to support student success [34]:
• Reference Services: Using AI-powered search engines and automated reference assistance, students can access relevant information more efficiently, reducing the frustration of sorting through dense clinical material.[34]
• Collection Development: AI analyzes usage patterns to ensure the collection meets the high research needs of a specialized medical program.[34]
• Information Literacy Instruction: Faculty and librarians collaborate to embed AI tools into assignments. This includes scaffolded research tasks where students compare AI-generated summaries with peer-reviewed sources to develop critical evaluation skills.[34]
The outcomes of these initiatives have been measurable. Faculty reported improvements in the quality and depth of student assignments, and students using AI-supported study guides showed a 15−20% improvement in assessment scores related to clinical terminology and case analysis.[34] This case study illustrates how AI can be embedded into specialized workflows to enhance, rather than replace, human learning.[34]
Global Adoption Trends and Regional Variations (2024–2025)
Global research conducted by Clarivate and ProQuest shows a steady rise in AI exploration and implementation, with adoption increasing from 63% in 2024 to 67% by 2025.[20, 35, 36, 37] However, this progress is described as “cautious,” with the majority of libraries (35%) still in the early evaluation stages.[35, 36, 37]
Differences by Library Type
Academic libraries are leading the way in active implementation. In 2025, 21% of academic libraries were in the initial implementation stage, compared to only 12% for all libraries.[35, 37] Public libraries remain more cautious; only 20% of public librarians expressed optimism about the benefits of AI over a five-year period—a decline from 26% in 2024—and 54% reported they have no plans or are not actively pursuing AI technologies.[35, 36] For public libraries, community engagement remains the primary mission, and concerns about privacy and security (65%) are significantly more pronounced.[35]
Regional and Role-Based Gaps
The pace of adoption varies significantly by region. Asia and Europe have continued to advance, with 37−40% of libraries in initial implementation or beyond.[37] In contrast, the United States reports lower optimism (7%) and confidence levels, likely influenced by severe budget constraints, which were cited by 62% of all respondents as a top challenge.[35, 37, 38]
A perception gap is also evident across professional roles. Senior librarians are more likely to prioritize library efficiencies and report higher confidence in AI terminology (43%) compared to junior librarians (36%).[35, 36, 37] This suggests that change management and inclusive professional development are essential for successful, long-term AI integration.[35]
| Metric | 2024 Survey | 2025 Survey | Trend |
|---|---|---|---|
| Overall AI Exploration / Implementation | 63% | 67% | Steady Increase [36, 37] |
| Academic Libraries in Implementation | 14% | 21% | Significant Growth [35] |
| Public Library Optimism (5-year view) | 26% | 20% | Declining Optimism [35, 36] |
| Number of AI Objectives Selected (Avg) | 3 | 4 | Broadening Use Cases [36] |
| Confidence in AI Terminology (Senior) | N/A | 43% | Higher Confidence in Leadership [35, 37] |
Future Outlook: The Physical and Digital Library of the AI World
The future of libraries lies in the creation of seamless access and immersive user engagement. Digital resources have largely replaced traditional print collections for routine information needs, but libraries of the future will prioritize connectivity beyond physical walls, utilizing mobile-friendly platforms and responsive design.[2]
Immersive technologies such as VR and AR will convert libraries into spaces for engaging experiences—virtual library tours, AR-enhanced exhibits, and VR learning environments will redefine user engagement.[2] AI will be the invisible engine powering these experiences, using user data to anticipate needs, provide personalized research assistance, and curate collaborative e-learning ecosystems.[2]
Conclusion: The Proactive, Cognitive Infrastructure
The transformation of libraries using AI platforms represents a fundamental paradigm shift from reactive service to proactive facilitation. This shift is not merely technological but cultural, necessitating new competencies in data literacy, prompt engineering, and ethical oversight for library professionals. The evidence from 2024 and 2025 demonstrates that while operational efficiency and user discovery are significantly enhanced by AI, the human element remains irreplaceable.
To successfully navigate this transition, libraries must:
• Prioritize AI Literacy: Professional development that integrates technical and ethical dimensions builds institutional readiness and confidence.[20]
• Adopt Ethical Frameworks: Ensuring transparency, mitigating bias, and protecting user privacy are non-negotiable for maintaining the library as a safe, trusted space.[22, 23, 39]
• Embrace Small Pilots: Manageable, human-supervised projects allow for scalability and let user-centered design guide the implementation.[20, 39]
• Advocate for Equitable Access: Supporting open access initiatives and ensuring that AI-enhanced services do not exclude users based on socioeconomic status is vital for the library’s mission.[2, 3]
As AI continues to evolve, libraries will increasingly serve as central hubs of a shared, responsible, and equitable future, reclaiming their place at the center of the scholarly and community information landscape.[32] By leveraging the transformative potential of AI while preserving their core values, libraries will remain indispensable pillars of knowledge in the 21st century.
——————————————————————————–
1. Academic Library with Generative AI: From Passive Information Providers to Proactive Knowledge Facilitators – MDPI, https://www.mdpi.com/2304-6775/13/3/37
2. Reshaping the library landscape: Exploring the integration of artificial intelligence in libraries, https://ijlsit.org/archive/volume/9/issue/1/article/3116
3. Reshaping the library landscape: Exploring the integration of artificial intelligence in libraries, https://www.semanticscholar.org/paper/Reshaping-the-library-landscape%3A-Exploring-the-of-Kumar-Jyoti/89989ffc57fa9b04a85ebfcda85ff00110d3df2d
4. AI Implementation in Library Metadata – Taylor & Francis, https://think.taylorandfrancis.com/special_issues/ai-implementation-in-library-metadata/
5. Library Management System : Alma – Ex Libris, https://exlibrisgroup.com/products/alma-library-services-platform/
6. Academic AI in Action: Enhancing Library Services with Alma, https://exlibrisgroup.com/blog/academic-ai-in-action-enhancing-library-services-with-alma/
7. Smarter Libraries through AI Technologies – Vendors, https://sites.google.com/usc.edu/libraryai/vendors
8. AI in Libraries – galibtech, https://galibtech.georgialibraries.org/emerging/ai-in-libraries
9. (PDF) Integration of AI-Driven Tools in Academic Library Services – ResearchGate, https://www.researchgate.net/publication/393729824_Integration_of_AI-Driven_Tools_in_Academic_Library_Services
10. Integration of AI-Driven Tools in Academic Library Services – IJFMR, https://www.ijfmr.com/papers/2025/4/50110.pdf
11. Effects of AI-driven tools on reference services and staff roles in …, https://www.emerald.com/rsr/article/doi/10.1108/RSR-02-2025-0008/1318048/Effects-of-AI-driven-tools-on-reference-services
12. How AI Is Modernizing Library Services in State and Local Government, https://vidizmo.ai/blog/ai-for-library-services
13. Liblime The Data-Driven Library: Informing Strategic Decisions Through Analytics, https://liblime.com/2025/12/14/the-data-driven-library-informing-strategic-decisions-through-analytics/
14. Predictive Analytics for Resource Planning and Acquisition in Libraries – ResearchGate, https://www.researchgate.net/publication/390662106_Predictive_Analytics_for_Resource_Planning_and_Acquisition_in_Libraries
15. Developing predictive analytics frameworks to optimize collection development in modern libraries, https://orionjournals.com/ijsru/sites/default/files/IJSRU-2023-0038.pdf
16. AI-Enabled Predictive Analytics for Collection Development in College Libraries., https://collegelibraries.in/index.php/CL/article/view/192
17. Case Study: How Follett Modernized K-12 Library Management with GenAI | Tribe AI, https://www.tribe.ai/case-studies/follett-case-study
18. Artificial intelligence (AI) enabled reference services in libraries – Credence Publishing, https://www.credence-publishing.com/journal/uploads/archive/202517512888567328756358.pdf
19. Smart Libraries: Transforming User Experiences with AI-Driven Assistance – ResearchGate, https://www.researchgate.net/publication/389322334_Smart_Libraries_Transforming_User_Experiences_with_AI-Driven_Assistance
20. Research on Artificial Intelligence (AI) Adoption in Libraries (2023–2025), https://learnworkecosystemlibrary.com/topics/research-on-artificial-intelligence-ai-adoption-in-libraries-2023-2025/
21. van Ballegooie | Transforming Library Data Analytics into Strategic …, https://journals.publishing.umich.edu/nasig/article/id/7753/
22. (PDF) Ethical and Privacy Challenges of AI in Libraries, https://www.researchgate.net/publication/396274272_Ethical_and_Privacy_Challenges_of_AI_in_Libraries
23. Advancing ethical AI practices to solve data privacy issues in library systems – Orion Scholar Journals Publication, https://orionjournals.com/ijmru/sites/default/files/IJMRU-2023-0063.pdf
24. Full article: Artificial intelligence algorithm bias in information retrieval systems and its implication for library and information science professionals: A scoping review, https://www.tandfonline.com/doi/full/10.1080/07317131.2025.2512282
25. District Court Issues AI Fair Use Decision: Using Copyrighted Works to Train AI Models Is Fair Use, but Using Pirated Copies to Build a Central Library Is Not | Insights & Resources | Goodwin, https://www.goodwinlaw.com/en/insights/publications/2025/06/alerts-practices-aiml-district-court-issues-ai-fair-use-decision
26. A Tale of Three Cases: How Fair Use Is Playing Out in AI Copyright Lawsuits | Insights, https://www.ropesgray.com/en/insights/alerts/2025/07/a-tale-of-three-cases-how-fair-use-is-playing-out-in-ai-copyright-lawsuits
27. Generative AI Training and Fair Use: The Anthropic and Meta Decisions – Lewis Rice, https://www.lewisrice.com/publications/generative-ai-training-and-fair-use-the-anthropic-and-meta-decisions
28. US Copyright Office Issues Report Addressing Use of Copyrighted Material to Train Generative AI Systems – McDermott Will & Schulte, https://www.mwe.com/insights/us-copyright-office-issues-report-addressing-use-of-copyrighted-material-to-train-generative-ai-systems/
29. “Next Gen Librarianship: Mastering AI and Data Science Roles”, https://kuey.net/index.php/kuey/article/download/11097/8644/20498
30. Can AI Become an Information Literacy Ally? A Survey of Library Instructor Perspectives on ChatGPT | Del Castillo, https://crl.acrl.org/index.php/crl/article/view/26938/34834
31. Transforming Academic Librarianship through AI Reskilling: Insights from the GPT-4 Exploration Program – UNM Digital Repository, https://digitalrepository.unm.edu/cgi/viewcontent.cgi?article=1224&context=ulls_fsp
32. Libraries’ opportunity to shape how AI transforms society | ALA, https://www.ala.org/news/2025/06/libraries-opportunity-shape-how-ai-transforms-society
33. The New Literacies: Prompt Engineering Skills and AI Literacy Among Librarians in Industry 5.0 | Request PDF – ResearchGate, https://www.researchgate.net/publication/397828254_The_New_Literacies_Prompt_Engineering_Skills_and_AI_Literacy_Among_Librarians_in_Industry_50
34. How One Small Library is Innovating with AI – Katina Magazine, https://katinamagazine.org/content/article/future-of-work/2025/how-one-small-library-is-innovating-with-ai
35. Pulse of the Library | Clarivate, https://clarivate.com/pulse-of-the-library/
36. Pulse of the Library: Reflecting the voices of librarians worldwide – Clarivate, https://clarivate.com/academia-government/blog/pulse-of-the-library-reflecting-the-voices-of-librarians-worldwide/
37. Clarivate Pulse of the Library Report Reveals Link Between AI Literacy, AI Implementation and Confidence, https://ir.clarivate.com/news-events/press-releases/news-details/2025/Clarivate-Pulse-of-the-Library-Report-Reveals-Link-Between-AI-Literacy-AI-Implementation-and-Confidence/default.aspx
38. Clarivate Pulse of the Library Report Reveals Link Between AI Literacy, AI Implementation and Confidence – PR Newswire, https://www.prnewswire.com/news-releases/clarivate-pulse-of-the-library-report-reveals-link-between-ai-literacy-ai-implementation-and-confidence-302598196.html
39. AI in Libraries: Revolutionizing Access to Digital Knowledge – Tribe AI, https://www.tribe.ai/applied-ai/ai-in-digital-libraries
Journal for Policy & Market Research 
Leave a comment