Artificial Intelligence/Robots

“Prompting Generative AI to Catalog: The Promise and the Reality”

The jagged technological frontier visualizes the boundary at which AI can be an asset versus a detriment for the user. Due to the inaccuracies of the generated MARC records in this project, MARC record generation barely landed on the favorable side of this frontier if using an effective prompt. However, even these records could not be trusted without a cataloger in the loop. Admittedly there is an art to cataloging, but it must be grounded in reality-based adherence to standards and norms, not one of creative writing.

https://tinyurl.com/4kjmxfae

“Clarivate Pulse of the Library Report Reveals Link Between AI Literacy, AI Implementation and Confidence”

The findings reveal a steady rise in artificial intelligence (AI) adoption, with 67% of libraries exploring or implementing AI tools, an increase from 63% in 2024. While the majority remain at the initial stages of evaluation, early adopters are pressing ahead and reporting greater optimism, particularly as they progress through implementation phases.

The report also shows that libraries are more likely to be in the moderate or active implementation phases of AI when AI literacy is part of the formal training or onboarding program (28%), librarians have dedicated time/resources (23.3%), or have managers actively encouraging development (24.2%).

https://tinyurl.com/49933npf

“Shifting Norms in Scholarly Publications: Trends in Readability, Objectivity, Authorship, and AI Use”

Academic and scientific publishing practices have changed significantly in recent years. This paper presents an analysis of 17 million research papers published since 2000 to explore changes in authorship and content practices. It shows a clear trend towards more authors, more references and longer abstracts. While increased authorship has been reported elsewhere, the present analysis shows that it is pervasive across many major fields of study. We also identify a decline in author productivity which suggests that `gift’ authorship (the inclusion of authors who have not contributed significantly to a work) may be a significant factor. We further report on a tendency for authors to use more hyperbole, perhaps exaggerating their contributions to compete for the limited attention of reviewers, and often at the expense of readability. This has been especially acute since 2023, as AI has been increasingly used across many fields of study, but particularly in fields such as Computer Science, Engineering and Business. In summary, many of these changes are causes of significant concern. Increased authorship counts and gift authorship have the potential to distort impact metrics such as field-weighted citation impact andh-index, while increased AI usage may compromise readability and objectivity.

https://arxiv.org/abs/2510.21725

“Experimental Evidence of the Effects of Large Language Models Versus Web Search on Depth of Learning”

A theory is proposed that when individuals learn about a topic from LLM syntheses, they risk developing shallower knowledge than when they learn through standard web search, even when the core facts in the results are the same. This shallower knowledge accrues from an inherent feature of LLMs—the presentation of results as summaries of vast arrays of information rather than individual search links—which inhibits users from actively discovering and synthesizing information sources themselves, as in traditional web search. . . . Results from seven online and laboratory experiments (n = 10,462) lend support for these predictions, and confirm, for example, that participants reported developing shallower knowledge from LLM summaries even when the results were augmented by real-time web links. Implications of the findings for recent research on the benefits and risks of LLMs, as well as limitations of the work, are discussed.

https://doi.org/10.1093/pnasnexus/pgaf316

“Evaluating Scopus AI Versus Traditional Searches for Literature Review about Prepectoral Breast Reconstruction: An Exploratory Study”

Results: The Scopus AI search retrieved 25 articles, while the traditional method identified 4. After removing duplicates, non-English texts, and non-relevant sources, 17 articles were included in the final analysis. Scopus AI provided automatic summaries, while manual review was required for the traditional method. No overlap was found between the two methods. Conclusions: AI tools like Scopus AI can enhance the speed and breadth of literature reviews, but human oversight remains essential to ensure relevance and quality. Combining AI with traditional methods may offer a more balanced and effective approach for clinical research.

https://doi.org/10.3390/medsci13040211

Paywall: “Tracing the Past, Predicting the Future: A Systematic Review of AI in Archival Science”

Our study highlights how integrating artificial intelligence (AI) into archival science can help address these issues. We begin with a thorough analysis of 45 papers published between 2011 and 2023 that met our predetermined criteria. . . . We investigated the key AI techniques and their applications in archives and records management functions. Our findings highlight key AI-driven strategies that promise to streamline recordkeeping processes and improve data retrieval in the immediate future.

https://doi.org/10.1002/pra2.1286

Paywall: “Fueling Conversations: AI Education across the iSchools in the US and Canada”

This study analyzed the AI-related courses in graduate and undergraduate programs offered by members of the iSchools organization in the United States and Canada. . . . Of the 51 iSchools, twenty-nine offered AI-related courses. The most covered areas include general AI, Machine Learning, Natural Language Processing, Deep Learning, and Robotics. Most courses focus on AI’s technical and applied facets, while a few cover the ethical, societal, cultural, and legal facets. Implications include the need for iSchools to offer AI courses that cover aspects beyond the technical, more undergraduate courses, and certificate programs that contribute to educating the labor force that needs upskilling.

https://doi.org/10.1002/pra2.1238

Paywall: “‘It’s Like Some Weird AI Ouroboros’”: Artificial Intelligence Use and Avoidance in Scholarly Peer Review”

This interdisciplinary perspective paper explores the evolving relationship between generative artificial intelligence (GenAI) and library reference services across academic and public libraries, with implications for Library and Information Science (LIS) education. . . . The paper traces historical developments in personalized reference, staffing trends, and technological transformations, arguing that GenAI tools should complement rather than replace human librarians. Additionally, the paper examines the impact of GenAI on archival reference, specialized services, and the emergence of conversational assistants as digital intermediaries. Ethical considerations are also addressed, including misinformation, epistemic agency, and belief formation.

https://doi.org/10.1002/pra2.1232

| Artificial Intelligence |
| Research Data Curation and Management Works |
| Digital Curation and Digital Preservation Works |
| Open Access Works |
| Digital Scholarship |

ACRL: “AI Competencies for Academic Library Workers”

This document contains two sections: mindsets (guiding orientations or dispositions) and competencies (skills, knowledge, behaviors, and abilities). Mindsets are presented in a single list. Competencies are organized into four categories: Ethical Considerations; Knowledge & Understanding; Analysis & Evaluation; and Use & Application. These parallel the categories Davy Tsz Kit Ng and colleagues identified in their content analysis of 18 articles about AI literacy. Each category contains four or five broad competencies. Each broad competency has a brief description and a corresponding list of related abilities.

https://tinyurl.com/4dxuauvy

"AI Adoption Jumps to 84% Among Researchers as Expectations Undergo Significant ‘Reality Check’"

The comprehensive study of more than 2,400 researchers worldwide finds that researchers remain optimistic about AI, with 85% reporting that it has improved their efficiency, and close to three-quarters saying it has enhanced both the quantity and quality of their work. Overall usage of AI tools surged from 57% in 2024 to 84% in 2025, including specific use for research and publication tasks, which grew significantly to 62% from 45%.

However, while AI usage has surged dramatically, researchers are significantly scaling back their expectations of what AI can currently do as they gain firsthand experience, moving beyond hype toward nuanced, evidence-based adoption. Last year, researchers believed AI had already outperformed humans for over half of the potential use cases presented. This year, that figure dropped to less than one-third.

https://tinyurl.com/y7ezcdjv

| Research Data Publication and Citation Bibliography | Research Data Sharing and Reuse Bibliography | Research Data Curation and Management Bibliography | Digital Scholarship |

Wiley Advances Research Exchange Platform with AI-Driven Automation, Streamlining Article Placement and Publication”

The new capabilities significantly enhance the ability of editors to match quality papers with the right journals and give confidence to authors in selecting the next journal for their submission, an otherwise time-consuming manual process. They include:

AI-powered transfer suggestions: When editors reject a quality manuscript that is not the right fit for a particular journal, AI technology suggests up to five better-matched Wiley journals. Transfers occur when editors recommend them and authors agree, reinforcing editorial independence and author choice.

Preserved peer review: Expert reviewer feedback travels with manuscripts when they transfer between journals, preserving valuable review work and accelerating editorial decisions.

Streamlined article resubmission: The automated transfer process means that authors don’t need to reformat or re-enter information, saving time and energy.

https://tinyurl.com/bddahdab

“Relatively Few Americans Are Getting News from AI Chatbots like ChatGPT”

About one-in-ten U.S. adults say they get news often (2%) or sometimes (7%) from AI chatbots like ChatGPT or Gemini. An additional 16% do so rarely, according to a recent Pew Research Center survey. Most Americans (75%) say they never get news this way. . . .

A third of those who use chatbots for news say they generally find it difficult to determine what is true and what is not. About a quarter (24%) say they find it easy to do so. But the largest share (42%) isn’t sure.

https://tinyurl.com/y4vz76xa

“AI Policies in U.S. Universities: A Critical Analysis of Policy Gaps and Library Involvement”

This posIT column critically examines AI policies and resources at 50 four-year universities—one from each U.S. state—to assess alignment with the Association of Research Libraries’ (ARL) Guiding Principles for Artificial Intelligence. Through content analysis of LibGuides, AI taskforce membership, campus events, and public-facing policies, the study reveals widespread adoption of AI resources but a significant lack of clarity, consistency, and librarian involvement in policy development.

https://doi.org/10.1080/01930826.2025.2560268

“Prestige over Merit: An Adapted Audit of LLM (Large Language Models) Bias in Peer Review”

Large language models (LLMs) are playing an increasingly integral, though largely informal, role in scholarly peer review. Yet it remains unclear whether LLMs reproduce the biases observed in human decision-making. We adapt a resume-style audit to scientific publishing, developing a multi-role LLM simulation (editor/reviewer) that evaluates a representative set of high-quality manuscripts across the physical, biological, and social sciences under randomized author identities (institutional prestige, gender, race). The audit reveals a strong and consistent institutional-prestige bias: identical papers attributed to low-prestige affiliations face a significantly higher risk of rejection, despite only modest differences in LLM-assessed quality. To probe mechanisms, we generate synthetic CVs for the same author profiles; these encode large prestige-linked disparities and an inverted prestige-tenure gradient relative to national benchmarks. The results suggest that both domain norms and prestige-linked priors embedded in training data shape paper-level outcomes once identity is visible, converting affiliation into a decisive status cue.

https://arxiv.org/abs/2509.15122

STM: Recommendations for a Classification of AI Use in Academic Manuscript Preparation

This document presents a classification of various ways that artificial intelligence (AI) can be used to assist in the preparation of academic manuscripts. It is intended to serve as a framework for publishers to individually develop policies on how AI may be used and should be declared by authors.

https://tinyurl.com/4v66r38p

“ARL/CNI Futurescape Libraries AI Toolkit Can Help You Thrive in the AI Landscape”

The Futurescape Libraries AI Toolkit integrates the ARL/CNI AI Scenarios along with priorities trialed and refined by strategic thinkers working in the research library field during a Strategic Implications forum in December 2024. . . .

Organized into flexible modules, the toolkit offers structured activities to help library leadership teams, staff, and external stakeholders:

Explore future possibilities

Test current strategies

Identify opportunities and vulnerabilities

Build readiness for long-term change

https://tinyurl.com/bdss8hac

“Keywords Are Not Always the Key: A Metadata Field Analysis for Natural Language Search on Open Data Portals”

Open data portals are essential for providing public access to open datasets. However, their search interfaces typically rely on keyword-based mechanisms and a narrow set of metadata fields. This design makes it difficult for users to find datasets using natural language queries. The problem is worsened by metadata that is often incomplete or inconsistent, especially when users lack familiarity with domain-specific terminology. In this paper, we examine how individual metadata fields affect the success of conversational dataset retrieval and whether LLMs can help bridge the gap between natural queries and structured metadata. We conduct a controlled ablation study using simulated natural language queries over real-world datasets to evaluate retrieval performance under various metadata configurations. We also compare existing content of the metadata field ‘description’ with LLM-generated content, exploring how different prompting strategies influence quality and impact on search outcomes. Our findings suggest that dataset descriptions play a central role in aligning with user intent, and that LLM-generated descriptions can support effective retrieval. These results highlight both the limitations of current metadata practices and the potential of generative models to improve dataset discoverability in open data portals.

https://arxiv.org/abs/2509.14457

“Implementing AI in Library-Led Programs to Foster Critical Information Literacy”

The spread of fake news and misinformation poses significant challenges to the integrity of information ecosystems, undermining public trust. Libraries, traditionally trusted sources of credible information, are in a unique position to address this issue through the integration of artificial intelligence (AI). This paper explores the potential of AI to detect misinformation and enhance critical information literacy. AI technologies like natural language processing and machine learning can analyze text patterns, verify sources, and identify fake news at scale. Tools such as fact-checking algorithms and real-time content monitoring systems can help librarians curate reliable resources and guide users in distinguishing credible information from misinformation. AI can also be employed to promote critical information literacy through personalized educational experiences, including chatbots and virtual assistants that offer on-demand guidance on evaluating information. Ethical considerations play a crucial role in AI implementation. The paper addresses concerns over biases in AI algorithms, data privacy, and the ethics of automated decision-making. Strategies for mitigating these risks include prioritizing transparency, accountability, and user-centered design. By upholding ethical standards, libraries can align AI use with their core mission of serving the public good. The study also highlights the practical challenges libraries face in adopting AI, such as resource constraints, staff training, and system integration. Case studies from pioneering institutions offer insights into overcoming these barriers. Libraries can implement AI to combat misinformation and foster critical information literacy while maintaining ethical principles. This approach strengthens libraries’ roles in ensuring informed, equitable access to information and positions them as key players in the fight against fake news.

https://doi.org/10.20944/preprints202509.1281.v1

“Chatbot Assessment: Best Practices for Artificial Intelligence in the Library”

In November 2019, the Leonard Lief Library implemented Ivy.ai, a proprietary chatbot on its website. This implementation was the first academic library installation of a vendor-supplied chatbot to be discussed in the professional literature. This chatbot functioned as a new tool that assisted users seeking information from the library website. User questions provided insight to the authors about the kinds of topics students searched for via the library website. In April 2023, the chatbot’s vendor began using OpenAI’s ChatGPT Application Programming Interface (API) to improve the chatbot’s functionality. This change, from a rules-based chatbot system to a transformer model, enhanced the chatbot’s ability to provide answers to patrons. To better understand this major change, the authors assessed the chatbot’s usage during the Spring 2023 semester. This assessment revealed the kinds of questions the chatbot struggled to answer, and possible reasons why. The assessment’s findings demonstrated how chatbots can successfully function as a enhancement to the library website. The article also presents best practices for libraries looking to implement or experiment with chatbots and contributes to the ongoing discussion of artificial intelligence in libraries.

https://tinyurl.com/5cmzhm3w

“IEEE Launches Pilot with Hum’s Alchemist Review”

Alchemist Review performs comprehensive manuscript analysis, automatically identifying crucial research elements including primary hypotheses, methodological approaches, and claimed contributions. Additionally, the platform leverages Grounded AI’s citation verification technology to ensure reference accuracy and detect potential retractions or contextual inconsistencies.

https://tinyurl.com/paekxz3p

“Redefining Research: Elsevier Announces Next-Generation AI-Powered Researcher Solution”

What will set the new solution for researchers apart:

One seamless assistant: Brainstorm ideas, plan projects, review literature, find collaborators, and discover funding opportunities – all in one space with a powerful AI assistant.

Trust Cards: Showing how evidence was used or inferred, highlighting confidence levels and providing risk assessments for potential inaccuracies.

Certified content only: Access comprehensive, peer-reviewed, cross-publisher academic content.

Curated datasets: Answers powered by publisher-neutral datasets e.g. Scopus abstracts and funding data.

Add your own content: Users can add their own content to supplement what is already included.

Privacy and security: Built with enterprise-grade security, Elsevier AI-powered solutions are developed in line with its Privacy Principles to safeguard personal data and privacy.

Publisher-neutral algorithms: An independent Advisory Board will be created to ensure results are prioritized and ranked based on quality, in a transparent, unbiased and responsible manner.

https://tinyurl.com/y576n7ku

“Reviewers Increasingly Divided on the Use of Generative AI in Peer Review”

A new global reviewer survey from IOP Publishing (IOPP) reveals a growing divide in attitudes among reviewers in the physical sciences regarding the use of generative AI in peer review. . . .

Key Findings:

41% of respondents now believe generative AI will have a positive impact on peer review (up 12% from 2024), while 37% see it as negative (up 2%). Only 22% are neutral or unsure—down from 36% last year—indicating growing polarisation in views.

32% of researchers have already used AI tools to support them with their reviews.

57% would be unhappy if a reviewer used generative AI to write a peer review report on a manuscript they had co-authored and 42% would be unhappy if AI were used to augment a peer review report.

42% believe they could accurately detect an AI-written peer review report on a manuscript they had co-authored.

https://tinyurl.com/32294sd5

“How Much Are LLMs Changing the Language of Academic Papers After ChatGPT? A Multi-Database and Full Text Analysis”

This study investigates how Large Language Models (LLMs) are influencing the language of academic papers by tracking 12 LLM-associated terms across six major scholarly databases (Scopus, Web of Science, PubMed, PubMed Central (PMC), Dimensions, and OpenAlex) from 2015 to 2024. Using over 2.4 million PMC open-access publications (2021-July 2025), we also analysed full texts to assess changes in the frequency and co-occurrence of these terms before and after ChatGPT’s initial public release. Across databases, delve (+1,500%), underscore (+1,000%), and intricate (+700%) had the largest increases between 2022 and 2024. Growth in LLM-term usage was much higher in STEM fields than in social sciences and arts and humanities. In PMC full texts, the proportion of papers using underscore six or more times increased by over 10,000% from 2022 to 2025, followed by intricate (+5,400%) and meticulous (+2,800%). Nearly half of all 2024 PMC papers using any LLM term also included underscore, compared with only 3%-14% of papers before ChatGPT in 2022. Papers using one LLM term are now much more likely to include other terms. For example, in 2024, underscore strongly correlated with pivotal (0.449) and delve (0.311), compared with very weak associations in 2022 (0.032 and 0.018, respectively). These findings provide the first large-scale evidence based on full-text publications and multiple databases that some LLM-related terms are now being used much more frequently and together. The rapid uptake of LLMs to support scholarly publishing is a welcome development reducing the language barrier to academic publishing for non-English speakers.

https://arxiv.org/abs/2509.09596

“Pay-per-Output? AI Firms Blindsided by Beefed up robots.txt Instructions.”

Announced Wednesday morning, the “Really Simply Licensing” (RSL) standard evolves robots.txt instructions by adding an automated licensing layer that’s designed to block bots that don’t fairly compensate creators for content.

Free for any publisher to use starting today, the RSL standard is an open, decentralized protocol that makes clear to AI crawlers and agents the terms for licensing, usage, and compensation of any content used to train AI, a press release noted.

https://tinyurl.com/mrxjmdvw

“Paper2Agent: Reimagining Research Papers as Interactive and Reliable AI Agents”

We introduce Paper2Agent, an automated framework that converts research papers into AI agents. Paper2Agent transforms research output from passive artifacts into active systems that can accelerate downstream use, adoption, and discovery. Conventional research papers require readers to invest substantial effort to understand and adapt a paper’s code, data, and methods to their own work, creating barriers to dissemination and reuse. Paper2Agent addresses this challenge by automatically converting a paper into an AI agent that acts as a knowledgeable research assistant. It systematically analyzes the paper and the associated codebase using multiple agents to construct a Model Context Protocol (MCP) server, then iteratively generates and runs tests to refine and robustify the resulting MCP. These paper MCPs can then be flexibly connected to a chat agent (e.g. Claude Code) to carry out complex scientific queries through natural language while invoking tools and workflows from the original paper. We demonstrate Paper2Agent’s effectiveness in creating reliable and capable paper agents through in-depth case studies. Paper2Agent created an agent that leverages AlphaGenome to interpret genomic variants and agents based on ScanPy and TISSUE to carry out single-cell and spatial transcriptomics analyses. We validate that these paper agents can reproduce the original paper’s results and can correctly carry out novel user queries. By turning static papers into dynamic, interactive AI agents, Paper2Agent introduces a new paradigm for knowledge dissemination and a foundation for the collaborative ecosystem of AI co-scientists.

https://arxiv.org/abs/2509.06917