From hype to strategy: navigating the reality of experimental strategic adoption of AI technologies in libraries
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Purpose The article highlights the value of adopting an experimental strategy for artificial intelligence (AI) adoption in libraries, with a specific focus on the University of Toronto (UofT) libraries as a case study. The experimental approach entails carrying out small-scale, effective, quick, and reversible experiments to increase awareness, reduce risks with adoption of incorrect, ineffective, or full-scale adoption; increase flexibility in adopting new technologies in the rapidly evolving AI industry; and increase open-mindedness to consider diverse perspectives even though they go against one’s held perceptions, and develop dynamic capabilities to innovate. To fully realize the revolutionary potential of AI technologies in libraries, it is crucial to adopt new technologies strategically, driven by experimentation, collaboration, and knowledge sharing. Design/methodology/approach Mixed-model research involving case study of UofT libraries and the author’s professional experiences in digitally transforming libraries is used to answer the research question. AI is an emerging area and hence its applications in libraries. Leveraging the author’s professional and research expertise, the findings from the case study are enriched, offering broader perspectives and more nuanced implications. Findings Libraries can recognize emerging opportunities, adapt to the shifting AI landscape, and effectively exploit AI technologies because of the development of dynamic capabilities and a focus on innovation. The UofT instance sheds light on the experimental strategy and acts as a lens to comprehend how to strategically think about the complete AI spectrum rather than keeping an eye on a few technologies that otherwise might just be overhyped in media outlets. A mix of centralization and decentralization of AI technology adoption experimentation is evident at UOT, where any librarian is free to test out a new tool and share their findings with their peers in the expectation that other libraries will embrace it as well. The reverse scenario is also conceivable (top management to individual libraries). UofT’s culture fosters collaboration and knowledge-sharing among librarians, promoting experimentation and innovation. Cocreation with patrons, including student entrepreneurs, enhances dynamic capabilities and informs rational adoption decisions. Looking at the results, some future research directions emerge that could strengthen the library’s focus on AI. The future research directions indicate the need for further investigation into experiment design, particularly focusing on experimentation policies, monitoring and evaluation of experimentation activities, and fostering greater collaboration with patrons. Additionally, exploring AI adoption factors at both organizational and individual levels is essential to create a supportive environment for these experiments. Conducting continuous AI experiments enables librarians to critically assess AI technologies by leveraging their experiences with various applications, allowing them to distinguish practical solutions from market hype and concentrate on options that truly enhance their library operations. Practical implications The article contributes to the knowledge of strategic AI technology adoption and the role of experimentation in libraries’ adoption of AI technologies. This paper offers practical guidance for libraries of all sizes and resource levels seeking to experiment with AI technologies. It encourages the creation of a collaborative environment where patrons and peers can come together to experiment and share knowledge. Additionally, it encourages libraries to explore various research directions—such as defining experimentation policies, integrating monitoring and evaluation (M&E) to assess the effectiveness of experiments, fostering collaboration, and leveraging AI adoption factors—to cultivate a culture of experimentation. This approach aims to increase the number of experiments and, consequently, the adoption of valuable AI technologies. Originality/value AI in libraries is rapidly evolving, but current literature remains underdeveloped and lacks comprehensive adoption frameworks. Investigating individual libraries’ AI practices and sharing these insights will enable collaborative learning, helping them improve overall adoption process, fostering further innovation with emerging technologies, and helping in development of a theoretical foundation or maturity of AI domain. The research outcomes hold significant value for a wide range of libraries, from those hesitant to adopt AI due to ethical concerns to those actively experimenting with AI technologies. The article uniquely recommends further research at the intersection of library AI-driven digital transformations, cocreation, monitoring and evaluation (M&E), adoption models, and AI experimentation policies that ethically balances library innovation focus and data privacies.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it