The study synthesizes literature on opening the ‘black box’ of artificial intelligence (AI) for credible information retrieval systems (IRS), and the emerging roles for library and information science (LIS) professionals. A rapid review of literature was adopted for the study. Literature was retrieved from EBSCOHost and Proquest, complemented by Google Scholar, using three eligibility criteria of publication source, language and year of publication. The review found that the AI black box is opened by explainability, which enhances the transparency, fairness and credibility of the IRS. This reinforces users’ trust in the system based on their understanding of its internal workings. Thus, black box models adversely affect the deployment of IRS, demanding explainable AI in IRS. The study proposed emerging roles for LIS professionals (including advocacy for transparent models in IRS, algorithm education, facilitating open science and data curating), in opening the AI black box for credible IRS.
AliSAbuhmedTEl-SappaghS, et al. (2023) Explainable artificial intelligence (XAI): What we know and what is left to attain trustworthy artificial intelligence. Information Fusion99: 101805.
4.
AndrikopoulouARowleyJGeoff WaltonG (2022) Research data management (RDM) and the evolving identity of academic libraries and librarians: A literature review. New Review of Academic Librarianship28(4): 349–365.
Blatch-JonesAJRecio SaucedoAGiddinsB (2023) The use and acceptability of preprints in health and social care settings: A scoping review. PLOS ONE18(9): e0291627.
7.
Borrego-DíazJGalán-PáezJ (2022) Explainable artificial intelligence in data science. Minds & Machines32: 485–531.
8.
BurrellJ (2016) How the machine ‘thinks’: Understanding opacity in machine learning algorithms. Big Data & Society3(1): 205395171562251.
9.
CarabantesM (2020) Black-box artificial intelligence: An epistemological and critical analysis. AI & Society35: 309–317.
CostaABuccioEDMelucciM, et al. (2018) Efficient parameter estimation for information retrieval using black-box optimization. IEEE Transactions on Knowledge and Data Engineering30(7): 1240–1253.
12.
DaiSLiuWZhouY, et al. (2024b) Cocktail: A comprehensive information retrieval benchmark with LLM-generated documents integration. Findings of the Association for Computational Linguistics: ACL 2024, pp. 7052–7074. https://doi.org/10.48550/arXiv.2405.16546.
13.
DaiSXuCXuS, et al. (2024a) Bias and unfairness in information retrieval systems: New challenges in the LLM Era. KDD ’24, 25–29 August, Barcelona. https://doi.org/10.48550/arXiv.2404.11457.
14.
GiustiniD.ReadK. B.DeardorffA., et al. (2021). Health sciences librarians’ engagement in open science: A scoping review. J Med Libr Assoc., 109(4):540–560. https://pubmed.ncbi.nlm.nih.gov/34858084/.
15.
HassijaVChamolaVMahapatraA, et al. (2024) Interpreting black-box models: A review on explainable artificial intelligence. Cognitive Computation16: 45–74.
16.
HershW (2024) Search still matters: Information retrieval in the era of generative AI. Journal of the American Medical Informatics Association31(9): 2159–2161.
17.
HombaliP (2022) The role of librarians in research data management: Challenges and opportunities. International Journal of Research and Analytical Reviews (IJRAR)9(1): 359–364. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4422179.
18.
HuangY.HuangJ. X. (2024). Exploring ChatGPT for next-generation information retrieval: Opportunities and challenges. Web Intelligence, 22, 31–44.
IgbinoviaMODanquahMM (2025) Artificial intelligence algorithm bias in information retrieval systems and its implication for library and information science professionals: A scoping review. Technical Services Quarterly: 1–27. https://doi.org/10.1080/07317131.2025.2512282.
LiuNFLinKHewittJ, et al. (2024) Lost in the middle: How language models use long contexts. Transactions of the Association for Computational Linguistics12: 157–173. https://aclanthology.org/2024.tacl-1.9/.
23.
MasindeJChenJWambiriD, et al. (2021) Research librarians’ experiences of research data management activities at an academic library in a developing country. Data and Information Management5(4): 412–424.
24.
McGillSC (2024) Preprint pointers from a long COVID scoping review: Considerations for source selection and searching. Journal of the Canadian Health Libraries Association45(2): 88–97. https://pmc.ncbi.nlm.nih.gov/articles/PMC11485164/.
25.
ModibaM.NgulubeP.MaruthaN. (2022). Infrastructure for the implementation of artificial intelligence to support records management at the council for scientific and industrial research in South Africa. ESARBICA Journal, 41, 159–171. https://www.ajol.info/index.php/esarjo/article/view/240091
NdunguMW (2024) Integrating basic artificial intelligence literacy into media and information literacy programs in higher education: A framework for librarians and educators. Journal of Information Literacy18(2): 1–18.
OgungbeniJIObiamaluARSsemamboS, et al. (2018) The roles of academic libraries in propagating open science: A qualitative literature review. Information Development34(2): 113–121.
30.
Page MJ, McKenzie JE, Bossuyt PM, et al (2021) The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ372(71): 1–9.
31.
PedreschiDGiannottiFGuidottiR, et al. (2019) Meaningful explanations of black box ai decision systems. Proceedings of the AAAI Conference on Artificial Intelligence33(01): 9780–9784.
32.
PedreschiDGiannottiFGuidottiR, et al. (2019) Meaningful explanations of black box AI decision systems. Proceedings of the AAAI Conference on Artificial Intelligence33(1): 9780–9784.
33.
PetchJDiSNelsonW (2022) Opening the black box: The promise and limitations of explainable machine learning in cardiology. Canadian Journal of Cardiology38: 204e213.
34.
RojaKMounicaNSSAishwaryaM, et al. (2018) Black-box optimization for information retrieval through dynamic parameter. International Journal of Creative Research Thoughts (IJCRT)6(1): 2130–2133. https://www.ijcrt.org/papers/IJPUB1801342.pdf.
35.
Sawchuk S, Gillis L and MacLeod L (2023) Supporting reproducible research with active data curation. In: Thompson K, Hill E, Carlisle-Johnston D, et al. (eds) Research data management in the Canadian context: A guide for practitioners and learners. London, Ontario, Canada: Western University-Western Libraries, pp. 198-212. DOI: 10.5206/QQSG2445.
TangQChenJYuB, et al. (2024) Self-retrieval: Building an information retrieval system with one large language model. https://arxiv.org/pdf/2403.00801v1.
38.
TayeMM (2023) Understanding of machine learning with deep learning: Architectures, workflow, applications and future directions. Computers12(5): 91.
VellidoA (2020) The importance of interpretability and visualization in machine learning for applications in medicine and health care. Neural Comput & Applic32: 18069–18083.
41.
WaddenJJ (2022) Defining the undefinable: The black box problem in healthcare artificial intelligence. Journal of Medical Ethics48: 764–768. https://jme.bmj.com/content/48/10/764.
42.
XuHShuttleworthKMJ (2024) Medical artificial intelligence and the black box problem: A view based on the ethical principle of “do no harm”. Intelligent Medicine4: 52–57.
43.
YangG.YeQ.XiaJ. (2022). Unbox the black-box for the medical explainable AI via multi-modal and multi-centre data fusion: A mini-review, two showcases and beyond. Information Fusion, 77, 29–52.
44.
ZednikC (2021) Solving the black box problem: A normative framework for explainable artificial intelligence. Philosophy & Technology34: 265–288.
ZhaiCWibowoSLiLD (2024) The effects of over-reliance on AI dialogue systems on students’ cognitive abilities: A systematic review. Smart Learning Environments11(28): 1–37.
47.
ZhaoWXLiuJRenR, et al. (2023) Dense text retrieval based on pretrained language models: A survey. ACM Transactions on Information Systems42(4): 1–60.
