Artificial Intelligence as a Tool for Promoting Information Literacy among Library Users

Synthesis: Offem and Arikpo argue that information literacy instruction is no longer sufficient on its own, because users now face machine-generated content at a scale that manual search-and-evaluate skills cannot address, and that AI should therefore be treated as an instrument for delivering information literacy rather than only as a new topic to teach about. Their proposal runs through intelligent retrieval, personalized and adaptive learning, chatbots and virtual reference, and automated fact-checking, each carrying part of the instructional load and extending support to remote users. The same technologies, they concede, produce fabricated references, dataset bias, privacy loss, a wider digital divide and passive dependence on machine output. Their resolution is to fold AI literacy into information literacy as a complementary component, with librarians rather than technologists leading the integration.

This page summarizes an opinion paper. It reports no new data: there is no methods section, sample, instrument or result, and every empirical claim in it is cited from other work or asserted.

Key Findings

1. The authors describe the paper as an "opinion paper" and a "scholarly opinion," and it contains no methods, sample, instrument, analysis or numeric result.
2. Their argument is structural: AI has become embedded in how libraries create, organize, retrieve and disseminate information, so information literacy instruction cannot remain a set of manual search-and-evaluate skills.
3. They propose using AI as a delivery mechanism for information literacy through intelligent retrieval systems, personalized and adaptive learning, chatbots and virtual reference, and automated fact-checking.
4. They claim these tools extend library support to remote users around the clock, an assertion not tested in the paper.
5. They name the risks on the same terms: hallucinations and fabricated references, bias encoded in training data, personal-data harvesting, a widening digital divide and passive dependence on machine output.
6. Their proposed resolution is to integrate AI literacy into information literacy instruction as a complementary component, teaching users to interrogate AI output rather than accept it.
7. They position librarians as indispensable to that integration, framing AI as a complementary tool and the profession as the guarantor of ethical, transparent and responsible use.
8. Every technology claim is attributed to other studies, including cited work by Wetzel and Kani (2024) on generative AI in information literacy instruction.

An argument about where instruction lives

The paper's contribution is a design claim, not a finding: if the information environment is now partly machine-generated, then the tools that generate it can also carry instruction. That reframing is worth taking seriously even without data, because it changes who is responsible for information literacy. In the traditional model the librarian teaches a skill and the student applies it elsewhere; in the model proposed here the library builds the retrieval and verification layer into the service itself, and the teaching happens where the search happens.

The weakness is that the proposal is stated at the level of capability rather than design. "Intelligent retrieval systems" and "automated fact-checking" are categories, not specifications, and the paper does not say what a library would build, buy or evaluate. A reader who accepts the argument still does not know what to do on Monday, which is the gap between an opinion paper and a framework.

The risks are the stronger half

The risks section is more concrete than the benefits section, and more useful. Fabricated references are a specific, checkable failure that instruction can address; bias in training data is a property of the tools a library selects; data harvesting is a procurement question. Each maps onto an existing library responsibility — collection judgment, vendor review, privacy policy — which is a more defensible basis for the authors' conclusion than the productivity claims.

That asymmetry is worth naming. The paper argues that AI should be folded into information literacy because it threatens the competency the library teaches. Read that way, the case does not depend on the tools improving outcomes; it depends on the tools changing what users encounter, which is true regardless of whether the promised benefits arrive.

What this means for practice

• Treat the proposal as a design brief, not evidence. Nothing here shows that AI-delivered instruction works; it argues that the delivery channel has moved.
• Start with the checkable risk. Fabricated references and unsupported claims are failures a short verification routine can catch, and they are the most defensible reason to teach AI evaluation.
• Put the librarian in the loop by design. The paper's strongest institutional claim is that librarians, not vendors, should own the integration; that requires a named role in whatever service is procured.
• Do not cite this for outcomes. It reports no data and should not be used as evidence that AI improves information-literacy learning.

Limitations

• The paper is an opinion piece with no empirical data, no methods, no sample and no analysis; it cannot support claims about effectiveness.
• The benefits it describes are asserted or cited from other work rather than demonstrated, and the technologies are named at the level of category rather than specification.
• The conclusion contains at least one overreach — that the paper "demonstrates" AI can enhance engagement and learning outcomes — which the design cannot support.
• No country, institution or library context is specified, so the recommendations are not tied to a setting or a resource level.
• No DOI is printed; the article is identified by its journal and volume only.
• The reference list contains at least two verified misattributions: the 2024 review it credits to Long and Magerko is Pinski and Benlian's, and Ng et al.'s "Conceptualizing AI literacy" is dated 2024 in an education-technology journal when it appeared in 2021 in Computers and Education: Artificial Intelligence. Several other citations are unrelated to the sentences they support.

Connected Concepts

• Information and Media Literacy
• Libraries
• Librarians
• AI Literacy
• Digital Divide
• Generative AI

Citation

Offem, C. O., & Arikpo, S. O. (2026). Artificial Intelligence as a Tool for Promoting Information Literacy among Library Users. International Journal of Jurisprudence, School Management and Civic Education, 1(1), 282–295.