From Curation to Context: Why Research Guides and Delivery Need a Rethink in an AI-Driven Academic Environment

By Tamir Borensztajn, Founder, WyderNet

I started my career as a librarian and earned my MLIS at Simmons University in Boston. In those early days, I spent a lot of time supporting students who arrived with a familiar mix of urgency and uncertainty about work that was due. Time was always a factor, but so was the challenge of quickly matching their specific academic need to the most relevant resources. That question of how to optimally support students and researchers in the moment of need has stayed with me ever since.

Libraries are exceptionally good at curation. By curation, I mean the work that subject and instruction librarians do to identify, evaluate, and recommend high quality resources in alignment with specific disciplines, courses, and research needs. This includes collection development, subject guides, reading lists, and research pathways that are carefully assembled to guide students and researchers toward authoritative, relevant material. Enormous professional judgment and effort go into selecting, structuring, and maintaining these resources. Yet in many institutions, an uncomfortable truth remains. A large portion of this work is underused or unseen altogether. This is not a failure of librarianship. It is a failure of scale, delivery, and visibility in the current library technology stack.

Artificial intelligence is forcing libraries to confront two long-standing challenges. The first is how subject and instruction librarians curate and recommend resources, drawing on deep disciplinary knowledge to guide students and researchers toward high quality material, and how that expertise can be extended at scale and kept current as courses, collections, and research directions change. The second is how those resources are actually delivered into the workflows where students, researchers, and faculty already do their work. If we do not address both, we risk building remarkable guides for audiences that never find them.

Curation Does Not Scale the Way Academic Work Does

Most guides are still created through manual effort. A librarian may review a syllabus, then select relevant materials, to structure into a guide. This produces high quality outcomes, but it is labor intensive and fragile. Courses change. Assignments change. Research directions change. Collections change. Yet guides often do not. They quietly drift out of sync with the academic reality they are meant to support.

At the same time, academic work has become more specific and more time bound. Students are not just studying a discipline. They are working on a particular problem for a particular assignment due on a particular date. Researchers are not just exploring a field. They are pursuing narrow questions under pressure to publish. Faculty are not just teaching subjects. They are designing learning experiences across multiple platforms with defined outcomes. There is a structural mismatch between how libraries curate and how academic work actually happens.

Artificial intelligence gives us a chance to close this gap. Academic artifacts such as syllabi, assignment briefs, research proposals, and project descriptions are not administrative paperwork. They are rich signals of intent. They tell us what someone is trying to accomplish. AI can be tuned to read these artifacts and help surface relevant resources across licensed collections, open access materials, institutional repositories, and curated digital assets. Not to replace librarians, but to extend their reach. Not to automate judgment, but to reduce the mechanical burden of matching context to content. Librarians remain firmly in control.

Availability Is Not the Same as Presence

Even if we perfect curation, we still face a harder problem. Delivery.

Libraries are very good at making resources available. They are far less successful at making them present. ‘If you build it, they will come’ is a mindset that worked in a world where the physical library was the primary gateway to knowledge. In a world where answers live in every pocket, libraries must now reach beyond the stacks and into the environments where academic work actually happens.

Students and researchers increasingly work in learning platforms, document editors, collaboration tools, and AI interfaces. This is where writing happens. This is where analysis happens. This is where decisions happen. In this reality, assuming that users will proactively seek out guides is optimistic at best.

The result is an awareness gap. Libraries provide access to high quality, trusted resources, but users either do not know they exist or encounter them too late to be useful. By the time a student discovers a guide, the paper is half written and momentum has already formed elsewhere. In these moments, convenience wins. Speed wins. Whatever is closest at hand wins. This is not a marketing problem. It is a structural one.

If libraries want to remain central to the academic process, they need to show up inside the workflow, not alongside it.

Intent Is Not a Mystery in Academia

One of the great ironies is that academia is full of explicit intent signals. Syllabi, assignments, deadlines, research proposals, and learning outcomes are structured declarations of purpose. They tell us what a student is expected to do, what a researcher is trying to answer, and what a faculty member is designing a learning experience to achieve.

Other industries spend millions trying to infer intent. Academia already has it. The challenge is using it.

AI can help interpret these signals and enable timely, contextual delivery of resources. A student uploads an assignment brief and relevant materials are surfaced. A researcher drafts a proposal and key literature appears. A faculty member designs a course and aligned resources are suggested. This is not about pushing content. It is about supporting work.

A Necessary Shift

This is not simply an evolution of existing practice. It is a shift in posture and paradigm.

Artificial intelligence gives us the ability to understand context and urgency in ways that were not previously possible. It allows us to move from generic curation to contextual guidance. From resources waiting to be found, to expertise delivered when it matters. From static guides to living, responsive support that can stay current as academic needs evolve.

But the principle remains unchanged. Librarians are the authority. They define quality. They define relevance. They define trust. AI should serve as connective tissue between academic intent and curated knowledge, helping libraries place the right resources in front of the right people at the right time, inside the workflows where learning and research actually happen.

The shift is from availability to presence. From access to impact. If we are willing to embrace this change, libraries are not displaced by AI. They are strengthened by it.