AI in Libraries: Inclusive Innovation or Threat to Traditional Learning?

Recently, my job hosted a seminar called Library Staff AI Exchange: Critical Perspectives & Inclusive Design. For context, I’m in my final year of my MLIS, and I’m working as a Library Assistant for a major university library. The presentation I found most compelling was titled Designing for Every Learner: Using Campus AI to Build Inclusive Learning Experiences. The presenter is a library intern from my parent organization’s School of Information, and I was interested to see how an outside perspective would suggest we implement AI into our space. The majority of the presentation was about creating good prompts and how to use different Gemini (Google’s AI platform) tools to accommodate patrons and students. The acronym PARTS is a simple way to create a good prompt: 

  1. Persona: is there a specific professional voice this should come from? 
  2. Audience: who is this for? 
  3. Recipient: what should this accomplish? 
  4. Tone: what should the reader do/feel/know after reading the AI output?  i.e. specify that the output should be in plain language 
  5. Structure: describe a document/general shape you liked the style of so that the response is like that. Be as precise as possible, as AI has issues with processing verbosity.   

These were the suggested uses of Gemini in the library: 

  • Gemini Notebook: a source-based notebook that can generate multiple modalities, only based on sources you provide it. It can also turn those sources into other formats, like an audio overview, a mind map, or flashcards.
  • Suggested use in the library: 
  • Upload digitized letters to create a chronological timeline of events, and other archival work. 
  • Upload an academic or research paper, and ask the AI to summarize it in simpler terms. 
  • Gemini Live: a real-time voice conversation with Gemini, so you talk instead of type. You can share your camera, so you can point at something and ask it to explain it.
  • Suggested use in the library: 
  • Patrons reading dense text on a reader can point their phone at the camera and ask the AI to decode it. 
  • Gemini Gems: A gem is a set of instructions you write once and reuse, instead of retyping the same detailed request every time.
  • Suggested use in the library: 
  • You can use a gem to create a LibGuide that prioritizes digital accessibility.

Reflecting on the presentation, I thought the PARTS acronym was very useful, especially if a user is new to AI and wants to use it for a specific purpose. However, I’m apprehensive about the suggested Gemini uses. For the Gemini Notebook example, I wondered what using it would mean for our archivists. Our institution has a huge archival collection, and our archivists are extensively and thoroughly trained to handle specific projects. While AI can’t replace the physical handling of materials, I’m also unsure if it’s accurate enough to replace the work of a trained archivist. In regard to the example of using Notebook to simply summarize an academic paper, I thought about what this would mean for the future of the Research 101 class that our librarians teach to our students every semester. The whole point of the class is to teach students, catered to their specific subjects like sciences or social studies, how to read research/academic papers on their own, and how to write their own research paper for their field of study. If we pivot to the use of AI, students won’t have the opportunity to learn how to do this for themselves, and in turn the class will become obsolete. While I think these ideas for AI implementation into the library sound great, especially for accessibility purposes, I’m unsure how they could realistically be implemented into the library without causing major disruption to our existing procedures. Perhaps we should consider using AI as a tool to use in the process of project creation, instead of using/viewing it as a replacement for our work. 

Leave a Reply

Your email address will not be published. Required fields are marked *