AI in Archival Practice: Large Language Models, Description, and Discovery

$250.00

Dates: August 2 - August 29

Credits: 1.5 CEUs or 15 PDHs

What happens when AI describes a collection, identifies subjects in historical records, or helps researchers find materials? Though familiar to archivists, these tasks raise new questions about how information is generated, interpreted, and presented.

Large language models (LLMs), which power tools such as ChatGPT, Claude, and Copilot, generate text, identify patterns, and organize information. In archives, they can suggest metadata, summarize documents, and support discovery. But convincing AI-generated descriptions may still be inaccurate, incomplete, or unsuitable for the records.

This course uses familiar archival activities to explore how LLMs work, where they help, and how to assess their limitations. Through practical examples, case studies, and course deliverables, participants examine AI-assisted description, metadata enrichment, discovery, and access, while considering provenance, context, representation, privacy, and professional judgment.

Learning Outcomes

By the end of the course, participants will be able to:.

  • Define LLMs and distinguish them from other AI technologies and traditional automation.
  • Identify AI applications in archival description, metadata, discovery, and access.
  • Evaluate the accuracy, context, provenance, and representation of AI-generated information.
  • Assess AI-assisted workflows with respect to archival standards, ethics, and human oversight.
  • Create a practical evaluation resource and relevant recommendations for their work.

Course Information

Session

Credits

1.5 CEUs or 15 PDHs

Registration dates

We accept registrations through the first week of classes, unless enrollment is full, and unless the class was canceled before it started due to low enrollment.

Course Description

What happens when AI describes a collection, identifies subjects in historical records, or helps researchers find materials? Though familiar to archivists, these tasks raise new questions about how information is generated, interpreted, and presented.

Large language models (LLMs), which power tools such as ChatGPT, Claude, and Copilot, generate text, identify patterns, and organize information. In archives, they can suggest metadata, summarize documents, and support discovery. But convincing AI-generated descriptions may still be inaccurate, incomplete, or unsuitable for the records.

This course uses familiar archival activities to explore how LLMs work, where they help, and how to assess their limitations. Through practical examples, case studies, and course deliverables, participants examine AI-assisted description, metadata enrichment, discovery, and access, while considering provenance, context, representation, privacy, and professional judgment.

Learning Outcomes

By the end of the course, participants will be able to:.

  • Define LLMs and distinguish them from other AI technologies and traditional automation.
  • Identify AI applications in archival description, metadata, discovery, and access.
  • Evaluate the accuracy, context, provenance, and representation of AI-generated information.
  • Assess AI-assisted workflows with respect to archival standards, ethics, and human oversight.
  • Create a practical evaluation resource and relevant recommendations for their work.

Lisa Cruces

Lisa Cruces is an independent consultant specializing in academic libraries, archives, and organizational development. Previously, Cruces has held positions at the Harry Ransom Center, University of Notre Dame, the University of Houston, and Texas State University, as well as advisory roles for non-institutional community archives and BIPOC collectives. Cruces completed her master of science in Information Studies at the University of Texas at Austin with a focus in archives and special collections and is currently pursuing her MBA in Human Resource Management at Texas State University. Her research focuses on assessing diversity, equity, inclusion, and accessibility in academia and workplace settings.

How to Register

To enroll yourself or other participants in a class, use the “Register” button that follows the description of each course. If the “Register” button does not show up, try loading the page in a different web browser. Contact us if you have technical difficulties using our shopping cart system or would like to pay for an enrollment using another method. On the payment page in the shopping cart system, there is a place to add notes, such as the names and email addresses of participants you wish to enroll. We will contact you to request this information in response to your processed payment if you do not include it in the “notes” field. Prior to the start of the workshop, we will send participants their login instructions.

Payment Info

Our shopping cart system allows you to pay with a credit card or with PayPal.

Alternatively, if it is an institutional payment, we can arrange to invoice you. Contact us by email, and we can make arrangements to suit your institution's business processes.

Special Session

Please contact us to arrange a special session of this class for a group of seven or more, with a negotiable discount, or to be notified when it is next scheduled.

Reviews

There are no reviews yet.

Be the first to review “AI in Archival Practice: Large Language Models, Description, and Discovery”

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

This site uses Akismet to reduce spam. Learn how your comment data is processed. (opens in a new tab)