Looking for DT Photo?
Select Page

Artificial Intelligence in Cultural Heritage Digitization: Practical Use Cases from the Field

by | Jan 5, 2026 | Artificial Intelligence, Case Studies, DT Roundtable

Artificial Intelligence in Cultural Heritage Digitization: Practical Use Cases from the Field

Artificial intelligence is rapidly moving from experimentation to production across libraries, archives, museums, and cultural heritage institutions. While the promise of AI is often discussed in abstract terms, its real value emerges when it is applied thoughtfully to specific digitization challenges—at scale, and with institutional goals in mind.

At Digital Transitions’ recent Cultural Heritage Roundtable, practitioners from across the field shared concrete, working examples of how AI is already improving digitization workflows, metadata quality, and access to collections. These sessions highlight a clear takeaway: AI is most effective when used as an assistive tool that enhances professional expertise, not replaces it.

From Capture to Context: Where AI Fits in Digitization Workflows

High-quality digitization has always required precision, consistency, and trust in the output. AI does not change those fundamentals—but it can dramatically reduce the time and labor required to move from raw capture to usable, discoverable digital assets.

During DT’s 2025 roundtable, we were shown three case studies for AI that proved to be highly effective:

  • Metadata Creation & Enrichment

  • Transcription & Text Extraction

  • Image Analysis & Intelligent Cropping

Each of these use cases supports a common objective: making large collections searchable, accessible, and sustainable without overwhelming limited staff and budgets.

AI-Assisted Transcription and Structured Metadata

Michelle Gollehon, Utah Historical Society

At the Utah Historical Society, AI is being used to bridge the gap between physical reference collections and digital access. Thousands of historical index cards—previously accessible only on site—are being digitized using a Digital Transitions Versa system and then processed with AI-powered transcription and data structuring tools.

The workflow combines high-resolution image capture with AI transcription to convert handwritten and typed cards into structured, searchable metadata. Crucially, human review remains part of the process, ensuring accuracy and contextual integrity while dramatically accelerating throughput. This approach allows small teams to scale projects that would otherwise take years to complete.

Watch the full presentation video here.

Smarter Metadata at Scale

Siera Erazo, NASCAR Hall of Fame

For institutions managing large photographic collections, metadata creation is often the biggest bottleneck. At the NASCAR Hall of Fame, AI is being used to extract targeted data points—such as subjects, events, and contextual details—from motorsports photography collections that span decades.

Rather than attempting to automate all descriptive work, AI is applied selectively to support cataloging priorities defined by curators and stakeholders. This targeted approach improves consistency, supports internal discovery, and reduces manual workload while maintaining curatorial oversight and ethical standards around bias, copyright, and transparency.

Watch the full presentation video here.

Image-Level Intelligence and Workflow Efficiency

Doug Peterson, Digital Transitions Head of R&D

AI is also being embedded directly into digitization workflows. During the roundtable’s technical track, Digital Transitions demonstrated how AI-driven image analysis—such as intelligent cropping and object detection—can improve productivity without compromising capture standards.

By combining large capture fields with AI-assisted cropping, multiple objects can be digitized in a single exposure and automatically separated downstream. This approach reduces handling time, supports consistent output, and aligns with FADGI-compliant imaging practices when paired with stable hardware and controlled lighting environments.

Watch the full presentation video here.

Best Practices: Using AI Responsibly in Heritage Work

Across all three presentations, several best practices emerged:

  • Start with a clear problem. AI works best when applied to a specific, well-defined challenge.
  • Keep humans in the loop. Review and validation remain essential for quality and trust.
  • Be transparent and ethical. Privacy, bias, and provenance must be considered from the outset.
  • Integrate, don’t bolt on. AI should complement existing digitization standards and workflows.

When approached this way, AI becomes a force multiplier—allowing institutions to do more with the same resources, without compromising professional standards.

Learn More from the Roundtable

These examples represent just a portion of the AI-focused discussions from Digital Transitions’ recent Cultural Heritage Roundtable. Full session recordings explore each use case in greater depth and provide practical insights for institutions considering AI adoption today.

To explore how Digital Transitions supports AI-ready digitization workflows—from capture hardware to software and services contact us or watch the full roundtable sessions here.

AI is no longer a future consideration for cultural heritage digitization. As these projects demonstrate, it is already delivering measurable value—when used with intention, expertise, and care.