JSTOR Integrations

JSTOR Seeklight is an AI-assisted collections-processing tool built as part of JSTOR Digital Stewardship Services. It helps archivists and librarians process digital collections faster while preserving professional oversight and archival values. Seeklight automatically generates descriptive metadata based on a Dublin Core–compatible schema, produces transcripts for typed, handwritten, or mixed-format documents, and groups related files into single archival items when appropriate, all handled via a cloud-hosted interface. Confidence scores accompany every AI-generated metadata field, and users can review, edit, or reject outputs; tags mark items as “AI-Generated,” “Reviewed,” or “Edited,” ensuring transparency and human supervision. It supports project-level summaries in the format of traditional finding aids, making collections easier to search and discover.