Reframing Museum Intelligence: A Survey of Large Language Models in Museums

Abstract

Large Language Models (LLMs) have recently emerged as a transformative technology for museums, enabling a shift from static content delivery toward interactive and intelligent systems. However, existing studies and deployments remain fragmented, with a lack of systematic analysis of how LLM capabilities align with the unique goals, constraints, and values of museum contexts. This gap motivates the need for a unified survey that views LLM-based systems not only as technical tools, but also as components embedded within cultural, educational, and institutional settings. This survey provides a comprehensive overview of how LLMs are being integrated into museums, covering both interactive intelligence for visitor-facing applications and backstage intelligence that supports curatorial and governance processes. By synthesizing prior work and emerging practices, we aim to offer valuable insights into research challenges, future directions, and practical considerations for developing LLM-enabled museum systems that are robust, responsible, and aligned with long-term cultural heritage objectives. CCS Concepts: • General and reference → Surveys and overviews; • Social and professional topics → Cultural characteristics; • Human-centered computing → Empirical studies in interaction design; • Computing methodologies → Artificial intelligence.

The authors' abstract, as published at the source. Journal on Computing and Cultural Heritage, 2026 · DOI ↗