Problem Museum Collection Management (CM) is a complex task. The variety of sources, from field notes to digital files, presents challenges in establishing relationships between sub-collections, provenance research, and ensuring data integrity. Objectives Use of Artificial Intelligence (AI) to streamline these processes. Exploration of two key applications of AI: - AI for CM: automating the identification of links, detecting inconsistencies, and enhancing data quality across collections and databases. - AI for Storytelling: investigating the potential for AI-driven storytelling to enhance engagement with museum collections. - Making data available for source communities, researchers and the general public. The first part of the project will focus on developing and fine-tuning AI tools for CM. This will help address issues such as automating the reading of historical handwritten labels to fill gaps and improve the accuracy of collection records. The second part will focus on two Congolese collectors, Ngwe and N’Kele, who collected thousands of specimens, some of which were used to describe new species. We will organize a hackathon with students from Catholic University Leuven (KUL), using AI-supported tools to come up with a way to tell their stories and increase the visibility of their work. Finally the results will be published on online repositories for outreach to the source communities, researchers and the general public. Methodology - Identification of field notes, correspondence, photographs, and publications related to Ngwe and N’Kele, followed by digitisation if necessary. - Translation of correspondence and field notes written in Lingala through the linguistic department of RMCA. - Development and implementation of a proof-of-concept method for accurately recognizing handwritten text, including accession numbers, collectors, and localities. The AI will also help detect discrepancies, such as misattributions of specimens to collectors. - Organisation of a hackathon exploring AI-supported tools. - Use of AI to generate narratives about Ngwe and N’Kele’s contributions. These will be visualised and shared via the communication channels of the RMCA and KUL. - Publicly making available of the cleaned dataset on platforms like GBIF, contributing to the global biodiversity knowledge base. A PhD student will oversee the AI-part, while a master’s student will assist with training the AI. Volunteers will offer support with digitisation. For the storytelling, a research assistant, supervised by an innovation manager, will organise the hackathon. Outcomes - The project will result in a comprehensive dataset that ensures clear provenance for specimens collected by Ngwe and N’Kele. - A functional AI proof-of-concept will help identify and resolve inconsistencies in large datasets, improving quality control in collections. - Based on the results of the hackathon, the project will develop guidelines for using AI to create engaging narratives that highlight key collectors such as Ngwe and N’Kele. - The AI tools developed will be adapted to other collections within RMCA. The collaboration between the CM and Linguistics departments will deepen the understanding of local naming practices for species. Visualisations and narratives will be tested in collaboration with the Public Services and Communication Department. - This project aligns with Belspo and RMCA strategies by focusing on provenance research, synergies across departments, and the dissemination of data through open AI technologies and online platforms, thus adhering to FAIR data principles. The tools created will also assist museum staff in daily quality control processes. Conclusion By applying AI to collection management, this project will provide a robust solution for improving data quality and accuracy in museum records. Storytelling through AI will also complement current curatorial practices by unlocking alternative ways to valorise key collections.