Updates

We see AI more and more in our daily work. That is why the Allard Pierson Museum, together with the Culture and Media Working Group of the Dutch AI Coalition (NLAIC), organised a conference on AI within the heritage sector last Monday at the Singelkerk and the Allard Pierson Museum in Amsterdam.

The programme included both lectures and workshops. Among others, the HAICu project was pitched by Prof. Lambert Schomaker during the plenary session. For more information, see the full programme here (in Dutch).

The HAICu project was officially kicked off last Friday with a meeting for all stakeholders in Utrecht. Within the programme, which recently received a 10.3 million euro grant from the National Science Agenda, AI and Digital Humanities researchers, heritage professionals and interested citizens work together to use artificial intelligence tools to make heritage collections more accessible.

The HAICu consortium consists of various parties, including heritage institutions, colleges and universities. From the University of Groningen, the Jantina Tammes School of Digital Society, Technology & AI collaborates on the project, among others. During the meeting in Utrecht last Friday, the various work packages of HAICu were presented in more detail. There was also a panel discussion, where partners were challenged to ask provocative questions.

The heritage world is dealing with a seemingly incessant growth of data. Within HAICu, artificial intelligence tools are being developed to make collections easier and more searchable. Due to the extensive collaboration between heritage institutions, the concept of multimodality plays a major role within the project. With the help of generative AI, different collections are linked, such as Delpher's newspaper archive and Sound and Vision's audiovisual sources.

The great polyphony of different collections provides a multiplicity of perspectives, important at a time when people increasingly live in their own bubble. The wide range of sources, e.g. text, image and sound, helps to create a balanced narrative. An important part of the HAICu project are the 'Innovation labs'. These labs are set up so that a wide audience and specific target groups (including journalists) can test the tools being developed by the consortium's researchers.

This article was originally published at Jantina Tammes School of Digital Society, Technology and AI, University of Groningen, 6 February 2024.

Sound and Vision, the Geheugen van Nederland, Beeldbank Groningen: these days, all sorts of heritage collections are available online. They are crucial sources of information for a wide range of stories. But just imagine being able to browse all these collections to find the context you need for a good story simply by entering a single search term. This is what the HAICu research project hopes to achieve with the aid of artificial intelligence (AI) tools. The project is also meant to make a valuable contribution to the ongoing development of AI. Professor Lambert Schomaker is heading the operation, thanks to €10.3 million of funding that the Dutch National Research Agenda will make available over the next few years.

You can read the full article here.

The NWO is providing 10.3 million euros for the HAICu project. This project deploys artificial intelligence (AI) to make digital heritage collections more accessible. It allows users to more easily interpret events from different perspectives and assess them for authenticity. The project will run for six years.

The University of Groningen is the lead institution for the project. In HAICu, AI and Digital Humanities researchers collaborate with various partners and interested citizens on scientific breakthroughs to unlock, connect, and analyze extensive digital heritage collections.

The extraordinary challenges of cultural heritage provide a unique opportunity to push the boundaries of AI. Future techniques must be able to be used outside the laboratory, learn from as few examples as possible and continuously learn from users. These techniques must take into account the societal demand for accountable and explainable methods for creating multimodal narratives of our cultural heritage that extend beyond current major language models.