Guiding AI: Three Choices for the Cultural Sector

AI is here, and it delivers results. Research by TNO (opens in new tab) shows that AI accelerates and simplifies work processes within organizations. At DEN, we see internally that the use of AI leads to productivity gains: routine tasks take us less time. ChatGPT now has more than 900 million weekly active users worldwide (opens in new tab), and research by the Kunstenbond shows that 59% of surveyed creators already use generative AI regularly or occasionally. AI is thus already part of daily practice, including in the cultural and creative sector. 

The question is therefore not only how we will use AI but how we ensure that AI strengthens the sector. From that conviction, DEN sees three tasks for the future. 

1. Ensure AI literacy among cultural professionals  

Generative AI is transforming cultural organizations and is therefore an important skill for all cultural professionals. AI can support but does not replace craftsmanship or imagination. 

This is why sector-wide AI literacy is essential. Cultural professionals must understand how AI works, precisely to consciously and critically determine where it can help and where collaboration with creators and colleagues remains central. This choice can only be made if you know what AI can and cannot do én experiment with it. 

Primarily, this means upskilling all cultural professionals so they can consciously choose how to use AI, for example, to make texts accessible, create strategic plans, take over routine tasks, or act as a sparring partner in the creative process. The Werktuig PPO funding scheme can help with this. We see in practice that the scheme for digital transformation is gaining traction and, within an organization-wide approach to AI, can be particularly effective. With this vision of utilizing AI, we see the task alongside protection differently: engage with AI organization-wide consciously. 

2. Make ‘protect and utilize’ the foundation of the task force 

The AI transition is too large to tackle alone. DEN is eager to lead the task force 'AI and Culture.' Essential for a task force approach is that participants embrace the broad interpretation of 'protect and utilize' described above as the foundation. Only then can we work on solutions instead of getting stuck in contradictions. 

The cultural sector largely consists of small organizations with limited size and budget. Not every organization needs or can reinvent the wheel on its own. In our opinion, this calls for thinking about shared services. By jointly investing in AI infrastructure, knowledge, and facilities, the sector can innovate faster and more safely. This means sector-wide AI facilities, shared licenses, or exploring European or sustainable AI solutions, such as Mistral and GreenPT, for organizations that consciously choose alternatives to American technology. This way, we not only build knowledge and collaboration but also digital autonomy. DEN sees a task for the task force here: jointly investing in these facilities so the entire sector can benefit. 

3. A representative AI model requires a new revenue model 

Good AI starts with good data. Many AI models today are primarily trained on freely available internet data, such as Reddit and Wikipedia. This leads to uniformity, cultural bias, and limited representation of languages, stories, and cultural perspectives. 

The cultural sector can make a unique contribution here. The Netherlands has an enormous wealth of cultural data: from museum collections, archives, and libraries to video recordings of theater performances, music, festivals, photography, and digital art. When this data is made available, it can contribute to AI systems that are more representative, diverse, and culturally rich. This wealth and diversity of sources contribute to AI systems that are higher quality and more representative. It is therefore important that these sources are used to train AI models. Consider Studio Erwin Olaf, which consciously chose to make Erwin Olaf’s work available for AI training (opens in new tab). Not to replace the unique artistry but to keep his vision alive and continue to bring his art to global attention. The AI era invites us to think about new revenue models. 

Conclusion 

The cultural sector has always shown that it does not merely undergo technological change but also gives it meaning. This is no different now. AI is changing the work within the cultural sector and its methods. The choice now is what role we all want to play in that change. 

DEN chooses to actively assist the sector in this. Through an organization-wide approach, organizing collaboration, and questioning existing systems. Together with the sector, we also give direction to AI. 

Ultimately, the future of AI is not determined by technology. It is determined by the choices we make today. If we as a sector make those choices consciously én actively engage with them, we build a future in which AI does not replace the cultural sector but strengthens it.