The European Commission (EC) intends to establish a digital infrastructure for advanced materials called the ‘Materials Commons’. This initiative intends to enhance the design, development, and testing of advanced materials. A key aspect of the platform will involve the use of artificial intelligence tools, which can potentially be used for optimized hazard and risk assessment of advanced materials. In this way, the Materials Commons initiative may substantially contribute to the development of safe advanced materials. The initiative may benefit from ongoing efforts to utilize AI and ML for predicting the toxicity of nanomaterials and advanced materials.

NanoCommons – a European digital infrastructure for advanced material innovation

This year, the EC announced its intention to establish a sustainable digital infrastructure for research and innovation in advanced materials, called the ‘Materials Commons’. This initiative aims to accelerate the design, development, and testing of new advanced materials in a controlled environment, using artificial intelligence (AI) tools such as machine learning (ML) techniques. The Materials Commons is intended to be a reliable platform for all stakeholders to share information about the innovation process, and the data on the platform are to follow the FAIR principles: Findable, Accessible, Interoperable, and Reusable. In this way, the Materials Commons also provides an opportunity to uncover data about safety and sustainability that are easily accessible for any user.

Use of machine learning in nanotoxicology

Over the past decade, considerable efforts have been made to establish databases containing FAIR safety and sustainability data on nanomaterials. Despite these efforts, the application of AI for assessing the safety of nanomaterials remains challenging due to a lack of high-quality data. A recent critical review of recent efforts to use ML in nanotoxicology highlighted the need for large, curated databases specifically tailored for modelling in nanotoxicology. The researchers believe that with improved data mining and ML techniques, it is possible to develop more predictive and reliable models. By combining ML with expert knowledge in nanotoxicology, we can gain a better understanding of the mechanisms of toxicity of advanced materials, which is crucial for effective risk assessment. Additionally, ML can assist in establishing safe-by-design criteria for the development of safer, more sustainable advanced materials, including nanomaterials.

Reflections by RIVM

The EC’s Materials Commons initiative, along with efforts to use ML in the safety assessment of nanomaterials, highlights the increasing impact of AI in society. In early 2024, the EC established harmonized rules on artificial intelligence. While the EC acknowledges AI’s potential to bring economic, environmental, and societal benefits, it also recognizes that AI may pose risks to public interests and fundamental rights, which could lead to material or immaterial harm.

As part of the Materials Commons initiative, there will be a focus on using AI and ML for optimized hazard and risk assessment. This is a positive development, as it implies that hazard and risk assessment can receive the necessary attention from the outset of materials innovation. As long as the digital infrastructure of the Materials Commons is not further clarified, it remains uncertain how much it will contribute to hazard and risk assessment of advanced materials. The Materials Commons may benefit from ongoing initiatives aimed at leveraging AI and ML to predict the toxicity of advanced materials.

Read the Dutch summary: Kunstmatige intelligentie en geavanceerde materialen