Standardizing Access to Sensitive Data with Machine-Actionable Access Conditions

The reuse of research data is widely acknowledged as a driver of scientific progress. However, access to sensitive datasets—such as those containing personal, health-related, or proprietary information—remains a significant challenge. In the Netherlands, researchers and data holders face a fragmented and inconsistent landscape of data access policies. Each major data provider, from Statistics Netherlands (CBS) to RIVM and various health registries, enforces its own procedures, legal terms, and review processes. This lack of standardisation creates confusion, lengthy delays, and administrative burdens for all parties involved. Although many data holders are willing to share sensitive datasets under specific conditions, the current system fails to clearly communicate these access conditions in a structured and machine-readable way, as mandated by the Accessibility principle (the A in FAIR). Metadata often lacks critical details about eligibility, permitted uses, and user obligations, making it difficult for researchers to judge whether and how they can gain access. This presentation offers a vision for improving access to sensitive data through the standardisation of access conditions using a machine-actionable language such as the Open Digital Rights Language (ODRL). Based on community consultation with Dutch researchers and data holders, the authors argue that encoding access policies in a digital, interoperable format—supported by user-friendly tools and a coordinating broker service—would lead to a more efficient, secure, and FAIR-compliant system. Such an approach could significantly increase the availability and responsible reuse of sensitive data for research.