Launching the Registry of Good Practices in AI-enabled Science

SCIANCE is launching the Registry of Good Practices in AI-enabled Science, which is a growing collection of practical examples demonstrating the effective and responsible use of AI across different scientific disciplines and research settings. The Registry is designed to capture the practical experience of the research community and identify lessons that can be shared across fields.

The first wave saw 107 nominations from the expert community, showcasing various approaches to AI-enabled research, such as the use of AI in data-intensive analysis and experimental workflows, and new forms of automation, collaboration, and scientific decision-making. The first selected cases are now available on the RAISE Digital Hub, with more to be added over time.

The Registry is designed to do more than simply collect examples; it is intended to help build an evidence base on what good AI-enabled science looks like in practice. SCIANCE is particularly interested in answering three questions:

Where in the research workflow is AI used, what role does it play, and how does it interact with researchers, data, tools and established scientific methods?

Which combinations of skills, infrastructure, data, organisational support, governance, collaboration and community standards make effective AI use possible?

How does the use of AI contribute to scientific discovery, research quality, productivity, reproducibility, access to knowledge, or the ability to address research questions that were previously difficult to tackle?

These questions complement the SCIANCE framework, which sets out good practice guidelines for different stages of the research workflow. The framework considers dimensions including impact, productivity, transparency, trustworthiness, openness, transferability, ethical awareness and frugality.

The first selected cases are now available on the RAISE Digital Hub, offering practical examples of AI-enabled research. More cases from the initial selection will be added progressively as the Registry develops.

The Registry aims to become an ever-growing community resource that documents emerging practices, supports interdisciplinary learning and promotes the responsible, effective and scientifically meaningful use of AI. The aim is to identify transferable lessons that will inform researchers, research organisations, research infrastructures and the wider European AI-in-Science community.

Explore the first cases on the RAISE Digital Hub – with more good practices coming soon.