Abstract: Staunton et al. (2021) argue that the increasing reliance on repositories for data exchange complicates the ability to monitor research practices, environments, techniques, and technologies. This challenge in tracking data exchanges undermines the quality and completeness of metadata, as well as the appropriateness of the data for potential application (p. 115; see also Rajesh et al., 2021). I apply two competing epistemic attitudes to science: the Book of Truths (BOT) and the Book of Conversations (BOC) (Davis-Stober et al., 2025). These attitudes were developed to address replication and the normative role of the peer-reviewed literature. Here, I apply them to risk variability. Across disciplines, norms for acceptable levels of uncertainty—or sigma—when annotating metadata vary significantly. For example, genome-wide association studies (GWAS) require highly stringent significance levels to minimize false positives due to the sheer number of comparisons involved. However, when GWAS data are applied in diagnostic medicine, the acceptable level of uncertainty may be relaxed, especially when the goal is to initiate treatment. This difference exemplifies that scientific domains prioritize distinct epistemic values when assessing whether a hypothesis is true. By highlighting these attitudes, I aim to illuminate how epistemic attitudes shape data exchange and complicate the ethical annotation of metadata across disciplines.

Abstract: The sharing of personal data in research is of high importance, as it is the basis for reproducibility and allows to reuse valuable datasets. In his talk, Evgeny Bobrov will give an introduction to the sharing of personal research data. After presenting basic concepts of data protection, he will discuss the steps which can be taken on the levels of consent, data anonymization, and data deposition to allow the sharing of personal data.

Abstract: Both artificial intelligence (AI) and citizen science hold immense potential for addressing major sustainability challenges from health to climate change. Alongside their individual benefits, when combined, they offer considerable synergies that can aid in both better monitoring of, and achieving, sustainable development. While AI has already been integrated into citizen science projects such as through automated classification and identification, the integration of citizen science approaches into AI is lacking. This integration has, however, the potential to address some of the major challenges associated with AI such as social bias, which could accelerate progress towards achieving sustainable development.

Abstract: Since the release of ChatGPT and other generative AI applications, research institutions as well as various stakeholders such as research funders and publishers have been discussing how the use of AI in research should be dealt with from the perspective of research integrity. While a consensus on some matters was quickly reached, other issues are still being debated. This has led to a rather heterogeneous policy landscape, which will likely remain in flux due to the fast-paced nature of AI. Additionally, a recent survey by Wiley found that nearly two-thirds of respondents indicated that the lack of clear guidelines prevents them from using generative AI to the extent that they would like.

In this presentation, participants will get an overview of the status quo of the current debate on AI and research integrity. The presentation has a practical bend covering how to disclose different use cases of AI in publications and what to keep in mind when it comes to working with AI in research, for example regarding peer review, writing grants or using AI-generated images. The presentation will also raise awareness about further issues that are still under debate such as recommendations on which tools to use, questions of access and ethical aspects.

Abstract: Machine learning is transforming scientific research by offering efficient, data-driven solutions for challenges across various fields. From the prediction of molecular properties to protein folding, ML accelerates problem-solving and opens up new avenues for exploration. However, there are specific challenges associated with applying machine learning to scientific research compared to other domains. Data availability can be limited and problems often possess inherent structure due to underlying physical laws. Incorporating these laws into machine learning models has the potential to enhance performance but may also have adverse effects. Likewise, different machine learning techniques are more suitable for particular scientific challenges. Open data and models are essential in addressing these aspects, as they improve data availability, promote reproducibility and accelerate model development. This lecture will explore the opportunities and challenges of integrating machine learning into scientific research with a focus on applications in chemistry and biology.

Abstract: Effective data infrastructures and competencies are essential for cultivating a collaborative data culture and driving AI progress. This lecture will explore the critical relationship between technological frameworks and the skills necessary for researchers to manage, share, and use data efficiently. We will examine the foundational role of robust data infrastructures in supporting AI initiatives, focusing on the need for large-scale, machine-interpretable data. Key topics will include transparency, reproducibility, and fairness in AI, along with the importance of making essential resources—publications, data, models, and code—accessible, interconnected, and scalable. Participants will gain insights into building a sustainable data ecosystem that enhances research outcomes and maximizes the potential of data in AI-driven projects.

Abstract: How can FAIR principles support the actual discovery and reuse of research data? This session invites participants to explore how metadata quality, repository design, and documentation practices influence the findability and reusability of datasets. Through guided exploration and shared reflection, we’ll examine published examples and consider what supports meaningful reuse—and what gets in the way. Participants will have the opportunity to experience typical challenges first-hand and use these observations to discuss relevant metadata standards, common issues, and the broader role of infrastructures in FAIR data practices. The session aims to connect practical engagement with critical insight and collaborative discussion.