idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
idw - Informationsdienst
Wissenschaft
Artificial intelligence is transforming research, from automated literature reviews to advanced data analysis. Alongside these opportunities, challenges arise. How can transparency be ensured when AI tools are used? Who carries responsibility for AI-generated content? What role should funders and policymakers play in building trustworthy infrastructures?
Dr. Angela Holzer, Programme Director at the German Research Foundation (DFG), explains in this interview why openness is vital for trustworthy AI, outlines the risks of neglecting transparency, and highlights how funding bodies can shape the long-term conditions for responsible AI in research.
How can openness, transparency, and collaboration help make artificial intelligence fairer and more trustworthy? This question is at the heart of the Berlin Summer School on Artificial Intelligence and Society 2025. From 15 to 18 September 2025, about 50 early-career researchers in computer science and the social sciences will meet at Berlin’s EUREF Campus. The Weizenbaum-Institute, the Cluster of Excellence "Science of Intelligence," and the Artificial Intelligence research center BIFOLD (Berlin Institute for the Foundations of Learning and Data) are jointly organizing the event.
Dr. Angela Holzer is Programme Director at the German Research Foundation. Her team is responsible for funding scholarly publications, as well as developing the strategy for Open Access and Open Science. In the following brief interview, she will discuss how openness can foster trust and integrity in Artificial Intelligence research. At the Berlin Summer School on Artificial Intelligence and Society 2025, she will deliver a keynote titled: "Is Open Science still relevant?"
What new challenges will open science face as a result of ongoing developments in the field of Artificial Intelligence (AI)?
Angela Holzer: The digital transformation has already reshaped the way research is conducted. The rise of AI will further influence the standards of scholarly work. AI has the potential to enhance research in many ways. It can facilitate certain research or publishing processes, thereby saving time and improving efficiency for researchers. For example, it can summarize literature, create visualizations, or bridge languages. AI also enables new methodological research approaches and innovations, such as advancing data analysis, modeling, and simulation.
To be beneficial, the adaptation of AI in science must meet certain preconditions. First, transparency is essential: researchers must disclose how and for what purposes AI has been used. Additionally, responsibility for research content must remain with the scientists themselves. Furthermore, copyright must be respected and acknowledged. Sensitive data must also be handled with care, and finally, AI-generated content must undergo expert scrutiny.
This also raises important questions for open science. Researchers or institutions must ensure that data, methods, and AI-generated outputs are findable, accessible, interoperable, and reusable (FAIR). Furthermore, while AI should support science, it cannot replace researchers' ethical and responsible conduct. For example, the German Research Foundation (DFG) has already addressed digital transformation through its current Code of Conduct for Safeguarding Good Research Practice. Similarly, in the context of AI, specific guidelines are needed to maintain integrity and trust in scientific work. One example is DFG’s position on the usage of AI in proposals ( DFG, German Research Foundation - DFG Formulates Guidelines for Dealing with Generative Models for Text and Image Creation).
What are the risks if AI-supported research is not conducted in a transparent and open manner?
Angela Holzer: There are multiple risks, ranging from the production of biased, flawed, or even fake research that is difficult to verify or correct, to the endangerment of academic freedom where transparency and data protection are not ensured.
However, open data also forms a crucial foundation for AI. The development of AI models and applications benefits from open science practices to the extent that these practices yield high-quality, openly accessible data that can be easily reused and used in a legally secure manner. Additionally, the trustworthiness of AI systems relies on the transparency and openness of AI systems, models, and algorithms. These examples illustrate why openness in AI-supported research is not a luxury but a prerequisite for safeguarding integrity and credibility in science.
What role should research funding bodies and policymakers play in order to establish open science and AI in the long term?
Angela Holzer: Funding bodies and policymakers must go beyond providing financial support — they need to shape the conditions under which open science and AI can thrive. This includes building and supporting infrastructures, such as the NFDI (Nationale Forschungsdateninfrastruktur) or Diamond Open Access infrastructures, embedding data protection and open science principles in regulations, and negotiating equitable contracts with publishers and platforms. They should also set ethical standards, e.g. regarding privacy regulations, and support open AI methods. Of course they also need to fund research on AI. It is equally important that they promote awareness and training so that researchers can handle these practices responsibly. Long-term success will depend on creating both strong infrastructures and a culture of openness.
In addition to these long-term measures, it is essential to support initiatives such as the Summer School in Berlin, which aims to engage today's young researchers and collaborate with them to establish a solid foundation for research in the era of AI. This exchange is essential for ensuring that openness and responsibility become or remain integral parts of future scientific practice.
Dr. Angela Holzer
Programmdirektorin
Gruppe Wissenschaftliche Literaturversorgungs- und Informationssysteme
E-Mail: angela.holzer@dfg.de
Telefon: +49 (228) 885-2568
Telefax: +49 (228) 885-713309
Kennedyallee 40
53175 Bonn
https://www.bifold.berlin/news-events/news/view/news-detail/angela-holzer-dfg-on...
Dr. Angela Holzer (DFG) on Open Science and AI
Copyright: BIFOLD
Criteria of this press release:
Journalists, Scientists and scholars
Information technology
transregional, national
Miscellaneous scientific news/publications
English
You can combine search terms with and, or and/or not, e.g. Philo not logy.
You can use brackets to separate combinations from each other, e.g. (Philo not logy) or (Psycho and logy).
Coherent groups of words will be located as complete phrases if you put them into quotation marks, e.g. “Federal Republic of Germany”.
You can also use the advanced search without entering search terms. It will then follow the criteria you have selected (e.g. country or subject area).
If you have not selected any criteria in a given category, the entire category will be searched (e.g. all subject areas or all countries).
