Artificial Intelligence Takes Center Stage at the 2025 Annual Assembly of the German National Academy of Sciences Leopoldina
This year’s Annual Assembly of the German National Academy of Sciences Leopoldina, held in Halle (Saale) on September 25 and 26, has focused intensively on the multifaceted advancements and societal impacts of artificial intelligence (AI). The gathering brought together a cadre of distinguished scholars from diverse fields to explore the technological breakthroughs in AI, its applications across various disciplines, and the challenging ethical questions it raises. Opening speeches by Dr. Lydia Hüskens, Deputy Minister President and Minister for Infrastructure and Digital Affairs of Saxony-Anhalt, alongside Dr. Rolf-Dieter Jungk, State Secretary at the German Federal Ministry of Research, Technology and Space, set the stage for a profound interdisciplinary dialogue.
Professor Dr. Bettina Rockenbach, President of the Leopoldina, emphasized the dual nature of AI’s evolution: while offering transformative possibilities for research, medicine, and communication, it also introduces complex dilemmas around risk management and ethical responsibility. Leveraging Leopoldina’s interdisciplinary expertise, the assembly sought to dissect various dimensions of AI — from foundational scientific advances to sector-specific applications in medicine, geosciences, and physics, establishing a comprehensive thematic landscape of inquiry.
The inaugural keynote delivered by computer scientist Dr. Cordelia Schmid offered a deep dive into the progressive development of AI technologies, highlighting current capabilities such as deep learning models, pattern recognition, and adaptive algorithms, while projecting potential future trajectories. Dr. Schmid delineated the evolution of neural network architectures and illustrated how advances in computational power and data availability have driven AI’s rapid growth. This was followed by an engaging panel discussion themed “Artificial intelligence in the services of humans – (How) can we achieve this?”, featuring AI researchers including Professors Dr. Niki Kilbertus, Nadja Klein, Cordelia Schmid, and innovation expert Professor PhD Dietmar Harhoff. Moderated by journalist Christoph Drösser, the conversation navigated the pressing question of embedding human-centric values within AI systems to guarantee beneficial deployment.
On the second day, the assembly shifted focus to the practical realms where AI is making significant inroads. Professor Dr. Sami Haddadin, an authority in robotics, elucidated advances in autonomous machine learning, emphasizing how contemporary robots acquire the ability to interpret sensory input, plan motion trajectories, and adapt dynamically to unpredictable environments. His discourse detailed the integration of reinforcement learning techniques with tactile sensing, enabling robots to “think” and self-improve beyond pre-programmed constraints.
Meteorologist Professor Dr. Susanne Crewell shed light on AI’s transformative influence on meteorology and climate science. By incorporating machine learning-driven modeling and big data analytics, AI is revolutionizing weather forecasting accuracy and enhancing our grasp of complex climate dynamics. Dr. Crewell underscored how convolutional neural networks and ensemble learning algorithms contribute to resolving nonlinear atmospheric processes and predicting extreme weather events with unprecedented precision.
Ethical considerations surfaced prominently in the lecture by Dr. Philipp Lorenz-Spreen, who examined AI’s interplay with social media platforms and its reverberating effects on democratic processes. Addressing the intricate behavioral feedback loops created by algorithm-driven connectivity, Dr. Lorenz-Spreen illuminated the challenges posed by content personalization, misinformation spread, and the amplification of societal polarization. His insights called for critical scrutiny into AI regulation frameworks to safeguard democratic integrity.
The concluding lecture by Professor Dr. Moritz Helmstaedter masterfully bridged artificial and biological intelligence, exploring mutual inspirations between computational models and neuroscientific discoveries. Helmstaedter highlighted how advances in understanding neural circuitry and brain plasticity have influenced AI architectures, while AI tools, in turn, provide novel methodologies to decode brain function and cognition, symbiotically accelerating progress in both realms.
A notable highlight of the assembly was the conferment of the “ZukunftsWissen – Early Career Award” to Professor Dr. Zeynep Akata, recognized for her pioneering work in explainable AI (XAI). Her research focused on generating interpretable AI systems integrating visual, linguistic, and conceptual modalities to ensure transparency in algorithmic decision-making. Akata’s contributions promise to enhance human trust in AI by elucidating complex model behaviors through multi-modal explanations.
The prestigious Cothenius Medal 2025 was awarded to Professor Dr. Kai Simons, honoring his seminal lifetime contributions to molecular biology, particularly his work on cell membrane organization and the molecular dialogue between viruses and host cells. Simons’ research underpins crucial biomedical understanding pertinent to virology and immunological applications.
This year’s assembly welcomed talented German students funded by the Wilhelm and Else Heraeus Foundation, providing them an exceptional opportunity to engage directly with cutting-edge scientific discourse. Additionally, postdoctoral fellows benefited from participation grants provided by the Friends of the Leopoldina Academy and the Alfried Krupp von Bohlen und Halbach Foundation, fostering early-career engagement in high-level scholarly exchange.
The scientific coordination of the event was managed by mathematician and computer scientist Professor Dr. Dr. Thomas Lengauer and physicist and computer scientist Professor Dr. Klaus-Robert Müller. Their visionary leadership shaped the event’s theme, promoting an interdisciplinary approach to understanding AI’s ramifications.
All lectures were held bilingually in English and German with simultaneous translation, and the entire assembly was livestreamed globally via the Leopoldina’s YouTube channel, enabling broad access to this critical discourse without prior registration. The event demonstrated how scientific communities can drive informed dialogue on AI’s societal integration, ensuring technologies progress in alignment with human values and ethical imperatives.
As the German National Academy of Sciences, the Leopoldina plays a pivotal role in providing evidence-based policy advice informed by extensive interdisciplinary assessment. Founded in 1652 and serving as Germany’s National Academy since 2008, the Leopoldina unites around 1,700 members from over 30 countries, representing virtually all scientific domains. Its commitment to independence and societal welfare enables it to act as a trusted advisor to political decision-makers, internationally recognized for contributing to global academic dialogue including G7 and G20 summits.
The 2025 Annual Assembly underscored a pivotal juncture where AI’s promise and responsibilities intersect, inviting researchers, policymakers, and the public alike to thoughtfully navigate the future of intelligent technologies.
Subject of Research: Artificial Intelligence and its Applications Across Scientific Disciplines and Society
Article Title: Artificial Intelligence at the Forefront: Insights from the 2025 Leopoldina Annual Assembly
News Publication Date: September 25-26, 2025
Web References:
- Leopoldina YouTube Channel Livestream: https://www.youtube.com/@nationalakademieleopoldina
- Annual Assembly Program: https://www.leopoldina.org/en/events/event/event/3237/
Keywords: Artificial intelligence, machine learning, robotics, explainable AI, AI ethics, climate modeling, neural networks, interdisciplinary research, AI in medicine, AI and democracy, cell membranes, viral-host interaction
