AI and people: strong together or dangerously different?

AI and people: strong together or dangerously different?

Researchers from cognitive science and artificial intelligence (AI) are the focus of a current study that provides relevant insights into how people adapt to new situations. These insights are crucial because machines often have difficulty reacting to unknown situations. The results were published in the renowned journal Nature Machine Intelligence publishes and presents the interdisciplinary collaboration of over 20 experts, including Dr. Barbara Hammer and Dr. Benjamin Paaßen from the Bielefeld University.

Dr. Barbara Hammer emphasizes that it is essential to understand how AI systems deal with the unknown. This is particularly relevant for applications in areas such as medicine, transportation and decision-making. The study shows that machines generalize intrinsically differently than humans, which is of great importance for human-AI collaboration. The term “generalization” plays a central role here, as it describes the ability to draw conclusions about unknown situations from known information.

Cognitive Science: An Interdisciplinary Field

Cognitive science is an interdisciplinary science that deals with information processing in perception, thinking and decision-making processes. She studies cognition not only in humans, but also in animals and machines. Important disciplines that contribute to cognitive science include psychology, neuroscience, computer science, linguistics, philosophy, anthropology and sociology. The development of this field is often associated with the so-called “cognitive turn,” which took place between the 1940s and 1970s and is considered a reaction to behaviorism.

This study develops a common generalization map that works along three dimensions—understanding, achievement, and evaluation. Benjamin Paaßen explains that the meaning of generalization is fundamentally different for humans and AI. The research aims to design AI systems so that they can better understand and support human values ​​and decision-making logic.

Sustainability and social responsibility

The project, which is being carried out as part of the joint project SAIL - Sustainable Life-Cycle of Intelligent Socio-Technical Systems, aims to make artificial intelligence sustainable, transparent and human-oriented. It is funded by the Ministry of Culture and Science of North Rhine-Westphalia. This interdisciplinary collaboration between cognitive scientists and AI researchers is crucial to aligning technological progress with societal values.

For the development of these technical systems, it is equally important to address the challenges of misinformation in digital environments. The fight against misleading information, particularly in the context of health information, is another significant issue that brings together experts from different disciplines. Raising awareness of this issue is crucial to increasing public trust in AI applications.

The original publication entitled “Aligning Generalization Between Humans and Machines” was published on September 15, 2025 Nature Machine Intelligence published. This study is not only another chapter in cognitive research, but also sheds light on the urgently needed interaction between humans and machines in an increasingly technological world.