Title | From Local Interactions to Global Conventions: Emergent Social Coordination in Humans and Machines
Location | Corvinus University, 1093 Budapest, Fővám tér 8. Room E.3.3005.
Abstract | Social conventions are the foundation of social coordination, shaping how individuals come together to form a society. In this talk, I will present theoretical and experimental findings that demonstrate the spontaneous emergence of social norms in human groups, as well as the existence of tipping points in social conventions. I will then explore the case of populations of large language models (LLMs). As AI agents increasingly communicate using natural language, understanding how they develop conventions is crucial for interpreting and managing their collective behaviour. I will show that LLM populations can establish social conventions and highlight how collective biases can emerge even when individual agents appear unbiased. The ability of AI agents to develop norms without explicit programming has significant implications for designing AI systems that align with human values and societal goals.
Ashery, A. F., Aiello, L. M., & Baronchelli, A. (2024). The Dynamics of Social Conventions in LLM populations: Spontaneous Emergence, Collective Biases and Tipping Points. arXiv preprint arXiv:2410.08948.
Bio | Andrea Baronchelli is a Professor of Complexity Science at City University of London, and the Token Economy theme lead at The Alan Turing Institute. His research investigates the emergent properties of decentralized socio-technical systems, including how individuals coordinate in social networks, shape and interact with blockchain technology, and how information spreads and influences social media. Baronchelli's work has been published in leading journals such as Nature, Science, and PNAS, and has received support from organizations including UKRI, PayPal, and the UK Government. In 2019, he was awarded the Young Scientist Award for Socio- and Econophysics by the German Physical Society.
This is a Hybrid event. You can attend on Zoom using the link above.