1. Introduction

In the realm of sociology and neuroscience, the concept of social intelligence (SI) has been a subject of interest since the early 20th century. Researchers have explored its significance in enhancing group competition, collaboration, and cooperation based on individual differences. Social intelligence involves the ability of social entities to flexibly adjust their decision-making processes based on the actions of others, modeling their goals and internal processes to adapt to shared environments. While social intelligence models have laid the foundation for advanced forms of collective intelligence, the application of these models in artificial intelligence (AI) research has been limited. This limitation stems from a lack of clear articulation of the elements and their interactions within social intelligence models, hindering their translation into practical AI models.

Given the inadequacy of a singular "City Brain" model to meet the demands of modern governance, this article proposes a theoretical model called "City Minds" based on observations of urban social organization. This model aims to combine urban science with next-generation AI technologies to construct a novel network tailored for complex and heterogeneous communities.

2. Understanding Social Intelligence

2.1 Key Features of Social Intelligence

Social intelligence, as highlighted by scholars like Kliemann and Chen, involves individuals adjusting their decision-making patterns based on the actions of others in a shared social environment. The ability to predict and react to the short-term or long-term behaviors of others within a common social context is crucial. Kingsbury has further summarized these interaction patterns within social groups, suggesting a multi-brain structure.

2.2 Definition of "City Minds"

"City Minds" is defined in this context as an advanced social intelligence model aimed at enabling AI to learn how a social community organizes, collaborates, and acts. It involves diverting information into multi-layered, three-dimensional decision-making mechanisms, ultimately seeking strategies for mutual benefit among heterogeneous entities to drive overall performance development.

2.3 Learning Modes of "City Minds"

The essence of the Minds model lies in a shift in learning modes. It encompasses three stages:

Learning Society Independently (LSI): A primary model of collective intelligence where each entity makes autonomous decisions but requires overall control through a collective model. This model enhances the autonomy and decision-making capabilities of entities.

Learning Society Collaboratively (LSC): Entities gain the ability to build information networks based on their individual needs, optimizing behavior around their development goals. This stage increases the heterogeneity among entities.

Learning Society Cooperatively (LSCO): Entities collaborate while pursuing their individual goals, striking a balance between cooperation and competition. This distinctive learning mode sets the Minds model apart from other collective intelligence models.

3. Nine Key Issues of "City Minds"

3.1 First Issue: Why Shift from City Brain to City Minds in Urban Intelligence?

The transition from a singular City Brain to City Minds is motivated by the limitations of relying on a single decision model. Minds overcome these limitations by providing a more adaptable and collaborative approach to governance.

3.2 Second Issue: How to Architect City Minds and Represent Decision-Making Entities?

The Minds model consists of four main types of entities: Core Brain (CB), Assistant Brain (AB), Distributed Brain (DB), and Terminal Brain (TB), each playing specific roles in decision-making at different levels of the urban system.

3.3 Third Issue: How to Establish Connections Between Minds?

Three types of connections are identified: Main-Subordinate Coordination, Hierarchical Coordination, and Community Coordination. These relationships form a complex network that facilitates information exchange and decision-making across different levels of the urban system.

3.4 Fourth Issue: How to Architect the Community Structure of City Minds and Define the Roles of Each Brain?

The community architecture of City Minds integrates the Main Brain, Assistant Brain, Distributed Brain, and Terminal Brain into a dynamic interactive system, ensuring collaborative development.

3.5 Fifth Issue: How to Simulate City Minds in the Digital Realm?

Simulating City Minds involves empowering individual nodes with collective perception and predictive capabilities, emphasizing active group perception and dynamic system predictions.

3.6 Sixth Issue: How Does City Minds Iterate?

The evolution of urban intelligence can be divided into three stages: Single Brain System, Low-Level Collective Intelligence System, and Minds System. This iterative process represents a historical progression in urban intelligence.

3.7 Seventh Issue: How to Map Virtual Minds Relations to Real Urban Governance?

Incorporating Minds into urban governance involves parallel AI scenario groups, enabling multidimensional interaction and resource coordination, and adopting decentralized decision-making models at the community level.

3.8 Eighth Issue: How to Facilitate Interaction Among the Material, Social, and Digital Worlds?

City Minds create a three-way interaction structure among the material, social, and digital realms. This interaction allows urban systems to learn, predict, and adapt, enhancing the overall intelligence of the city.

3.9 Ninth Issue: How to Architect Connections Between City Minds?

Interconnecting City Minds involves establishing close information exchange not only between decision-makers but also across various levels and types of brains, fostering a robust network for decision-making.

4. Conclusion

The shift from "City Brain" to "City Minds" represents a transformative leap in the field of urban intelligence. This evolution addresses limitations in decision models, enhances collaboration, and adapts to the diverse and complex needs of urban environments. Society Brains embody a multi-brain structure that reflects the collaborative and dynamic nature of social communities, presenting a promising direction for the future of AI research and urban governance.

In summary, Society Brains revolutionize artificial intelligence by transitioning from learning individual intelligence to understanding and replicating the intelligence of social communities. City Minds, as a manifestation of this evolution, provides a framework for more adaptive and collaborative urban governance, paving the way for a new era of artificial intelligence applications.

AI Writing From Article: WU Zhiqiang, GAN Wei, LI Shuran, et al. Society Brains: Theoretical Model and Key Issues[J]. Urban Planning Forum,2023(6):20-26