Generative AI Meets Agent-Based Modelling and Simulation

Dear Colleagues,

The intersection of generative Artificial Intelligence (AI) and Agent-Based Modelling and Simulation (ABMS) represents a transformative frontier in computational social science and complex systems research. ABMS provides a powerful bottom-up framework for simulating complex phenomena by modeling the behaviour and interactions of individual agents within specified environments. Meanwhile, LLMs and other generative AI technologies have emerged with unprecedented capabilities in natural language understanding, text generation, code synthesis, logical reasoning, and human-like dialogue.

Despite the rapid growth of research at this intersection, there remains a critical need for structured, cross-disciplinary exploration of how, where, and why generative AI technologies can support and enhance the ABMS process. Traditional ABMS, rooted in symbolic AI paradigms, emphasises explicitly defined rules and transparent system dynamics. In contrast, generative AI operates through data-driven pattern recognition and emergent linguistic structures. These complementary yet distinct paradigms present both significant opportunities and considerable challenges.

Generative AI offers substantial potential to enhance the ABMS workflow throughout the ABMS cycle, from problem formulation through literature synthesis, system analysis via mechanism extraction, conceptualisation through translating narratives into structured models, to implementation via code generation and results communication through stakeholder-tailored reporting. LLMs can also serve as cognitive components within simulated agents, enabling novel approaches to modelling human-like reasoning.

However, this integration presents significant challenges: bias propagation from training data, opacity of LLM decision-making, difficulties with verification and validation, risks of hallucination, and fundamental tensions between ABMS's transparent, rule-based nature and neural networks' black-box characteristics. Ethical considerations surrounding data privacy, misinformation, and equitable access must be carefully addressed.

This Special Issue seeks to advance the responsible and critical integration of generative AI technologies as tools to support ABMS research and practice. We welcome both theoretical contributions that explore foundational questions about this integration and empirical studies that demonstrate novel applications, methodologies, and validation approaches. We particularly encourage cross-disciplinary submissions from researchers and practitioners in areas such as Computer Science and Artificial Intelligence, Computational Social Science, Sociology, Political Science, Operations Research, Management Science, Economics, Complex Systems Science, Network Science, Environmental Science, Public Health, and Urban Planning. By bringing together diverse perspectives from multiple disciplines, we aim to establish best practices, identify key challenges, and outline future directions for this rapidly evolving field.

We invite submissions addressing (but not limited to) the following topics: theoretical and methodological advances in integrating generative AI with ABMS; applications of generative AI across the ABMS cycle; domain-specific applications, demonstrating how generative AI enhances ABMS in these areas; critical evaluations of LLM-augmented ABMS; and emerging techniques and technologies relevant to LLM-augmented ABMS.

Prof. Dr. Ender Özcan
Dr. Peer-Olaf Siebers
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Algorithms is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.