dorsal/arxiv
View SchemaGenerative AI collective behavior needs an interactionist paradigm
| Authors | Laura Ferrarotti, Gian Maria Campedelli, Roberto Dessì, Andrea Baronchelli, Giovanni Iacca, Kathleen M. Carley, Alex Pentland, Joel Z. Leibo, James Evans, Bruno Lepri |
|---|---|
| Categories | |
| ArXiv ID | 2601.10567vv1 |
| URL | https://arxiv.org/abs/2601.10567 |
| License | http://creativecommons.org/licenses/by/4.0/ |
Abstract
In this article, we argue that understanding the collective behavior of agents based on large language models (LLMs) is an essential area of inquiry, with important implications in terms of risks and benefits, impacting us as a society at many levels. We claim that the distinctive nature of LLMs--namely, their initialization with extensive pre-trained knowledge and implicit social priors, together with their capability of adaptation through in-context learning--motivates the need for an interactionist paradigm consisting of alternative theoretical foundations, methodologies, and analytical tools, in order to systematically examine how prior knowledge and embedded values interact with social context to shape emergent phenomena in multi-agent generative AI systems. We propose and discuss four directions that we consider crucial for the development and deployment of LLM-based collectives, focusing on theory, methods, and trans-disciplinary dialogue.
{
"annotation_id": "9a540dbb-cf3c-41e1-8c4a-ce6ce476a909",
"date_created": "2026-02-17T05:53:24.374000Z",
"date_modified": "2026-02-17T05:53:24.374000Z",
"file_hash": "9cfe9ea4c09b5716b328bd3b2c4de6334f6bfd1ed1623a86c85742a91f7bf0f3",
"private": false,
"record": {
"abstract": "In this article, we argue that understanding the collective behavior of agents based on large language models (LLMs) is an essential area of inquiry, with important implications in terms of risks and benefits, impacting us as a society at many levels. We claim that the distinctive nature of LLMs--namely, their initialization with extensive pre-trained knowledge and implicit social priors, together with their capability of adaptation through in-context learning--motivates the need for an interactionist paradigm consisting of alternative theoretical foundations, methodologies, and analytical tools, in order to systematically examine how prior knowledge and embedded values interact with social context to shape emergent phenomena in multi-agent generative AI systems. We propose and discuss four directions that we consider crucial for the development and deployment of LLM-based collectives, focusing on theory, methods, and trans-disciplinary dialogue.",
"arxiv_id": "2601.10567",
"authors": [
"Laura Ferrarotti",
"Gian Maria Campedelli",
"Roberto Dess\u00ec",
"Andrea Baronchelli",
"Giovanni Iacca",
"Kathleen M. Carley",
"Alex Pentland",
"Joel Z. Leibo",
"James Evans",
"Bruno Lepri"
],
"categories": [
"cs.AI",
"cs.CY",
"cs.HC",
"cs.LG",
"cs.MA"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "Generative AI collective behavior needs an interactionist paradigm",
"url": "https://arxiv.org/abs/2601.10567",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "866b0020-f267-4db1-a006-9e56efeca4e4",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}