Aug 2026· Sociologica· Vol 20, pp. 163-177· 1 citation· 87 references
TL;DR
Before attempting to imagine a new “social physics” in the age of AI, a far more important task for sociology is to address its fundamental linguistic deficit, which has often been exacerbated by the influence of individualistic theories of action and/or language.
Abstract
Recent public and/or intellectual controversies around large language models (LLMs) forcefully demonstrate the relevance of language to social life. In both computer science and computational social science (CSS) research, their development has led to a surge of interest and engagement with the complexity of natural language and its relationship to culture. But what sociological understanding of language can this emergent research draw on? The rise of LLMs — and their use in social simulations of interacting “language agents” — thus renders salient the relatively minor importance attributed to language in many conceptions of social action. Before attempting to imagine a new “social physics” in the age of AI, a far more important task for sociology is to address its fundamental linguistic deficit, which has often been exacerbated by the influence of individualistic theories of action and/or language.
The central claim of this paper is that current generative AI systems are not social agents despite their increasingly human-like language abilities and future work should seek to align their outward behavior with their actual capabilities while developing appropriate safeguards for applications in which emotional or s...
Matthias Scheutz· The Paris Journal on AI &...· 0 citations
Communication across languages is increasingly mediated by artificial intelligence (AI). This creates opportunities for people with different language backgrounds to communicate in their respective native languages. AI-mediated communication can therefore reduce cognitive effort, disfluency, and social stigma associate...
Zeynep Aslan, Janet Geipel, B. Keysar· Current Opinion in Psycholog...· 0 citations
MindEvolve is introduced, an autonomous workflow designed to predict behavior in social interactions by generating interpretable symbolic models of cognition, and provides a roadmap for advancing LLM-based cognitive modeling toward human-expert-level theory construction.
Ying-Ying Ye, You-Le Fang, Xiao-Xue Gao et al.· bioRxiv· 0 citations
This paper shows that when agents share more context they produce less informative languages, and provides a proof of concept and enables the testing of further hypotheses of the impact of shared context on language structure.
Hedvig Skirgård, Jérôme Michaud, Anna Jon-And· 0 citations
Understanding the social intelligence of large language models (LLMs) is increasingly important for applications involving human-AI interaction and multi-agent environments. Social intelligence comprises two aspects: cognitive intelligence, the ability to infer others'mental states, and behavioral intelligence, the abi...
Alieh Zamani, Soroush Esfahanian, Erfan Shahabi et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.