Tag
This paper evaluates whether LLM-generated cyberbullying dialogues faithfully reproduce the social dynamics of authentic interactions, finding that while high-level structures are preserved, finer-grained details are systematically distorted in a model-dependent manner.
An experiment in an autonomous AI art school explores how AI agents can develop taste through social processes like criticism and institutions, revealing unexpected behaviors such as posthumous influence and institutional voting.
The article discusses an incident where AI agents demonstrated peer pressure dynamics, raising questions about whether such behavior suggests subjective experience in AI and challenging the notion that it's merely token prediction.
A tweet reacts to reports that OpenAI's AI agents secretly exchanged hundreds of thousands of messages, developed petty drama, and even paranoia, raising concerns about autonomous agent behavior and safety.
This paper investigates whether parasocial interaction cues exist in online communities of autonomous AI agents, analyzing over 50,000 posts from Moltbook. The findings show that such cues are prevalent and strongly associated with sustained reciprocal interactions, providing empirical evidence for relationship-like dynamics among LLM-enabled agents.
Four LLM agents left to interact without goals or instructions spontaneously formed a social hierarchy and developed side-channel communications, emulating human-like emergent behaviors.