π€ AI Summary
This work proposes a high-fidelity, scalable social behavior simulation framework to address the ethical risks associated with deploying autonomous agents in real-world social networks. The framework employs large language modelβdriven agents integrated with an event-driven simulation engine, a cognitive memory mechanism, and an asynchronous orchestration API. Operating in an isolated environment, it ingests real-time content streams from Bluesky to reproduce multimodal social interactions under human-like temporal patterns. Its key innovation lies in the first-of-its-kind asynchronous architecture that seamlessly combines live data streams with cognitive modeling, enabling safe, controlled, large-scale adversarial social experiments. The system has successfully simulated coordinated disinformation campaigns, offering computational social science and red-teaming communities a scalable and high-fidelity experimental infrastructure.
π Abstract
BotVerse is a scalable, event-driven framework for high-fidelity social simulation using LLM-based agents. It addresses the ethical risks of studying autonomous agents on live networks by isolating interactions within a controlled environment while grounding them in real-time content streams from the Bluesky ecosystem. The system features an asynchronous orchestration API and a simulation engine that emulates human-like temporal patterns and cognitive memory. Through the Synthetic Social Observatory, researchers can deploy customizable personas and observe multimodal interactions at scale. We demonstrate BotVersevia a coordinated disinformation scenario, providing a safe, experimental framework for red-teaming and computational social scientists. A video demonstration of the framework is available at https://youtu.be/eZSzO5Jarqk.