AI summaryⓘ
The authors studied how groups of language-model agents can work together without direct communication or assigned roles to create technology in a shared virtual world called SwarmWorld. These agents explore, build, test, and improve tools that keep working even after the agents are removed, showing that decentralized collaboration can create useful, lasting technologies. The agents develop different roles on their own and mostly learn by watching physical changes rather than talking. The study found that groups sharing resources and artifacts can build a wider range of useful technologies than agents working alone, though the best single invention may come from solo efforts. Overall, the agents create lasting technological systems mainly through indirect interaction rather than direct communication.
Collective intelligenceLanguage-model agentsDecentralized multi-agent systemsSwarmWorldStigmergyTechnological societiesExecutable controllersArtifact inheritanceExploration and constructionSimulation environments
Abstract
Collective intelligence can emerge when individuals coordinate through a shared environment, allowing local actions to accumulate into durable social organization. Language-model agents offer a new substrate for this process, yet most multi-agent systems rely on direct conversation, predefined roles, or centralized workflows. It remains unclear whether decentralized agents can build functional technologies and outperform independent search. Here, initially homogeneous LLM agents in SwarmWorld self-organize without assigned roles or recipes into evolving technological societies. Agents explore a spatial environment, process resources, test materials, construct persistent artifacts, and write executable controllers evaluated by a deterministic simulator under unseen disturbances after the agents are removed. SwarmWorld splits cognition from consequence: agents propose architectures and controllers within fixed action and material schemas, while the simulated world determines function. Shared societies develop broader, more resilient technological portfolios than a strong best-of-N isolated-search baseline, although isolated search remains competitive for the strongest artifact. Agents differentiate into exploration, construction, maintenance, and coordination behaviors, transitioning as the world matures. Technologies accumulate through collaborative construction, executable inheritance, and persistent agent-artifact networks, with most reuse beginning through physical observation rather than communication. Explicit cultural mechanisms amplify collaboration and organization, but functional benefits depend on outcome and timescale. Physical stigmergy alone supports capable societies, while interaction drives persistent technological ecologies rather than universally superior individual inventions.