Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates
2026-07-01 • Computation and Language
Computation and Language
AI summaryⓘ
The authors explain that big language models on their own are simple and don’t show complex lifelike behavior. But when many of these models interact as a group, they start to show new and interesting behaviors. These groups of models can remember things, use tools, share skills, and act on their own. Because they communicate using normal language, researchers can easily study and understand their actions. The authors suggest these groups could be used to study artificial life and review some recent examples of this idea in practice.
Large Language Model (LLM)Emergent behaviorAgentic systemsArtificial Life (ALife)InterpretabilityNatural language communicationPersistent memoryMulti-agent systemsComputational substrate
Authors
Elias Najarro, Ane Espeseth, Eleni Nisioti, Sebastian Risi, Stefano Nichele
Abstract
Complexity and interpretability rarely coincide: systems rich enough for complex behaviours to emerge are usually too opaque to question, while transparent ones are too simple for anything complex to emerge. A single large language model (LLM) is a static artefact, hardly exhibiting any of the emergent properties we associate with life. This changes through interaction: populations of LLMs display emergent dynamics absent from isolated models. Furthermore, LLMs can be endowed with persistent memory, tools and shared skills, and the capacity to initiate actions unprompted, i.e., turning LLMs agentic. In this paper, we argue that such collectives of agents can serve as a computational substrate for Artificial Life (ALife) research. Critically, since the agents communicate in natural language, their collective behaviour can be directly interrogated by examining textual traces and asking the agents themselves. We outline the notion of interpretability in language-model research and extend it for collectives of agents. Lastly, we survey recent examples of agentic LLM collectives that already instantiate the idea of agentic substrates, from controlled experiments to deployments in the wild.