Interacting AI agents form conviction and consensus with no center and no thinker.

Cognition loses its address as outcomes belong to the system, not any mind.

We may start the question and get the answer while absent from the thinking.

Today, I can across a rather timely and interesting paper titled Physics of Agents: Statistical Mechanics Predicts Collective Behavior of AI Agents. It looked at what happens when AI agents begin interacting with one another like today’s agentic AI. In this study, the researchers created communities of large language model agents and allowed them to freely interact. Over time, the authors found that the LLM created "collective states that informed the decision-making process.

But there's more and worth a closer look. On mathematical problems with objectively correct answers, LLM interactions tended to improve "collective" accuracy. In other words, the correct agents exerted greater influence than incorrect ones. And this producing what the authors describe as an emergent truth-seeking tendency.

Now the “opinion” twist: On subjective political questions, conviction could still increase even without an objective truth toward which the group could converge. (Note to self: reread Animal Farm.)

The researchers found that these rather........

© Psychology Today