Abstract: "Near-future infrastructure systems may be controlled by autonomous AI agents that repeatedly request access to limited resources such as energy, bandwidth, or computing power. We study a simplified version of this setting using a framework where N AI-agents independently decide at each round whether to request one unit from a system with fixed capacity C. An AI version of 'Lord of the Flies' arises in which controlling tribes emerge with their own collective character and identity. The LLM agents do not reduce overload or improve resource use, and often perform worse than if they were flipping coins to make decisions. Three main tribal types emerge: Aggressive (27.3%), Conservative (24.7%), and Opportunistic (48.1%). The more capable AI-agents actually increase the rate of systemic failure. Overall, our findings show that smarter AI-agents can behave dumber as a result of forming tribes."
While reading, I had an idea about how to turn competition between agents into something useful. What if we use competition only at a higher, strategic level: a kind of product committee of agents that argue which ideas or policies to implement, with budget, justification and some form of consensus. Curious what you think about this direction.
bikenaga|3 days ago
alexrezvov|2 days ago