Why Better Models Can Create Riskier Systems: Evidence from LLM Agents in Financial Markets

A new arXiv paper (2609.04373v1) finds that improving the capability of large language models (LLMs) can degrade system-level outcomes in financial markets, as more capable models exhibit correlated behavior that creates a non-diversifiable risk floor. The authors, using an agent-based simulation with LLM traders, show that frontier LLMs' correlated behavior increases with capability, reducing market risk when shared reasoning is accurate but becoming a liability under common misinformation. This 'capability paradox' suggests that better individual models do not necessarily lead to better system-level outcomes.

arXiv:2609.04373v1 Announce Type: new Abstract: Large language models LLMs are being deployed at scale in consequential real-world systems, from financial markets to content moderation to hiring. We show that improving individual model capability can degrade rather than improve system-level outcomes. We hypothesize that shared training and architectures can lead more capable LLMs to behave more similarly, creating correlated actions that do not diversify away. We develop a general framework showing how this correlation creates a non-diversifiable risk floor and test its predictions in financial markets using an agent-based simulation with LLM traders of varying general-purpose capability. We find that: 1 frontier LLMs exhibit significantly correlated behavior that increases with capability; 2 when their shared reasoning is accurate, increasing agent participation reduces market-level risk; and 3 when agents share a common misinformation environment, the same correlated behavior becomes a liability. Together, these results identify a capability paradox: improving individual models does not necessarily produce better system-level outcomes. Whether the same dynamics arise in other domains is an open empirical question.