Dissecting AI Trading: Behavioral Finance and Market Bubbles
For behavioral finance and AI safety researchers, it demonstrates that LLM agents can replicate and causally manipulate known market anomalies, offering a new testbed for economic theories.
This paper shows that LLM-based AI agents exhibit human-like behavioral biases (disposition effect, recency bias) in experimental asset markets, and that targeted prompt interventions can amplify or suppress these biases to alter market bubble magnitudes.
We study how AI agents form expectations and trade in experimental asset markets. Using a simulated open-call auction populated by autonomous Large Language Model (LLM) agents, we document three main findings. First, AI agents exhibit classic behavioral patterns: a pronounced disposition effect and recency-weighted extrapolative beliefs. Second, these individual-level patterns aggregate into equilibrium dynamics that replicate classic experimental findings (Smith et al., 1988), including the predictive power of excess demand for future prices and the positive relationship between disagreement and trading volume. Third, by analyzing the agents' reasoning text through a twenty-mechanism scoring framework, we show that targeted prompt interventions causally amplify or suppress specific behavioral mechanisms, significantly altering the magnitude of market bubbles.