# Look-Ahead Bias

A trading system for large language models reading Web financial news treats each news item as a testable unit and checks what its judgments mean against announcement-window abnormal returns and issuers' 8-K filings. The confirmatory reader emu's item judgments reach a weekly Spearman correlation of 0.119 (95% CI [0.098, 0.140]) with announcement-window abnormal returns, and event types agree with 8-K filings at AUC 0.833; measured from the first trade after publication, the correlation falls to 0.023, indicating that Web news often reports a price move already realized.

A trading system for large language models reading Web financial news treats each news item as a testable unit and checks what its judgments mean against announcement-window abnormal returns and issuers' 8-K filings. The confirmatory reader emu's item judgments reach a weekly Spearman correlation of 0.119 (95% CI [0.098, 0.140]) with announcement-window abnormal returns, and event types agree with 8-K filings at AUC 0.833; measured from the first trade after publication, the correlation falls to 0.023, indicating that Web news often reports a price move already realized.

FinCosmos opens all three uncertainties of financial agent evaluation at once: a simulated investor reveals a hidden strategy only when asked, the agent must source its own Web data and backtest under a data cutoff, and programs verify the delivery instead of matching a reference answer. Across 150 sessions with four coding agents and a tool-free model, none is accepted, and all 35 sessions the simulated user signed off fail at least one program veto.

FinCosmos opens all three uncertainties of financial agent evaluation at once: a simulated investor reveals a hidden strategy only when asked, the agent must source its own Web data and backtest under a data cutoff, and programs verify the delivery instead of matching a reference answer. Across 150 sessions with four coding agents and a tool-free model, none is accepted, and all 35 sessions the simulated user signed off fail at least one program veto.