42AI Raises $62.5M with Behavioral World Model

The system is deployed with ten major brokerages to manage liquidity, margin, and hedging. It analyzes user interactions across KYC, support calls, and transactions to infer psychological tendencies. The model uses transformer-based architecture to generate probabilistic decision forecasts, maintaining consistent behavioral traits during simulations. Performance metrics include an F1 score of 0.8–0.85 for trade direction and volume, with 98% accuracy reported on transaction flow trends.

Summarised from the original article by an open-source model, then checked automatically: every number and English name must appear in the source. The original article is authoritative.