Tel Aviv-based 42AI, which is developing a "behavioral world model" that tries to predict how humans will act, emerges from stealth with a $62.5M seed

Your financial brokerage may soon predict your trading behavior with 98% accuracy, altering how they manage risk and liquidity on your behalf.

Tel Aviv-based startup 42AI has emerged from stealth, announcing $62.5 million in seed funding for its "behavioral world model" designed to predict human actions. The company, which spent three years in development, integrates psychological research, neuroscience, and genetics to analyze individual tendencies rather than just past behavior. 42AI's model processes various data points, from support calls to financial transactions, to understand internal motivations and predict future decisions with probabilistic accuracy. Currently deployed in several financial brokerages, the system claims high accuracy in predicting trading flows and direction, generating seven figures in revenue. The funding was primarily from founders and Infinity Group, with plans to expand applications into gaming, fraud prevention, and cybersecurity, and eventually into robotics and AI agents.

The core innovation of 42AI's model lies in its attempt to derive behavioral predictions from intrinsic psychological traits and motivations, rather than solely relying on extrapolation of past actions. By analyzing a broad spectrum of user interactions, including sentiment analysis in communications and reaction to seemingly unrelated events, the company aims to build a more robust and potentially more accurate predictor of human decision-making, particularly in dynamic environments like financial markets.

This emergence signals a growing trend towards deeper, more nuanced AI models capable of understanding and anticipating human behavior, moving beyond pattern recognition to infer underlying psychological drivers. While currently focused on finance, the ambition to apply this to gaming, cybersecurity, and even human-robot interaction suggests a broader market for AI that can model complex human decision-making processes, potentially transforming customer interaction and agentic AI design across industries.