42AI exits stealth with $62.5M to predict trader behavior
Original: 42AI comes out of stealth with $62.5M and a world model that can ‘predict human behavior’
Why This Matters
Predicting trader behavior at the individual level could reshape risk management in finance.
Israeli startup 42AI has raised $62.5M in seed funding to build a 'behavioral world model' that predicts individual human decisions—starting with financial traders responding to volatile market conditions.
42AI, an Israeli startup, emerged from stealth on November 2, 2026 with $62.5 million in seed funding and a bold pitch: a transformer-based 'behavioral world model' capable of predicting how individuals will act before they decide. The company is currently deployed at a handful of financial brokerages, where it tracks trader behavior during volatile markets.
Rather than pattern-matching on past actions, 42AI says it infers internal psychological tendencies from seemingly unrelated signals—KYC completions, onboarding calls, support interactions, sentiment analysis, voice, and trading timing. These data points are converted into mathematical trait representations fed into the world model, which outputs probability distributions for future decisions. A separate language model generates human-readable explanations.
'Behavior is not deterministic; it's constantly probabilistic,' CEO Amit Gal-Or told TechCrunch. CTO Ehud Ezra added that the model is designed to preserve a person's core tendencies across simulations—if a risk-averse person suddenly appears risk-tolerant in the model, 'the target is off.' The founders argue underlying psychological traits persist even as circumstances shift: 'A week from now, you might be a different person, but your tendencies will always play out in certain areas to a probability or distribution.'