42AI Exits Stealth With $62.5M Seed and a Model That Predicts Traders

Israeli startup 42AI has come out of stealth with $62.5 million in seed funding and a claim that its "behavioral world model" can predict how individual traders will act — even in situations those traders have never faced before.

The company is already running the system at 10 large brokerages, which use it to manage liquidity, margin requirements and hedging exposure, CEO and co-founder Amit Gal-Or told TechCrunch. Revenue has passed "seven figures," he said, without giving a precise number.

What does the model actually do?

42AI says its model doesn't extrapolate from past behavior. Instead, it tries to read a person's underlying motivations by watching how they react to seemingly unrelated events — KYC checks, onboarding, support calls, money transfers and trading itself.

"When the person is talking to support or sales in the initial process […] we're looking at the sentiment, we're looking at the voice, we're looking at a lot of interactions that they're doing, how many times do they call, and so forth," said CTO and co-founder Ehud Ezra. "And all of those data points are going into the model. Later on, you have the interactions: how you buy and sell, and when, and we're looking at the market at that stage."

The startup converts those observations into mathematical representations of traits and feeds them into a transformer-based world model that outputs probabilities for decisions under different circumstances. A separate language model generates explanations for the predictions.

"So we're trying to look at how they're looking at the world. In a world model, every individual is the world," Ezra said.

How accurate is it?

Gal-Or claims the model reaches 98% accuracy when predicting market trades — but he clarified that figure measures whether trading flows fall within a predicted range, not whether each individual trade was called correctly.

The model consistently scores 0.8 to 0.85 on F1 — the machine-learning metric that balances correct predictions against false alarms and missed events — for predicting trading direction and volume.

The founders are blunt about the limits. "There's a natural ceiling because you really don't have access to everything that a person experiences," Gal-Or said. Ezra added that accuracy drops when the system misses situational context — say, a trader changing strategy on a friend's advice.

Training itself rewards stability: if a usually risk-averse person suddenly registers as a risk-taker in simulation, "the target is off," Ezra said.

Where did the data come from?

42AI spent three years building the system, drawing on psychological research, neuroscience and genetics, plus input from psychology and PTSD experts and Dan Shechtman, who won the 2011 Nobel Prize in Chemistry.

The team first studied thousands of people through psych assessments and behavioral scenarios, then added government and biobank datasets covering more than 2 million people, Gal-Or said. Customer datasets have since expanded to cover almost 30 million people. Gal-Or argues privacy holds: "the model never knows who the person is," since it doesn't need personally identifiable information — though 42AI still builds individual behavioral profiles from linked customer activity.

Customers pay either a flat fee or a percentage of the value the platform helps them capture. Gal-Or claimed one brokerage generated roughly $30 million in additional value through dynamic hedging.

Who funded it — and what's next?

The $62.5 million came from the founders and Infinity Group over three years, and wasn't entirely external money, Gal-Or said. Infinity was founded by his father, Amir Gal-Or — also 42AI's chairman and a co-founder. The company now plans to raise more.

Beyond finance, 42AI is exploring gaming, fraud prevention and cybersecurity, and eventually wants robots and AI agents to use the model to understand people. That's distant: Ezra acknowledged the model must be trained on rules and context for each new domain.

Then there's the ethics question — what stops a brokerage from exploiting predictions about its own customers? Ezra called it "a valid concern" and pointed to regulations, monitoring and an internal ethics committee, though the founders didn't explain what would concretely stop customers from gaming the people being profiled.

Gal-Or was unmoved. "We are very proud that actually we are part of the solution; we are the solution to this problem," he said.