LOTAN MAGAL AND ASKIT: AI FOR HOW PEOPLE ACTUALLY DECIDE
Lotan Magal is building AskIt to predict human decision-making — applying AI to understand how people choose, decide, and act in complex situations.
Many AI products optimize for classification: labels, segments, probabilities. AskIt sits closer to decision science: what drives choices under uncertainty, how preferences shift with context, and how organizations can anticipate behavior rather than only describing it after the fact. That distinction matters for product teams, policy designers, and operators who need to reason about humans as dynamic agents, not static profiles.
AskIt helps organizations anticipate behavior and design better products, policies, and experiences based on how people actually think — not how models assume they think when trained on simplified datasets.
Magal's personal through-line is resilience under ambiguity: continuing to build when signals were noisy and stopping would have been the rational short-term move. That mindset matches the category, where progress depends on iterating through messy human data and refusing to confuse a clean demo with a robust system.
At pre-seed, AskIt is pre-revenue but focused on the hardest early work: defining evaluation standards for decision prediction that can survive contact with reality.
AskIt is a bet that the next frontier of applied AI is not only language and vision, but judgment under constraints — the invisible layer that determines whether products, incentives, and interventions work when they leave the slide deck.