Simile

Simile is an AI company based in Palo Alto, California. It is building a foundation model for human behavior. The company partners with real people to create AI agents that copy their preferences and decision patterns, then combines those agents into simulated populations. Organizations use them to test product launches, pricing changes, marketing campaigns, and policy decisions before making them for real, and to see what people would likely do and why. Customers include CVS Health, Gallup, Wealthfront, and Deloitte, with uses ranging from customer research to earnings-call preparation and market-entry decisions.

Simile was founded in 2025 by Stanford researchers Joon Sung Park, Michael Bernstein, Percy Liang, and Lainie Yallen. It grew out of Park's "Smallville" work on generative agents: AI systems that remember, plan, and interact like real people. The company left stealth in February 2026 with a $100 million Series A led by Index Ventures, then raised a $200 million Series B led by Greenoaks five months later, at a $2 billion valuation. Revenue grew fivefold in that period, and Fortune 100 companies have run tens of millions of simulations on its models. The long-term goal is bigger than market research: models accurate enough to simulate whole populations, so decisions can be rehearsed before they reach the real world.¹