Simile, a one-year-old startup building AI agent “twins” that model how consumers behave, has raised $200M in Series B funding led by Greenoaks, according to the New York Times DealBook. The round values the company at $2B, double its $1B Series A valuation from six months ago.
The company has now raised $300M total, with both rounds closing within a six-month window. That pace of capital deployment is unusual even by current AI fundraising standards.
How Simile Works
Simile’s platform analyzes real consumer data to build synthetic “agentic twins” that predict how people will respond to product launches, pricing changes, marketing campaigns, and brand repositioning. Companies run scenarios against these AI consumer models instead of assembling real focus groups or surveying actual users.
Microsoft has reportedly used Simile’s platform to gather feedback on products without surveying real humans, according to the Times report.
The Category Bet
The valuation doubling in six months reflects a specific investor thesis: enterprises will increasingly prefer to test strategies against synthetic AI agents rather than recruit human participants. The advantages are straightforward: speed (hours instead of weeks), cost (no recruitment, incentives, or facility rental), and the ability to test scenarios that would be impractical or impossible with real panels.
The question is whether synthetic consumer models are accurate enough for high-stakes product decisions. Traditional market research firms like Nielsen and Ipsos have spent decades building methodologies to account for response bias, social desirability effects, and the gap between stated preferences and actual behavior. Whether AI agent twins trained on historical consumer data can replicate those nuances, or simply reflect the biases baked into their training data, remains an open question.
A $2B Bet in Year One
A $2B valuation for a company that didn’t exist 12 months ago places Simile among the fastest-growing AI startups by valuation. For context, it took Anthropic two years to reach a comparable milestone. Greenoaks, which led the round, has a track record of early-stage bets on companies that establish new software categories (previous investments include Brex and Figma).
The round also signals growing investor appetite for “agentic simulation” as a category: AI agents used not to execute tasks, but to model the behavior of the humans those tasks are designed to serve.