Simile (https://simile.ai), a Palo Alto and New York-based AI simulation company, has raised $200M in Series B funding led by Greenoaks at a $2B valuation. Simile builds the world's first AI simulation of society populated by generative AI agents based on real humans. Companies use these 'agentic twins' to test decisions before real-world rollout. The capital will accelerate its foundation model for human behavior and scale deployments for Fortune 100 clients.
Rapid Series B Follows Hypergrowth
The timing comes amid surging interest in synthetic user research. Aaru raised $50M Series A in December 2025 at a $1B valuation, while BluePill AI secured $6M Seed in November 2025. Simile's approach grounds agents in real human interviews and transaction data rather than generic LLM personas, addressing gaps in fidelity for high-stakes enterprise decisions.
Traditional Research Cycles Collapse
Market research services reached $96.77B in 2026. Response rates for traditional surveys have collapsed into single digits while panel costs have risen sharply. Enterprises face weeks or months of delays for insights that often fail to predict actual behavior. Only 8% of researchers fully trust synthetic alternatives, creating a trust gap even as 89% now use AI tools.
Grounded Digital Twins Drive Accuracy
Simile trains a proprietary foundation model on consented interviews and behavioral data to create digital twins that replicate how specific individuals respond to products, policies, or environments. CVS Health built 100,000 AI patient twins from 2.9M records to test pharmacy experiences, replicating in hours what previously took months. This differs from competitors that rely on prompt-engineered personas without real data grounding.
"Our mission is to simulate all eight billion people on Earth, accurately."
Greenoaks and Index Signal Conviction
Greenoaks led the round, co-leading with Index Ventures. Participants include Hanabi Capital, A*, BCV, CVS Health Ventures, and Definition. The $300M total raised in five months from Series A signals strong investor belief in simulation as the next frontier beyond language models.
Synthetic Research Attracts Billions
The broader synthetic data generation market stands at $528M to $791M in 2026 and is projected to grow at 31-35% CAGR. The synthetic user research category has drawn over $1.5B in venture funding. Structural drivers include collapsing survey response rates, LLM accuracy reaching 85% on benchmarks like the General Social Survey, and demand for faster decision intelligence across healthcare, finance, and policy.
Stanford Roots Anchor Technical Moat
CEO Joon Sung Park pioneered the generative agents field through his Stanford PhD project Smallville. Co-founders include Stanford professors Michael Bernstein and Percy Liang. This academic foundation underpins the company's 85% accuracy claims and enterprise traction with clients including Gallup, Wealthfront, Suntory, and Deloitte.
Explosive Traction Validates Model
Revenue grew 5x within five months of the February 2026 launch. Tens of millions of simulations have run for Fortune 100 organizations, and the company grew to over 50 employees. Selection for the Disney Accelerator further validates the platform's strategic potential.