Simulation: the new Scaling Law — Joon Sung Park, Simile AI
August 21, 2026
AI Summary
5 min read"Simulation is a lot like painting," says Joon Sung Park, CEO of Simile AI. "The best paintings teach you something deep about the subject." Park started his career as a professional painter before pivoting to computer science, and that artist's eye for capturing the "essential essence" of a person now drives his company's mission: building high-fidelity simulations of human beings. His 2023 paper on "generative agents" (the "Smallville" paper) became one of the most cited AI papers of the year, and his follow-up work showed that these digital twins can replicate a person's behavior and attitudes with 85% accuracy. The core insight: large language models are great at reasoning, but terrible at being you.
The Behavior Foundation Model
The fundamental problem, Park explains, is that frontier models like GPT-4 or Claude are trained to be "super rational objective machines." They excel at reasoning tasks because they're trained on web data—Wikipedia, social media, professional forums. But this data represents "self-exposed attitudinal data," not deep behavioral truth. "They have yet to learn really deep behavioral nature of people," Park says. "Not just what people say they don't mind, but what they actually do in real life."
Continue reading the full summary in the app — free to try.
Read Full Summary →Free • No credit card required
Never miss an episode of Latent Space: The AI Engineer Podcast
Get every new episode summarized in your inbox — free, ~5 minutes to read.
No spam. Unsubscribe anytime.
What you'll learn
- 1 Simulation: the new Scaling Law — Joon Sung Park, Simile AI
- 2 (00:02) **Introduction and Background** - Joon shares his path from artist to AI researcher
- 3 (01:46) **Origin of the Generative Agents (Smallville) Paper** - How the foundational paper came together
- 4 (05:19) **Simulation vs. Personal Assistants** - Why they chose world simulation over automation
- 5 (09:53) **When to Train vs. Prompt** - The intuition behind model-level changes
- 6 (11:23) **The Behavior Foundation Model** - Three buckets of data
- 7 (14:49) **How They Collect Behavioral Data** - Running real experiments with real stakes
+ Full timestamped outline available in the app
Guests on this episode
Show Notes
When we first dicsussed the Summer of Simulative AI in 2024 we knew it would be a brief summer, but it has recently come back with a vengeance with SimGym in April and now Simile AI’s $2B Series B, backed by GreenOaks and Index Ventures with prominent backers like Fei-Fei Li and Andrej Karpathy, running tens of millions of simulations for Fortune 100 clients like CVS and 85–99% accuracy vs human focus groups.
Time to catch up on why this Second Summer of simulation is working!
From creating Smallville, the landmark 2023 paper on Generative Agents that showed AI characters could remember, plan, socialize, and develop emergent behaviors, to now building foundation models of human behavior, Joon Sung Park is trying to answer a much bigger question: what if we could simulate the world before making decisions in it? In this episode, the Simile co-founder and CEO joins us to unpack the path from generative agents to digital twins, why today’s frontier models still fail to capture how humans actually behave, and what it would take to eventually simulate all 8 billion people on Earth.
We go deep on Simile’s approach to modeling human behavior: long-form interviews, observational and transaction data, randomized controlled trials, population-level and individual-level models, and post-training on the causal mechanisms behind why people make decisions. Joon explains how his research created digital twins that reproduced human behavior and attitudes 85% as accurately as people reproduced their own responses, why models optimized to be rational can be bad simulations of irrational humans, and why understanding “social physics” may require changing model weights rather than simply prompting frontier LLMs.
We also explore the much larger ambition behind simulation: testing products and policies before deploying them, finding counterintuitive paths toward desired outcomes, modeling emergent behavior across entire societies, and potentially tackling problems like climate change, democratic instability, and UBI. Joon reflects on scaling laws for simulation, the economics of data-center-scale simulated worlds, the connection to Thomas Schelling and psychohistory, why simulation is surp
More from this podcast
Latent Space: The AI Engineer Podcast →