SciFi Futures
← Capabilities
Social & Economic Systems/Integrated system

Generative-agent social simulation

Simulating populations of individually modeled people with AI agents to predict how a society responds to policies, events and information.

State of the art

Park et al. — Generative Agent Simulations of 1,000 People (2024) reproduce individual attitudes about as well as people's own two-week retest consistency.

What’s blocking it
  • Agents mirror LLM biases rather than real behavioral variance
  • No validation against real outcomes over months or years
  • Simulating millions of agents with interactions is costly

Who’s working on it

  • Joon Sung ParkResearcher
    Stanford University (founder and CEO, Simile)

    Built the Smallville generative agents (UIST 2023 best paper) and then agents from two-hour interviews with 1,052 people that replicated those individuals' survey answers

  • SimileStartup

    Commercialising interview-grounded agent populations for market and policy research; raised $100M Series A in February 2026 and $200M Series B at a $2B valuation in July 2026

  • Michael BernsteinResearcher
    Stanford University, Computer Science

    Co-author of the generative-agents and agent-bank work and the main voice on what these simulations can and cannot validly predict

  • Percy LiangResearcher
    Stanford University (Center for Research on Foundation Models)

    Co-founder of Simile and the evaluation methodologist pushing for held-out behavioural benchmarks rather than plausibility demos

  • Stanford HAI ↗Academic lab

    Hosts and funds the agent-simulation work and the survey-replication datasets that make external validation possible

  • Matthew SalganikResearcher
    Princeton University

    His Fragile Families Challenge (160 teams, PNAS 2020) is the benchmark result showing how poorly individual life outcomes are predicted, and the bar any agent simulation must beat

What to invest in

KindInvestmentScale
researchProspective forecasting tournaments pitting agent simulations against real policy outcomes$$ $10M – $100M

Sci-fi concepts in reach

Psychohistorydirect

Sources