# Simile tested digital twins against real people

Markdown copy of https://realalpha.app/s/simile-tested-digital-twins-against-real-people/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

- Speaker: [Joon Sung Park](https://realalpha.app/people/joon-sung-park/) · Joon Sung Park is cofounder and CEO of Simile.
- Host: [Molly O'Shea](https://realalpha.app/people/molly-oshea/) · Founder, Sourcery · Ex-Upfront Ventures investor; has interviewed Alex Karp and Palmer Luckey.
- Show: Sourcery
- Episode: [The $2B AI Startup Simulating Human Behavior with 85% Accuracy](https://realalpha.app/v/-6Kg5N7JL7I/), Oct 26, 2026
- Moment: 8:52 to 11:09 of the episode (2 min), playing on https://realalpha.app/s/simile-tested-digital-twins-against-real-people/; the episode from there: https://realalpha.app/v/-6Kg5N7JL7I/?t=532

## Quote

> “at the end of 2024, for the first time really in the field, we could actually show that we can **predict people's behavior** with 85% plus accuracy”
> — [Joon Sung Park](https://realalpha.app/people/joon-sung-park/) at [10:23](https://realalpha.app/v/-6Kg5N7JL7I/?t=623) · Sourcery

## Takeaways

1. The study recruited a representative sample of 1,000 people.
2. Participants returned after two weeks for behavioral tests.
3. Their digital twins predicted behavior with 85%+ accuracy.

## About this segment

Joon Sung Park describes training digital twins from interviews with 1,000 representative U.S. participants, then testing their predictions against those people. The study reached 85% accuracy and gave Simile its technical foundation.

From “The $2B AI Startup Simulating Human Behavior with 85% Accuracy”, Oct 26, 2026. [Watch the full episode](https://realalpha.app/v/-6Kg5N7JL7I/)

## Transcript

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

[8:57](https://realalpha.app/v/-6Kg5N7JL7I/?t=537) But then over time, we wanted to go beyond that. So here, really the initial evaluation really was about believability, that we start to see what we call the illusion of life in these agents. But as we mature as a field, what we started to do was to run evaluation that is now about more of a faithfulness and accuracy. So the experiment that we ended up running following that was we ran an experiment where we then took this architecture to create simulations of 1,000 real people.

[9:28](https://realalpha.app/v/-6Kg5N7JL7I/?t=568) So that started with us actually going out, recruiting a large number of real people that sample representatively from the U.S. So it's 1,000 representative samples of the U.S. population across gender, race, ethnicity, and so forth. And then we actually collected data from these people. So we would actually do, this time, human interview. So we actually built an AI agent that can talk to voice-to-voice with people, have those agents talk to people, voice to voice.

[9:54](https://realalpha.app/v/-6Kg5N7JL7I/?t=594) We would ask questions that's a little bit like the questions we ask agents in Smallville. Tell me the story of your life. Where did you grow up? What were some of the hard decisions you had to make? Who matters to you? We would use that data to then train AI agents that can basically be the twins of these people. And then we would send people away for two weeks, bring them back to a lab, and then this time we would actually ask people to complete a battery of surveys, behavioral studies, respond to stimuli, and had their agents predict what they would say. And at the end of 2024, for the first time really in the field, we could actually show that we can predict people's behavior with 85% plus accuracy with these agents. That's ultimately what gave the company foundation for today's technology. Of

[10:43](https://realalpha.app/v/-6Kg5N7JL7I/?t=643) course, the technology itself has shifted quite a bit, but that was the initial building block that we could see that, ah, these models and these agents can now actually predict your own behavior, and this is an amazing building blocks. But at the same time, it's a large data operation that we do not simply create simulations out of vacuum. We believe strongly that the simulation has to be grounded, and it has to be evaluated alongside people.