# Robots should learn the world, not copy human behavior

Markdown copy of https://realalpha.app/s/robots-should-learn-the-world-not-copy-human-behavior/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

- Speaker: [Bernt Børnich](https://realalpha.app/builders/bernt-bornich/) · Bernt Børnich is founder and CEO of 1X.
- Host: [Ti Morse](https://realalpha.app/people/ti-morse/) · Host, Relentless · Interviews hard-tech founders on Relentless; guests include Sam Altman, Brian Armstrong and Hadrian's Chris Power.
- Show: Relentless
- Episode: [Manufacturing 50,000 Humanoid Robots Next Year | Bernt Børnich, 1X](https://realalpha.app/v/DXSs6yIL00w/), Sep 18, 2026
- Moment: 28:51 to 31:17 of the episode (2 min), playing on https://realalpha.app/s/robots-should-learn-the-world-not-copy-human-behavior/; the episode from there: https://realalpha.app/v/DXSs6yIL00w/?t=1731

## Quote

> “you learn from all the data about **how people drive** it's very useful data”
> — [Bernt Børnich](https://realalpha.app/builders/bernt-bornich/) at [30:15](https://realalpha.app/v/DXSs6yIL00w/?t=1815) · Relentless

## Takeaways

1. Human demonstrations show what happens, not necessarily ideal behavior.
2. A model should learn how the world and other agents work.
3. The desired behavior can then be selected for a specific task.

## About this segment

Bernt Børnich argues robots should learn how the world and its consequences work, rather than copy human actions as the target. That can help them meet requested standards even when people's behavior is imperfect.

From “Manufacturing 50,000 Humanoid Robots Next Year | Bernt Børnich, 1X”, Sep 18, 2026. [Watch the full episode](https://realalpha.app/v/DXSs6yIL00w/)

## Transcript

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

[28:56](https://realalpha.app/v/DXSs6yIL00w/?t=1736) is not perfectly folded so if you emulated me like it's it's it's folded but it doesn't like you see creases you know if most people's environments are imperfect and the way that humans actually operate in the real world is just constantly imperfections how do you collect like the best real world perfection data i don't think you do um i think you just learn how the world works and and like we learned that like there will be wrinkles in your t-shirt if you do like you do and sometimes you will have the like random example where like it's actually straight right and and and both of them are equally important because how you learn how to get from one transition from one stage to another um and it more becomes like how do you ensure like that have the intelligence emerge over this so that if i say actually i want my laundry to be like

[29:53](https://realalpha.app/v/DXSs6yIL00w/?t=1793) perfectly straight then the robot understands how to do that it's a pretty good path there uh it's a pretty well explored problem especially self-driving is a good example of this like most drivers are terrible drivers for driving it's a very it's a very thing you you want the robot to do exactly you know perfect job every time yeah but like if you just clone human behavior in traffic you're not gonna get that no not at all uh so you can't actually just do that right you learn from all the data about how people drive it's very useful data uh you learn what happens if you do something you don't learn how to drive you learn the consequence of driving right then when you have that understanding of how to operate a car and how traffic works and how typically like other cars move and like how pedestrians like might move and like predicting

[30:44](https://realalpha.app/v/DXSs6yIL00w/?t=1844) all these things now you can learn how to operate in a manner that is desirable given that someone tells you what's the desired behavior but i don't think models necessarily should explicitly learn through cloning human behavior that's that's kind of like a very um that's how we used to do things like years ago but like that's not how you do these things anymore you learn how the world works and And then you can essentially search in that space for like whatever optimal solution that you would want to your problem.