SimReal: RL environments where AI agents learn real work and are graded by what actually happens
What made chess, Go, and code fall to AI?
They gave intelligence a world that answered back. A winner. A failing test. A way to try again and get better.
Humanity has always learned this way. A bridge that held. A harvest that failed. A decision that changed a life. The world turned our experience into judgment.
But beyond games and code, the world’s answers are harder to read.
We build worlds that answer back. Real economic work, recreated. Closing the books. Pricing a policy. Paying a claim. The same situation, replayed. One decision changed. The real rules determine what follows.
A thousand chances to learn before the stakes are ours.
We turn humanity’s hard-won experience into worlds intelligence can learn from, so it can help us build, discover, and care for one another at a scale we have only imagined.
Join us. Let’s give intelligence a world to grow in.
Built by a team with experience from
Connecting
Stanford University
Data and products supported by people from
University logos show a selection of the 21 universities in our network. Logos identify where members of our network studied or work, and the universities whose student societies partner with us. They do not imply endorsement by these institutions or firms.
