Human preference. Measured at scale.

A model trained on our data beats every frontier model at predicting what people choose, even on content it never saw.

40M+
preference pairs
200K+
monthly users
20+
languages

Each dot is ten thousand preference pairs

Tested against frontier models

Frontier models misread people. Our data doesn't.

“Which option will this person pick?” On content never seen in training

Our data
45.2%
Claude Opus 5.5
40.0%
GPT-6
39.9%
Gemini 4
38.0%
Llama 4
35.1%
Bars start at pure guessing, 31% · 16,000 real decisions · open model trained on our data
5 of 5

frontier models beaten at predicting what real people choose, even on content our data never saw

+4 to 5

points better at predicting people when Claude Opus or GPT-6 is given what our data taught a model

70.5%

right at picking the answer people prefer, as a judge trained on our data. GPT-6 65.0%, Claude Opus 62.8%

Pre-registered tests · 95% confidence ranges · full results in the data room

Ask any population anything.

We design a gamified survey, run it with real users, and deliver labeled data. First data in two weeks.

01  Scope02  Design03  Ship04  Measure
Personality

A profile built from choices, not questionnaires.

Taste

What people choose, and what it predicts they will like next.

Social intelligence

How well people predict what others will choose, scored against a right answer.

Emotional intelligence

How people read tone, and how they react to it.

Boundaries

Where the line sits for different people, and how it moves with the room.

Social and political views

Where populations stand, inferred from choices and reported only in aggregate.

Also measurable: humor, brand affinity, trust, honesty under pressure

Before a model can move a population, someone has to measure what moves people. That is where OREX comes in.

Built the audience. Shipped AI at Amazon.

Nate Fox
Nate Fox
Founder & CEO

40M+ recorded choices, with zero paid acquisition.

Designed the identity, consent, and measurement architecture that turns choices into structured preference data. Patent pending on behavior-driven consent and identity capture. Previously led growth and product at Google, Lyft, and Robinhood.

Fred Clarke
Fred Clarke
Co-Founder & Head of Product

Shipped production AI at Amazon before it was a category.

Built multi-model orchestration inside Alexa in 2017, routing between competing models to choose the next step in a conversation. 25+ years across Amazon and Microsoft. 9 patents.

Every number tested on data it was not fit on. Five of our own findings retracted.

See the evidence

The human signal models are missing.

We publish our methods. Bring your hardest ML lead.

License a corpus

Choices with every alternative kept, a score for each judge, and the stated reason.

Commission a study

Any population, any question. First data in two weeks.

Evaluate a model

Score how a model reads people against a live human baseline.

Request a sample