Where the evidence stands
We publish our own evidence, including what is not yet proven.
Most instruments in this market claim more than their data carries. Areté's standing rule is the opposite: never claim more than the evidence carries. Here is the ledger, plainly.
Measured
The mechanics are real. Cognitive signature from forced-choice calibration, state from your own baseline, HRV and sleep against your own norm, decisions stamped with the state they were made in. Every mechanism is test-pinned in code.
Synthetic
No accuracy figure on people exists yet. Our synthetic check answers a narrower question: how consistent the instrument is with itself, on answers made to follow its own rules. It is not a measure of how well it reads a person, and we do not quote it as one. Human validation is being set up.
Not yet
Human validation has not happened yet. A pre-registered protocol for the first labeled cohort (n≥20, blinded administration, triangulated reference labels) is published in full, here, before the first participant. Negative results will publish. That is the point.
Never
Areté is a self-insight instrument, not a clinical or diagnostic tool, and it never claims otherwise on any surface.
The founding validation cohort
Twenty operators and practitioners, named in the validation report if they wish, first to hold the instrument to its own standard. If honesty like this is what you want from your tools, apply.