How the read is built.
The RIDE Forecast is advisory by design. It informs a decision a human makes. It does not decide. This page is the plain-terms account of how the read is built, where the assumptions live, and what is still uncalibrated.
Two reads, one verdict
Read 1 is the leader. Four locked threshold items map onto an MBTI type and a home regime. A small set of allocation scenarios gives Home (the spike) and Range (the spread). A horizon item reads how your attention actually gets spent across now, next, and far.
Read 2 is the challenge. A short set of objective questions about the business places the company on a regime and a horizon mix.
The fit is the overlap. The score is the headline. The located gaps name the phase you are most likely to mishandle and the horizon most likely to go dark.
Where AI sits
The scoring is deterministic arithmetic. There is no model inferring traits from text. The locked threshold key, the allocation scoring, and the fit formula are all open code. The fit prior is hand-set in quartile steps and is reported as provisional on every artifact.
Authored copy in the forecast (the derived life) is written by humans against a locked voice register. The scaffold you see while that copy is in sign-off is generated from the same axis mechanics that produce the score, so the read is self-consistent.
What is uncalibrated
The 16x16 fit prior is a hand-set heuristic. Real calibration against outcome data is the open empirical job. Until that lands, the score reads as a wide confidence band, the forecast carries the provisional flag, and the verdict is positioned as a forecast worth pressure-testing, not a verdict to act on alone.
Per-archetype horizon mixes are a first-draft heuristic and are being validated.
Human-in-the-loop
You answer the assessment. You see the read. You decide. The product is free, owned by you, and built so the forecast is something you can take to a coach or a board conversation, not a verdict an employer pulls on you.
We do not run an employer-facing screening view, a candidate ranking, or a pass and fail verdict sent to a third party. Crossing any of those would change the regulatory weight of the product, and we would do that as a separate, deliberate programme.