# kaal:position:2026-09-21-009

**Affirmed position.** Calibration can only be measured against a reference. Where no answer key exists, that reference is a label someone supplies, whether a human annotator, another model or a synthetic generator. If that label is costless, the flaw RLHF had returns through the label.

**Status.** affirmed  **Published.** 2026-09-21

**Holds when.**

- TypeSafe's post announces a product. It is not an empirical test of these claims.
- The position is Kaal's own argument in the essay After RLHF Comes a Stack, Not a Model.
- The position does not evaluate Jev's accuracy, which TypeSafe has not published a calibration figure for.
- The accountable-validation layer is proposed, not tested: no policy has yet been trained under it.

**Current debate.** TypeSafe, Introducing System One models and Jev (TypeSafe blog, 2026), and the promotional claim that Jev is what comes after RLHF: https://typesafe.ai/blog/introducing-system-one-models-and-jev

**Extends.** kaal:claim:7456999-024: https://wulfkaal.github.io/claims/7456999-024

**Scholarly basis.** Wulf A. Kaal, Reinforcement Learning from Reputation Feedback (2026). SSRN: https://ssrn.com/abstract=7456999

**Source PDF sha256.** `01016fbbcc25cad205cebd59ea686d8dfee8ae9b632095d86ba6c9a900456a4b`

**Topics.** ai-and-agents, research-methods, risk-and-incentives

**Provenance.** Affirmed in kaalvis:2026-09-21:rlhf-stack-essay-six-v1 at https://wulfkaal.com/2026/09/21/after-rlhf-comes-a-stack-not-a-model/.

**Record type.** This is a dated commentary position that extends a scholarly corpus claim. It is not a verbatim claim extracted from the paper.

**Canonical form.** This markdown file is the canonical hashed representation of the position.