Blind review on The Broken-Born program

The blind-review protocol and verdict scale

Mon, 01 Jan 0001 00:00:00 +0000

The blind-review protocol is the fixed procedure the program uses to test a manuscript: a fresh AI subagent receives a copy of the manuscript and the protocol and nothing else, checks every citation against its source, re-runs every number in its own code, re-derives the mathematics independently, audits the argument for consistency and omitted constraints, and returns a grade of A, B or C with the single weakest link named. The author then rules on each finding, re-running disputed numbers in separate scripts, and scores the forecast that was committed before the referee was dispatched. The grade scale comes from the referee prompt of the skeptic review console, the tool of rounds 1 and 2; rounds 3 and 4 ran the full five-step form with a referee that could execute code. The protocol’s own stated limit is that the referee is the same model family as the author.

The skeptic review console

Mon, 01 Jan 0001 00:00:00 +0000

The skeptic review console is a single HTML page, artifact #15 (review/skeptic-review-console.html), that sends a manuscript to a fresh Claude model with no conversation context, an adversarial referee prompt and web search turned on, and prints the review it gets back. It was the tool of review rounds 1 and 2, both on #14, the standalone manuscript proposing superheavy hidden-sector solitons as dark matter. Its prompt defines the A/B/C verdict scale that the later rounds kept. The referee it calls can search the web but cannot execute code, the limitation behind the referees’ arithmetic errors in those rounds. The page embeds #14 version 2 verbatim and is kept as the record of rounds 1 and 2; the proposal in it has since been retired.

Pre-registration and its calibration record

Mon, 01 Jan 0001 00:00:00 +0000

Pre-registration is the practice of writing down, and committing to the repository, the verdict a blind review is expected to return and the weakest point the referee is expected to find, before the referee is dispatched. Because the forecast is fixed in advance it cannot be adjusted to the outcome, and scoring it afterwards measures how well the author judges the author’s own work. The record so far: the verdict was forecast correctly in rounds 1 and 2, missed in round 3 and forecast correctly in round 4; the location of the weakest link is now forecast better, and its severity is still underestimated. Since round 3 the forecast is kept out of the commit subject, because a referee subagent sees recent commit subjects.

The measured referee failure mode

Mon, 01 Jan 0001 00:00:00 +0000

The measured referee failure mode is the program’s record of where its blind reviews have gone wrong, and on which side. In rounds 1 and 2 the referees, which could search the web but not run code, were reliable on structure, literature and logic and unreliable on arithmetic: their three confident order-of-magnitude indictments were all their own unit errors, and none survived. That finding made the iron rule load-bearing and was the main reason the project moved to Claude Code, where a referee can execute code. In round 3 the failure moved to the author’s side: the author’s scripts had checked the arithmetic of a model with a wrong physics input, so a check that tests only arithmetic passed it. Round 4’s decisive defect was again on the author’s side, in a check run on a reduced model.

Review rounds 1 and 2

Mon, 01 Jan 0001 00:00:00 +0000

Review rounds 1 and 2 were the program’s first two blind reviews, both of #14, the standalone manuscript proposing superheavy hidden-sector solitons as dark matter. Both ran through the skeptic review console: a fresh referee with web search and no code execution. Round 1 graded version 1 B, flawed but salvageable, with medium confidence, and every valid finding was folded into version 2; round 2 graded version 2 “B near B/C” and called the derivation of the hidden coupling numerology. Round 2’s findings became the pending list R1–R7, from which version 3 was cut for round 3. The rounds are known only from the handoff’s summary: the referees’ reports and the full adjudications were never brought into the repository.

Review round 4

Mon, 01 Jan 0001 00:00:00 +0000

Review round 4 was the blind review of the toy-model note, papers/reciprocity-toy-models.md, which builds two explicit field theories in which Born reciprocity is a symmetry of the action and asks whether the lowest-energy state breaks it. A fresh referee that could execute code reproduced the note’s algebra and most of its numerics, but found that the note’s one claim of quantum symmetry breaking rested on a potential that is unbounded below once the field H is allowed to vary. It graded the note B, flawed but salvageable, and the author agreed; the pre-registered forecast had also been B, though it had placed the weakest link correctly and underestimated how serious it was. Every revision was accepted; version 2 of the note applied them on 2026-10-20, with three departures, and the round added two rules to the project’s method.