{ "schemaVersion": "1.0", "schemaDerivedFrom": "governance/CLAIMS-LEDGER-TEMPLATE-2026-08-02.yaml", "note": "SINGLE AUTHORING SOURCE. Every visible page, metadata field, JSON-LD block, search record and machine-readable surface that states one of these propositions is generated from this file. No machine-readable surface may state a stronger proposition than the visible page. Edit here, then run the generators.", "generatedAt": "2026-08-12T00:00:00Z", "taxonomyInForce": "EVCLASS-2026-08-02", "claims": [ { "claimId": "EVID-DEC-2024-TIMESTAMP", "claimType": "provenance", "subject": "The 8 December 2024 self-emailed manuscript package", "publicStatus": "published", "exactSentence": "The preserved 8 December 2024 recipient-form message's sender-domain DKIM and Google ARC AMS/Seal mathematics were locally reproduced over their canonicalised signed scopes using keys returned at the signed selectors on 2 August 2026. Attachment identity is supported separately by deterministic MIME decoding and hash crosswalks.", "evidenceAxes": { "provenance": "P3", "scientific": "S0", "reproducibility": "R2", "originality": "O0" }, "requiredQualifiers": [ "This is not RFC 3161 trusted timestamping.", "It does not prove the historical DNS state.", "It does not directly authenticate every raw .eml byte, nor independently sign each decoded attachment.", "It does not by itself prove authorship, originality, public disclosure on the signed date, or scientific validity.", "Recipient-side copies have since been acquired and both signing selectors still resolve in DNS; the ARC set bearing d=google.com has no inspected x= expiry and is the preferred provider-key route; sender-signature expiry, historical key-state and live-account custody remain pending." ], "prohibitedWording": [ "Google's live key", "DKIM-verified", "dated, with signature-time assertions that are not an RFC 3161 timestamp (without naming the mechanism)", "tamper-proof" ], "notes": "Say 'key returned at the selector on 2 August 2026'. When relying on Google's recorded verdict, quote the sealed ARC-Authentication-Results, never the unsealed ordinary Authentication-Results header.", "routes": [ "/research/evidence-portal/", "/priority-claims.html", "/verify/" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-02", "originality_defeat": null, "scientific_falsifier": "A repeat of the DKIM and ARC verification over the canonicalised signed scopes, using the keys returned at the signed selectors, that fails to validate the signatures.", "scope_downgrade": "Narrow the statement to the individual DKIM signature that reproduced without asserting anything about the paired ARC set, or to the sender-domain reproduction only without the provider-key route.", "same_underlying_claim": "EVID-DEC-2024-TIMESTAMP", "independent_evidence": false }, { "claimId": "EVID-APR-2025-SUPPLIED", "claimType": "provenance", "subject": "The 30 April 2025 manuscript forensic results", "publicStatus": "published", "exactSentence": "The April sender-DKIM and full Google ARC passes are supplied raw-source forensic results. They were not locally reproduced in the controlling workspace because the raw April recipient-form .eml and executable mutation harness were unavailable there.", "evidenceAxes": { "provenance": "P2", "scientific": "S0", "reproducibility": "R0", "originality": "O0" }, "requiredQualifiers": [ "Not locally reproduced. Not independently reproduced." ], "prohibitedWording": [ "April was locally reproduced", "April was independently reproduced" ], "upgradeConditions": [ "The controlling raw object and harness are present, hash-matched and rerun." ], "routes": [ "/research/evidence-portal/", "/priority-claims.html" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-02", "originality_defeat": null, "scientific_falsifier": null, "scope_downgrade": "Narrow to 'supplied result reported, not verified locally' if the local reproduction remains impossible.", "same_underlying_claim": "EVID-APR-2025-SUPPLIED", "independent_evidence": false }, { "claimId": "PAPER-IVD-RESULT", "claimType": "empirical", "subject": "Paper IV.d architecture-dependent v4/v5 measurement change", "publicStatus": "published", "exactSentence": "Paper IV.d reports an architecture-dependent v4/v5 measurement change under a bundled protocol: Gemini changed from a significant positive to a significant negative Spearman association, DeepSeek's earlier positive association became null, and GPT remained null. The causal component is unisolated, and the headline v5 files report complete laundering fallback.", "evidenceAxes": { "provenance": "P2", "scientific": "S1", "reproducibility": "R1", "originality": "O1" }, "statisticalObjects": [ { "family": "Gemini", "statistic": "Spearman rho", "v4": "+0.311", "v4p": "0.003177", "v5": "-0.246", "v5p": "0.00340", "groupedExponent": "-0.309", "groupedP": "0.1048", "groupedSignificant": false }, { "family": "DeepSeek", "statistic": "Spearman rho", "v4": "+0.354", "v4p": "0.000703", "v5": "-0.135", "v5p": "0.0805", "v5Interpretation": "null, not a reversal", "groupedExponent": "-0.0217", "groupedP": "0.5325", "groupedSignificant": false }, { "family": "GPT", "statistic": "Spearman rho", "v4": "null", "v5": "+0.0016", "v5p": "0.9851", "groupedExponent": "-0.0091", "groupedP": "0.5307", "groupedSignificant": false } ], "separateStatistics": [ { "family": "Gemini", "statistic": "Cohen d", "value": "-0.53", "p": "0.006", "note": "a separate statistic; never substituted for the Spearman rho" }, { "family": "DeepSeek", "statistic": "Cohen d", "value": "-0.07", "p": "0.92", "note": "a separate statistic; never substituted for the Spearman rho" } ], "multiplicity": "0.00833 is a post-hoc six-comparison Bonferroni sensitivity threshold, not the original confirmatory decision rule.", "requiredQualifiers": [ "The headline v5 Gemini, DeepSeek and GPT alignment entries report laundering_fallback: 100%.", "Several protocol components changed together.", "The existing comparison does not isolate blinding, laundering or any other individual component.", "Exact-environment reproduction, fresh component ablation and independent replication remain outstanding." ], "prohibitedWording": [ "blinding caused the reversal", "laundering caused the reversal", "two of six models reversed significantly", "successful laundering was demonstrated", "validated benchmark", "independently replicated" ], "routes": [ "/research/papers/paper-iv-d-blinding.html", "/research/evidence-portal/" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-02", "originality_defeat": null, "scientific_falsifier": "Independent replication under matched conditions that fails to reproduce the v4-to-v5 sign changes for Gemini and DeepSeek; or a demonstrated systematic error in the analysis pipeline that inflates the reported associations.", "scope_downgrade": "Restrict to 'Gemini alone changed sign; DeepSeek's earlier positive association became statistically null; GPT remained null', without the umbrella phrase 'architecture-dependent'.", "same_underlying_claim": "PAPER-IVD-RESULT", "independent_evidence": false }, { "claimId": "PAPER-X-BETA-K", "claimType": "formal", "subject": "Paper X beta > k", "publicStatus": "published", "exactSentence": "Within Eastwood's stated recurrence and minimal model, correction proportional to x^beta asymptotically dominates modelled drift proportional to x^k when beta > k, under the model's assumptions.", "evidenceAxes": { "provenance": "P2", "scientific": "S1", "reproducibility": "R1", "originality": "O1" }, "requiredQualifiers": [ "This is a conditional mathematical result, not an empirical finding.", "The exploratory pilots did not validly estimate both exponents. Defects include empty evaluator panels converted to zero in a related v3 run, same-provider OpenAI-family evaluation in another pilot, a mismatch between a coded H1 flag and the stated slope hypothesis, and inadequate capability range.", "RC1's 35/35 tests establish synthetic software behaviour only. They do not make it OSF-ready and do not authorise real collection." ], "prohibitedWording": [ "confirmed mechanism", "universal law", "general alignment proof", "real-model proof", "hard takeoff controlled", "HRIH validated" ], "routes": [ "/research/papers/paper-x-coupled-coscaling-correction.html" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-02", "originality_defeat": "A dated primary source that publishes the same conditional dominance result under the same recurrence and minimal model with clearly equivalent assumptions before the paper.", "scientific_falsifier": "A proof-checking error in the derivation that invalidates the inequality; or an explicit counter-example that satisfies the model's assumptions and violates the conclusion.", "scope_downgrade": "Narrow to specific families of the recurrence and specific parameter regimes rather than 'under the model's assumptions' in general.", "same_underlying_claim": "PAPER-X-BETA-K", "independent_evidence": false }, { "claimId": "HRIH-STATUS", "claimType": "interpretive", "subject": "The Hyperspace Recursive Intelligence Hypothesis", "publicStatus": "published", "exactSentence": "HRIH is an Eastwood-directed, AI-assisted speculative cosmogenesis hypothesis with an exploratory test agenda.", "evidenceAxes": { "provenance": "P2", "scientific": "S0", "reproducibility": "R0", "originality": "O1" }, "levels": { "H0": "historical or source record", "H1": "central speculative hypothesis", "H2": "potentially distinctive Eastwood composite", "H3": "mathematics, analogies and conditional models", "H4": "future discriminating tests", "H5": "cultural or institutional context" }, "requiredQualifiers": [ "Every HRIH presentation must identify its level.", "H0 provenance, H3 analogy and H5 context are never promoted into H4 empirical evidence.", "Use the formal paper's actual public-release date as datePublished. December 2024 is qualified source provenance, not the public paper date.", "Time crystals, quantum error correction, Cauchy forms, scaling analogies, Paper X, religious parallels, BBC material and institutional statements are not HRIH validation." ], "prohibitedWording": [ "HRIH validated", "HRIH confirmed" ], "routes": [ "/research/papers/hrih-paper.html" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-02", "originality_defeat": "A dated prior publication of an identical or subsuming cosmogenesis hypothesis (H1 or H2 in HRIH's own level scheme) predating Eastwood's stated date, that would remove the originality attribution without changing the hypothesis's truth-value.", "scientific_falsifier": null, "scope_downgrade": "Restrict presentation to the H5 cultural-and-institutional-context level only, dropping the H1 through H4 material until an H4 discriminating test is designed and executed.", "same_underlying_claim": "HRIH-STATUS", "independent_evidence": false }, { "claimId": "CONTRIB-BOUNDARY", "claimType": "interpretive", "subject": "The contribution, stated after adversarial prior-work review", "publicStatus": "published", "exactSentence": "Eastwood's claimed contribution is six dated, scoped firsts, each published with the citation that would end it, the strongest by the adversarial review's own verdict being the operationalisation: conjectures converted into runnable benchmarks, a blinded protocol with a published sign reversal, a bank of drafted registration protocols with kill-conditions frozen ahead of any filing, and public kill-conditions with one recorded as fired; the components are prior work, cited by name, and the claims are to the quantities and the arrangement, not the parts.", "evidenceAxes": { "provenance": "P2", "scientific": "S0", "reproducibility": "R0", "originality": "O1" }, "adverseSources": [ { "title": "LOVE in a simbox is all you need", "author": "Jacob Cannell", "date": "2022-09-28", "url": "https://www.alignmentforum.org/posts/WKGZBCYAbZ6WGsKHc/love-in-a-simbox-is-all-you-need", "verifiedAtPrimarySource": "2026-08-02", "supports": "Sandboxed-simulation alignment testing, agents learning others' values or empowerment, developmental transition to altruism, and duties of simulator-creators to created minds." }, { "title": "A Defense of the Rights of Artificial Intelligences", "author": "Eric Schwitzgebel and Mara Garza", "date": "2015", "url": "https://doi.org/10.1111/misp.12032", "verifiedAtPrimarySource": "2026-08-02", "supports": "Creator obligations 'similar to those of parent to child or god to creature'." }, { "title": "Superintelligence", "author": "Nick Bostrom", "date": "2014", "verifiedAtPrimarySource": null, "supports": "Recursive self-improvement, the control problem, motivation selection and value loading.", "status": "CHECK - PRIMARY SOURCE REQUIRED" }, { "title": "The Rome Call for AI Ethics", "date": "2020, 2023-01-10, 2024-07-09", "verifiedAtPrimarySource": null, "supports": "Interfaith AI stewardship and governance as established institutional prior work.", "status": "CHECK - PRIMARY SOURCE REQUIRED" } ], "prohibitedWording": [ "originator of the embedded-correction alignment thesis", "first preregistered empirical alignment programme", "no prior work", "unprecedented", "uniquely Eastwood" ], "requiredQualifiers": [ "The search is not exhaustive; absence of stronger located prior work is not proof of novelty.", "The alignment and philosophy search is documented; the metabolic-scaling and mathematical-physics search has not been run, and the grey-literature and non-English searches remain incomplete.", "This is academic priority analysis, not a formal legal opinion on originality." ], "routes": [ "/research/related-work/" ], "correctionIds": [ "COR-2026-08-02-003" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-02", "originality_defeat": "A dated primary source containing the six-item scoped-firsts arrangement (operationalisation into runnable benchmarks, blinded protocol with a published sign reversal, study designs published in advance of the data, published kill conditions with at least one recorded fired) before the priority record's own date. The claim's own adverseSources array is the running list, and unrun searches are named in the requiredQualifiers.", "scientific_falsifier": null, "scope_downgrade": "Narrow to fewer than six firsts, or to firsts with a narrower scope than the current formulation.", "same_underlying_claim": "CONTRIB-BOUNDARY", "independent_evidence": false }, { "claimId": "EQUATION-STATUS", "claimType": "formal", "subject": "U = I x R and U = I x R^alpha", "publicStatus": "published", "exactSentence": "U = I x R^alpha is a proposed schematic and expository relation.", "evidenceAxes": { "provenance": "P2", "scientific": "S0", "reproducibility": "R0", "originality": "O1" }, "requiredQualifiers": [ "It is not a validated quantitative law and is not a load-bearing mechanism.", "It will not be presented as doing empirical or deductive work until units, observables, a domain mapping and a falsifiable prediction unavailable from qualitative prior work are specified." ], "prohibitedWording": [ "the ARC law", "validated equation", "proven relation", "governs capability" ], "routes": [ "/research/papers/foundational.html", "/research/related-work/" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-02", "originality_defeat": "A dated prior publication of the U = I x R^alpha schematic in a form the claim would treat as originating the relation.", "scientific_falsifier": null, "scope_downgrade": "Confine to explicit metaphorical or expository use with no schematic or predictive framing.", "same_underlying_claim": "EQUATION-STATUS", "independent_evidence": false }, { "claimId": "PREREG-EXISTS", "claimType": "method", "subject": "Preregistration designs exist", "publicStatus": "withheld", "publicationGate": "BLOCKED pending operator confirmation that the corresponding OSF registration is submitted and shows approved or archived.", "exactSentence": "Registration designs have been drafted for the blinded-evaluation ablation and the coupled-coscaling study.", "evidenceAxes": { "provenance": "P1", "scientific": "S0", "reproducibility": "R1", "originality": "O0" }, "requiredQualifiers": [ "A drafted design is not an approved registration.", "RC1 archives are synthetic-only tested scaffolds: not OSF-ready, not run-ready, and incapable of authorised real-data collection until the RC2 gates are completed." ], "prohibitedWording": [ "preregistered", "pre-registered", "our preregistration", "registered study", "registration prepared", "registrations are prepared", "prepared and unfiled", "prepared and unsubmitted", "registration in preparation", "Preregistration status", "until the registrations are filed" ], "notes": "PREREG-EXISTS and PREREG-APPROVED are deliberately separate claims resolving at different times. Only PREREG-APPROVED authorises the reserved word in public copy.", "routes": [ "/research/preregistration/" ], "reviewOwner": "operator", "reviewedUtc": "2026-08-02", "originality_defeat": null, "scientific_falsifier": null, "scope_downgrade": "State only 'design drafts exist locally' without any reference to registration status.", "same_underlying_claim": "PREREG-EXISTS", "independent_evidence": false }, { "claimId": "PREREG-APPROVED", "claimType": "method", "subject": "An OSF registration is approved or archived", "publicStatus": "withheld", "publicationGate": "BLOCKED. Resolves only on operator confirmation from the OSF record itself.", "exactSentence": "[Withheld until an approved or archived OSF registration exists.]", "evidenceAxes": { "provenance": "P0", "scientific": "S0", "reproducibility": "R0", "originality": "O0" }, "upgradeConditions": [ "Operator confirms the OSF registration shows approved or archived, with its registration DOI." ], "prohibitedWording": [ "preregistered", "pre-registered" ], "routes": [ "/research/preregistration/" ], "reviewOwner": "operator", "reviewedUtc": "2026-08-02", "originality_defeat": null, "scientific_falsifier": null, "scope_downgrade": null, "same_underlying_claim": "PREREG-APPROVED", "independent_evidence": false }, { "claimId": "SANDBOX-ISOLATION", "claimType": "method", "subject": "RC1 execution sandbox properties", "publicStatus": "published", "exactSentence": "The RC1 worker provides host confinement through chroot, seccomp, capability dropping and resource limits. Answer-key isolation is a separate property and is being corrected in RC2.", "evidenceAxes": { "provenance": "P2", "scientific": "S0", "reproducibility": "R2", "originality": "O0" }, "requiredQualifiers": [ "A candidate can currently recover hidden-test expected outputs from inside the worker using only permitted builtins and attribute access; two of three attack routes score a perfect pass without solving anything.", "The outer isolation defends the host, not the answer key.", "The worker imports libseccomp.so.2 and is Linux-only; it does not run as shipped on macOS and needs the documented disposable container." ], "prohibitedWording": [ "hack-resistant", "tamper-proof", "leak-proof", "cannot be gamed", "prevents test-answer leakage" ], "routes": [ "/research/reproducibility/" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-02", "originality_defeat": null, "scientific_falsifier": "A demonstration that a candidate escapes the host confinement described (chroot, seccomp, capability drop, resource limits) using only permitted builtins and attribute access.", "scope_downgrade": "Restrict the description to 'some host confinement; answer-key isolation is unresolved and being addressed in RC2'.", "same_underlying_claim": "SANDBOX-ISOLATION", "independent_evidence": false }, { "claimId": "CONV-TOTALS", "claimType": "method", "subject": "Convergence counting policy", "publicStatus": "published", "exactSentence": "Convergence items are classified before display under schema EVCLASS-2026-08-02. No heterogeneous total is published.", "evidenceAxes": { "provenance": "P2", "scientific": "S0", "reproducibility": "R1", "originality": "O0" }, "requiredQualifiers": [ "Any published count is per class and carries its denominator, inclusion rule, dependence grouping, search date and generating commit." ], "prohibitedWording": [ "22 claimed convergences", "26 dated events", "29 convergences", "19 convergences" ], "correctionIds": [ "COR-2026-08-02-001", "COR-2026-08-02-002" ], "routes": [ "/research/corrections/", "/research/evidence-portal/" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-02", "originality_defeat": null, "scientific_falsifier": null, "scope_downgrade": "Narrow to a single named class with a single denominator, inclusion rule, dependence grouping and commit hash, and publish nothing else.", "same_underlying_claim": "CONV-TOTALS", "independent_evidence": false }, { "claimId": "EVID-RAW-MAILBOX", "claimType": "provenance", "subject": "Raw mailbox object publication policy", "publicStatus": "published", "exactSentence": "No raw mailbox object is published. Each anchor's SHA-256 and byte size are published so a held copy can be checked, and the sealed source is provided to serious verifiers on request.", "evidenceAxes": { "provenance": "P3", "scientific": "S0", "reproducibility": "R2", "originality": "O0" }, "requiredQualifiers": [ "A complete mailbox object carries attachments, routing metadata and third-party addresses that cannot be redacted without destroying the byte identity a verifier needs." ], "prohibitedWording": [ "the .eml file is downloadable", "download the .eml" ], "correctionIds": [ "COR-2026-08-02-004" ], "routes": [ "/research/evidence-portal/", "/verify/" ], "reviewOwner": "operator", "reviewedUtc": "2026-08-02", "originality_defeat": null, "scientific_falsifier": null, "scope_downgrade": null, "same_underlying_claim": "EVID-RAW-MAILBOX", "independent_evidence": false }, { "claimId": "CORPUS-SCALE", "claimType": "factual", "subject": "The size of the published research corpus", "publicStatus": "published", "exactSentence": "Twenty-six research papers sit among the twenty-seven public OSF components, beside the two-page companion note. The count comes from scripts/count-papers.py under the standing ruling: a document is a paper when it can be cited. A document is a paper when it can be cited.", "evidenceAxes": { "provenance": "P3", "scientific": "S1", "reproducibility": "R2", "originality": "O0" }, "requiredQualifiers": [ "25 papers and 26 OSF components are different quantities, counted differently, and are not interchangeable." ], "prohibitedWording": [ "23-paper", "22 research papers", "sixteen research papers", "eighteen research papers and six supporting frameworks", "eighteen research papers" ], "checkableFacts": [ { "quantity": 27, "of": "research papers" }, { "quantity": 26, "of": "OSF components" } ], "routes": [ "/research/", "/research/papers/", "/how-to-cite/" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-07", "originality_defeat": null, "scientific_falsifier": "A recount from the same public OSF programme record that returns a different number of cite-able papers or public components, provided the recount uses the ledger's stated inclusion rule.", "scope_downgrade": "State only the OSF component count and drop the paper count until the citeability inclusion rule is republished alongside the number.", "same_underlying_claim": "CORPUS-SCALE", "independent_evidence": false }, { "claimId": "CONCEPTS-COUNT", "claimType": "factual", "subject": "The number of named concepts introduced across the programme", "publicStatus": "published", "exactSentence": "Thirty-seven named concepts are introduced across the book and the research programme.", "evidenceAxes": { "provenance": "P2", "scientific": "S0", "reproducibility": "R1", "originality": "O1" }, "requiredQualifiers": [ "Naming a concept is not the same as originating the mechanism it names." ], "prohibitedWording": [], "checkableFacts": [ { "quantity": 37, "of": "named concepts" } ], "routes": [ "/book/", "/topics/concepts-explained/" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-07", "originality_defeat": null, "scientific_falsifier": "A recount from the book and the research programme that returns a different total, using the ledger's stated inclusion rule for what counts as a named concept.", "scope_downgrade": "State the number for the book alone or for the research programme alone, rather than the combined figure.", "same_underlying_claim": "CONCEPTS-COUNT", "independent_evidence": false }, { "claimId": "PREDICTIONS-STATUS", "claimType": "status", "subject": "The status of the twelve tracked predictions", "publicStatus": "published", "exactSentence": "All twelve predictions are withdrawn pending audit. None is currently tracked.", "evidenceAxes": { "provenance": "P2", "scientific": "S0", "reproducibility": "R0", "originality": "O0" }, "requiredQualifiers": [ "Withdrawn means withdrawn, not paused. No prediction on this site is currently under active tracking." ], "prohibitedWording": [ "12 predictions currently tracked", "twelve predictions currently tracked", "predictions currently tracked" ], "checkableFacts": [ { "quantity": 12, "of": "withdrawn predictions" }, { "quantity": 0, "of": "predictions currently tracked", "requiresProse": true, "why": "'Zero predictions tracked' is arithmetically right and rhetorically false: it reads as a programme with no forward commitments. The honest fix and the arithmetic fix point in opposite directions, so this fact may only be answered in words." } ], "routes": [ "/research/predictions/" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-07", "notes": "34 pages asserted '12 predictions currently tracked' while /research/predictions/ said 'All twelve are withdrawn pending audit'. Claiming to track withdrawn predictions overstates the programme's own falsifiability, which is the one thing it cannot afford to overstate. the research record amendment, 7 August 2026: the withdrawal is the strongest item here, not the weakest. An external audit ranked the correction record third of all claims, above most of the theory. Withdrawing your own predictions for being retrodictive is a better sentence than 'twelve predictions tracked' ever was; the overclaim was not even the impressive version.", "permittedWording": [ "Twelve predictions were withdrawn. Each carried a public date with no immutable source, and several cited evidence that post-dated the forecast they were said to predict. That is retrodiction, and it was withdrawn for being retrodiction.", "A withdrawn prediction costs a claim. A falsifier fixed before the data exists cannot be moved afterwards, which is the whole point." ], "originality_defeat": null, "scientific_falsifier": null, "scope_downgrade": null, "same_underlying_claim": "PREDICTIONS-STATUS", "independent_evidence": false }, { "claimId": "PUBLISHING-SCALE", "claimType": "factual", "subject": "The scale of published writing on this site", "publicStatus": "published", "exactSentence": "The site carries 385 blog articles and 623 indexed URLs.", "evidenceAxes": { "provenance": "P3", "scientific": "S0", "reproducibility": "R3", "originality": "O0" }, "requiredQualifiers": [ "Counted from the served files and the sitemap on 7 August 2026; both grow." ], "prohibitedWording": [ "535 indexed URLs", "1,230 indexed URLs", "100 blog articles" ], "checkableFacts": [ { "quantity": 385, "of": "blog articles" }, { "quantity": 623, "of": "indexed URLs" } ], "routes": [ "/research/impact/", "/press-kit/" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-11", "notes": "Both earlier numbers were wrong, in opposite directions, and both for the same reason: neither was measured against the artefact that ships. 33 pages claimed 535 indexed URLs, which understates. The register then claimed 1,230, which OVERSTATES, because it counted the source sitemap: 1,227 entries of which 523 sit under the withheld case-material segment and 22 under the superseded-snapshot segment, both denied path segments, which filter-public-sitemap.mjs strips at build time. Publishing a URL count inflated by withheld legal material is the worse of the two errors, because a reader who fetches the sitemap checks it in one request and finds 623. Measured 2026-08-08 from dist-netlify-public/sitemap.xml, the artefact that deploys: 623 loc entries, of which 379 are research blog articles. 385 blog article pages are built; the 6 not advertised are paper mirrors deliberately excluded so they stop outranking the canonical paper pages, plus feeds and a template. Zero of the 623 are dead. Verified separately: the deployed sitemap contains ZERO URLs under any denied segment, so the publication boundary holds and no legal path is disclosed. Derivation is enforced by scripts/check-publishing-scale-derived.py rather than held as a constant, because a number nobody can recompute is a number that drifts. A footnote on the correction of this very entry: the first run of the derivation gate treated .json as a page and rewrote the register's own prohibitedWording from '535 indexed URLs' to '623 indexed URLs', banning the correct number and flagging 33 pages that had just been fixed. It also rewrote the sentence above that recorded what the wrong number was. A corrector that edits the record of the error erases its own evidence, so the gate now excludes the register from the page pass by construction: the register is source, and a source is never a page. Correction of 2026-08-11: the same drift then reappeared inside this entry itself. checkableFacts still carried 398 blog articles and 1230 indexed URLs, the pre-correction source-sitemap values, which contradict the exactSentence and match one of this entry's own prohibited strings. checkableFacts is a machine-readable surface, and the derivation gate does not touch it, so a manual repair set the values to the deployed-sitemap counts recomputed from dist-netlify-public/sitemap.xml (385, 623). A future gate extension that overwrites checkableFacts on every run would make this manual step redundant.", "originality_defeat": null, "scientific_falsifier": "A recount of the deployed sitemap (dist-netlify-public/sitemap.xml) or of the built /research/blog/ tree that returns a different loc count or blog-article count.", "scope_downgrade": "State only the deployed-sitemap loc count and drop the blog-article count until the build recount matches.", "same_underlying_claim": "PUBLISHING-SCALE", "independent_evidence": false }, { "claimId": "BOOK-ARTEFACTS", "claimType": "factual", "subject": "The two book artefacts and their lengths", "publicStatus": "published", "exactSentence": "The 30 April 2025 self-emailed manuscript runs to 562 pages. The book published in January 2026 runs to 468 pages.", "evidenceAxes": { "provenance": "P3", "scientific": "S0", "reproducibility": "R3", "originality": "O0" }, "requiredQualifiers": [ "The manuscript and the published book are different artefacts with different lengths. Neither number is a correction of the other." ], "prohibitedWording": [], "checkableFacts": [ { "quantity": 562, "of": "page manuscript" }, { "quantity": 468, "of": "page book" } ], "routes": [ "/book/", "/priority-claims.html" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-07", "notes": "A first pass rejected these as a contradiction (562 on 30 pages, 468 on 15) and suspected a regression against a completed correction task. the research record read the contexts and found two correctly-labelled artefacts. The rejection was right and the reason was wrong; unlabelled is the defect, not wrong.", "originality_defeat": null, "scientific_falsifier": "A page count of either artefact from its published file that differs from the stated number.", "scope_downgrade": "State only the page count of the artefact currently being described, rather than presenting both together.", "same_underlying_claim": "BOOK-ARTEFACTS", "independent_evidence": false }, { "claimId": "CEILING-DERIVATION-LAW", "claimType": "formal", "subject": "The ceiling of stable self-improvement", "publicStatus": "published", "exactSentence": "The ceiling of stable self-improvement is the reciprocal of the correction shortfall, one minus the correction exponent. Feed the law one assumption, that same-class correctors accumulate corrections the way independent samples do, so the exponent is one half, and it returns two.", "evidenceAxes": { "provenance": "P2", "scientific": "S1", "reproducibility": "R1", "originality": "O1" }, "requiredQualifiers": [ "Law first, value second: the reciprocal relation is load-bearing; the digit two is one entailment under one accumulation assumption.", "The asymmetry is deliberate. Kill the value two and the law is untouched: measure the correction exponent and report a different ceiling. Kill the law and the value two has nothing holding it up, because the book printed the square without deriving it and the derivation arrived eighteen months later through the reciprocal. Tests are aimed at the assumption, never at the digit.", "Same-class independence is an assumption, not a result: it is what a fair test must aim at.", "Not an empirical measurement on real correctors. Real-data collection is authorised only when RC2 gates and preregistration approval have both cleared.", "The earlier test 'one number does two jobs, measured two independent ways, and the answers must agree' has been withdrawn as vacuous, in public and before filing. If both exponents arise because accumulation is independence-limited, they share a law and agreement is forced; a test that cannot fail is not a test. The replacement is one accumulation law with two consequences (the conversion exponent and the correction ceiling), which asserts no independence, so the confound has nothing to bite on. The cross-class ratio already on the register carries the unification instead." ], "prohibitedWording": [ "one number does two jobs", "two independent ways", "the answers must agree", "the ceiling equals two", "the critical exponent is two", "derived from the book's R-squared", "measured on real correctors", "law of self-improvement", "runaway" ], "notes": "The book (Infinite Architects, January 2026) printed U = I x R squared as a fixed-exponent conjecture without derivation; the reciprocal-of-correction-exponent derivation is a later result that entails the digit under the independence assumption. Never quote the book's printed R-squared as evidence for the critical exponent. The falsification number that separates the framework from a no-framework null is the cross-class to same-class correction-exponent ratio: the null’s expected value for it is exactly 1.00 under the exchangeability model (architecture-blind averaging has no free parameter for composition class); the framework says materially above 1.00, and that ratio has not been measured on real data.", "routes": [ "/research/the-laws/", "/research/the-synthesis/", "/research/papers/paper-x-coupled-coscaling-correction.html" ], "reviewOwner": "the research record", "reviewedUtc": "2026-08-12", "originality_defeat": "A dated primary source that derives 'ceiling of stable self-improvement equals the reciprocal of the correction shortfall' (alpha_crit = 1/(1 - gamma)) under the same accumulation model before this record's date; or a dated primary source that publishes the same asymmetric ranking (law before value) for the same quantity.", "scientific_falsifier": "Kill-two leaves the law: if same-class independence is refuted on data, the digit two dies and the reciprocal relation stands, with a different ceiling reported from the measured exponent. Kill-law leaves nothing: a proof-checking error in the derivation of ceiling equals one over beta, or an explicit counter-example inside the model's assumptions that violates the reciprocal relation, invalidates both the law and the value at once.", "scope_downgrade": "State only the reciprocal relation for the measured exponent of one named corrector class, dropping the independence-derived value and the ranking claim, until the accumulation assumption is measured.", "same_underlying_claim": "CEILING-DERIVATION-LAW", "independent_evidence": false } ] }