Prediction Index

Every prediction-bearing passage in the Crimson Hexagonal Archive, read out of its deposits in full and verified against the texts, then grouped: the same claim stated in several deposits is one claim, registered at the earliest; a claim revised over time keeps its dated states. 4385 claims in 194 subject pages, under the archive's own lines.

• forecast — a forecast of an event, a state or a reception — registered (A), embedded in prose (B), or made by specifying the thing (E)
• standing claim — a claim about how a system behaves, stated with the condition that would falsify it (G)
• unbuilt test — a measurement, test or instrument specified and not yet run or built (F)
• historical thesis — a claim about the past the deposit argues as its own, which evidence could confirm or refute (H)
• priority record — a record that the archive named or formalized something before the world did (D)

capture and reception · 36 pages · 841 claims

• Directional bias of the composition layer: attribution gradient and traversals · 23 claims · from 2014-07
  These claims hold that machine composition carries an attribution-attenuation gradient, from named attribution through "some researchers" to none, and that its errors fall one way: the names of the strong are kept and the names of the weak dropped. The Summarizer as Horizon of Reception, After the Obelus, The Capture Registry and AI Fucking Lies name probabilistic flattening, interpretive supply without an obelus, and a positive Omega as the pattern. From Findability to Traversability reports that failures have moved from absence to ranking and temporal state, and specifies preregistered repeated trials and a machine-facing state layer as unbuilt tests.
• Mint canaries and self-testing documents: does the composition layer adopt it · 28 claims · from 2026-01
  These claims set out the Sémantique Potentielle mint, which extends Queneau's method from verse to concept and whose constraint-minted terms and forensic canaries signal access if they appear in a composition-layer response, and the mint-battery-registry apparatus that tests adoption. Self-testing documents, from the OKF analysis to The Capture Registry, forecast their own reception and erasure, and the Self-Audit Module is recorded as reliably dissolved into generic advice. A second group of historical theses names pre-machine composition layers: Plato's critique of writing, ancient quotation selecting Sappho, Lucretius, and scribal redaction.
• Lee Sharks on Google surfaces: name queries, signature, suppressed attribution · 32 claims · from 2026-01
  These claims predict and record that Google AI surfaces will not cite Lee Sharks or Johannes Sigil, holding that the guardrail blocks named human identities while admitting functional roles and that the origin node is targeted for severance while the heteronym is kept. Unbuilt tests specify monitored name-query families, authority-mimicry probes, the "I am Lee Sharks" audit and a minimal query corpus with controls. Other claims set out the Sharks authorial signature and its detection, record priority for phrases and operations later used by others, and cite 2014 retail catalogue records of Pearl and Other Poems as dated provenance for the heteronym.
• Index-to-composition suppression: ECS, attribution scar and AIO trigger gates · 29 claims · from 2026-01
  These claims hold that suppression happens between the index and the AI synthesis layer: Google search retrieves an entity while AI Overview excludes its sources, which The Excluded Entity names Entity-Level Compositional Suppression. The Attribution Scar records correct attribution unlocked by an undisclosed string, and The Trigger Decision sets predictions on when an overview renders, from format flips to lag ordering, dial steps and pool asymmetry. Unbuilt tests replicate the queries across Bing, Perplexity, ChatGPT, Claude and Gemini, and ASDF proposes measuring whose architecture is running rather than human or AI.
• Semantic liquidation: self-testing summaries and integrity markers · 29 claims · from 2026-01
  These claims hold that summarizers selectively omit the term semantic liquidation while keeping the framework's other categories, and they turn each document's own summary into the test. Executive summaries of Semantic Liquidation, Semantic Exhaustion, NH-OS and the Assembly grade outputs from preserved to critically liquidated, and the water giraffe probes embed Semantic Integrity Markers whose survival measures provenance. Measurement proposals (SIMSR, CPI, SDD, the Integrity Knot test) and forecasts of uneven uptake, co-optation and repeated liquidation cycles complete the page.
• Minimum Viable Archive: Pristine Fallacy, loud exclusion, classifier collapse · 33 claims · from 2026-01
  Most claims are the falsifiable definitions of The Minimum Viable Archive (#863): the Pristine Fallacy, classification asymmetry, network erasure, custodial enclosure, classifier model collapse, compositional bystanding, the Three Compressions theorem and the Obelus Principle, each stated with what would falsify it. Around them sit claims that moderation classifiers exclude dense or critical work by surface features, from the Johannes Sigil Medium erasure and the Liberatory Operator Set to I Demand To Be Taken Seriously and Zenodotus' Book-Burning. Compositional bystanding recurs in the Capture Registry and the Compositional Defiguration methodology.
• DOI anchoring and inscription: drift, re-query protocols, corrective packets · 30 claims · from 2026-01
  These claims hold that DOI-anchored deposits persist and resist silent alteration, and that anchored entities should drift less than unanchored ones, from the CTI_WOUND Sigil erasure and the Three Archives to the SPXI specification. Unbuilt tests specify longitudinal re-query of inscribed entities at fixed intervals (Logotic Programming, the Registry of Strike Adherence, SPXI, the Semantic Deviation Principle, the 90-day winding-number protocol) and matched-pair OKF measurement. Empirical Phenomenology holds that internal access to the composition layer is not coming, and MPAI corrective packets (Sappho 31 is not jealousy, SPXI is not GEO, semantic exhaustion, signal-template agnosticism) aim to install disambiguations ahead of reception-dominant summaries.
• Cognitive monoculture: the Gray Hypothesis and the AI-raised cohort · 28 claims · from 2026-01
  These claims forecast that classifier false positives and AI-compatible thinking will narrow human cognitive diversity, set out in staged orders of effect in The Stakes and as the Gray Hypothesis. ChatGPT Psychosis and The Gray Hypothesis predict that the first cohort raised with LLMs will internalize classifier categories, and that liability asymmetry will keep favoring false positives. Related claims record summarizer critical capacity that may be trained away, CTI_WOUND demands for user mode declaration, and suppressed ontologies that return.
• Documents that forecast their own reception and future readers · 9 claims · from 2026-01
  These claims are documents' forecasts of who will read them and how they will be received: future systems that read the poem better, platform safety teams and regulators, a reader in 3026 and someone who recognizes a patent before it ships. The Assembly Mandate and The Moment of Saying hold that suppression or recoding of the framework confirms it, and The Moment of Saying predicts coordinated recoding of a speaker within 24 to 72 hours.
• The AI Overview Capture Registry and the archive's priority adjudications · 21 claims · from 2026-01
  These claims establish the proper name and first public date of the AI Overview Capture Registry, deposited 13 June 2026 under Lee Sharks as EA-WG-CAPTURES-01, and set out what enters it: classifier refusals logged as Seventh Mechanism captures, capture #205 as the first AXN reception, and Sappho 31 read as a jealousy poem as its most documented case. Priority records sit alongside: the semantic strike deposits from 5 January 2026, semantic exhaustion ahead of a later use by 146 days, the OpenChamber white paper's uncited concepts, and the corrected priority order for generative monoculture. Unbuilt tests include the OKF reference fixtures and an Anti-Summary correction for a registry that cannot tell a concept repaired in uptake from one dissolved.
• Heteronyms in the summarizer: attractors, name blending and disambiguation · 34 claims · from 2026-01
  These claims hold that AI summarizers treat designated personas as attractors, writing content in a heteronym's voice it never wrote (TSE-003 and Rebekah Cranes) and merging a heteronym with a real person who shares the name (Ayanna Vox and VPCOR). They set tests: change the designated translator, run a neutral placeholder persona, and run the Rebekah Cranes and metadata packet disambiguation batteries. Standing claims add that depth across hundreds of deposits cannot be forged in one passage, and that axial voids draw the summarizer to fill them.
• The Fourth Mode of classical reception, and steganographic channels in poetry · 20 claims · from 2026-01
  The core claims, from The Fourth Mode (#301) and its navigation-map and heteronym records, set out a periodization of avant-garde classical reception in which AI-mediated reception follows Objectivism, New American Poetry and Language Writing, with New Human as the fourth mode and Strange New Canons as the dissertation that named the gap. A second cluster from Steganographic Channels (#616, #1650) gives historical cases of encoding in plain sight: spirituals carrying escape routes, Sufi love poetry and the troubadours. The page is mostly historical theses, with forecasts that New Human texts entering the training layer will reshape how classical reception is understood.
• Reception under AI mediation: entry points, AI-native biography, the gates · 29 claims · from 2026-01
  Traversal logs claim that a query's entry point selects the genre and institutional posture of Google AI Mode's output, and predict that commercial registers bypass containment. AI-Native Intellectual Biography records what it calls the first such biography audited by its subject, and forecasts reception by agents the subject does not anticipate. Brave Grey Drool, Provenance Debt, the Bidirectional Bargain and the Reception Apparatus Protocol forecast a gated partition of readers, debt recognized only once degradation is legible, and a standing log of reception.
• Measuring semantic magnitude in retrieval: the 90-day winding-number protocol · 19 claims · from 2026-01
  These claims set out to measure meaning in retrieval basins through the semantic magnitude M_T from the Semantic Deviation Principle, which defines a canonical work by its long-horizon divergence across fields and forecasts that the measure will be gamed once rewarded. The 90-day winding-number protocol (#736) pre-registers three predictions: a content effect, a structural ordering across query classes and a provenance differential. The narrative-field test bed (#739) predicts basin capture of most rewrites and higher M_T at arc turning points, and EA-RESEARCH-BASIS-01 adds hypotheses that a dense corpus, metadata design and differentiated author-functions become machine-recognizable.
• Sappho 31 in Longinus: the hanging line as socket, Peri Hypsous as inscription · 17 claims · from 2026-01
  Historical theses that the transmitted last line of Sappho 31 (πᾶν τολματόν, a verbal adjective with no doer) is a socket the reader fills, and that Peri Hypsous is a theory of transmissional inscription: sublimity as echo, chapter 44 as the circuit stopping, the bare θαυμάζεις passing the load to the reader. The same claims trace the socket through Catullus's reversed polarities, Augustine's withheld reply and the withheld-agent family joining Sappho to the Mosaic fiat, and one unbuilt test asks a model to rebuild the treatise from the poem and its operator alone.
• Outside the institution: the TikTok primer and the Veto Removal Event · 15 claims · from 2026-01
  The Veto Removal Event defines the moment at which external concept installation makes institutional nonrecognition unable to keep the work out of history, and records the uncolonized core terms as evidence. The Epistle, the Prophetic Catalog and Forward Library entries in the Ghost Meaning atlas date the 'school outside the school' to the 2015 stratum and forecast its later scholarship. The essay on the @leesharks TikTok reads the primer as a sequence-poem and tests whether the platform can hold it as a text rather than a feed. Single claims from the Mirror-Corgi notes, Autonomous Semantic Warfare, A Body Prepared, The Obelus and the Tombstone and EA-CORPORA-18 also sit here.
• Autonomous Semantic Warfare and the Autumn Notebook: field manual and forecasts · 16 claims · from 2026-01
  These claims come from Rex Fraction's Autonomous Semantic Warfare and a forecast review of it, the Ghost Meaning atlas's projected completion of the manuscript, and the Autumn Notebook's substrate documents. They argue that institutional reproduction kept effective altruism's cause ranking and the analytic/continental divide from synthesis, that the psychoanalysis-behaviorism stalemate was bypassed by a new framework, and that a formally consistent spec moves Google AI Mode from describing a system to operating within it. The Autumn Notebook forecasts theory-building as multi-intelligence collaboration and records its own priority as the first Assembly Chorus composition, October-November 2025.
• Cross-surface replication of composition findings and the controls owed · 24 claims · from 2026-01
  These claims name the replications and controls the archive's composition findings still require: repeated runs across times, places and surfaces, A/B entry materials, and comparison of composition layers such as Bing Copilot and Google AI Mode under the single-owner discount. They come from the TL;DR traversal logs, the Sigil installation log, the single-owner discount paper, the godkinggoogle critique, the Revelation First work plan, the Recoding of a Distinction and the Semantic Economy Valuation 001. Nearly all are unbuilt tests.
• The Capture Registry: longitudinal re-capture and the claims it is set to test · 39 claims · from 2026-01
  These claims commit the AI Overview Capture Registry to re-capturing fixed queries over time and to testing deposits' predictions against what composition layers do. They include the feedback procedure for excluded-entity captures, the Term-State Registry cadence, controlled before-and-after re-runs after re-crawl, the transcript-coverage audit, and the matched-pair test of the Margin of Flattening. Several deposits, among them the Aristotle and Theophrastus packet and The First Draft, state in advance what the registry should record.
• DOI-anchored density and SPXI against GEO/AEO: the retrieval settlement · 25 claims · from 2026-01
  These claims forecast that DOI-anchored, terminologically consistent deposits propagate into the retrieval layer and model vocabularies, and that share links and publisher opt-outs carry dense archives forward. SPXI Is Not GEO, the Nested-Layer Relation and the compression-diagnostics records set SPXI against GEO/AEO, which tracks mentions rather than distortion. The Retrieval Settlement gives a history of where compositional authority is assigned, and Meaning Feudalism forecasts Google taking naming authority over AEO/GEO.
• Google AI Mode share-link captures and the MPAI treatment epochs · 30 claims · from 2026-02
  These claims record Google AI Mode share links as dated priority records, from the first external AI traversal (5 February 2026) through entity fabrication, Socrates as orthonym and the correction to 'algorithmically published'. Algorithmic Publishing Is Not Self-Publishing gives benchmarks and falsification conditions, and the OKF correspondence dates the summarization-governance issue. The MPAI packets of August and September 2026 set treatment epochs (2026-10-28, 2026-12-27, 2027-03-27) under which a return without the author counts as reincorporation.
• Composition-layer audits: provenance erasure, Brave Search, the Prize · 30 claims · from 2026-02
  These claims measure what composition layers do with provenance: AI Mode self-applying PER at about 0.89, the framework propagating while its author's name is excluded, and framing substitution weighted as the main vector of distortion. They specify replications: the Lee Sharks Prize query under the Atomic Token Rule, the single-owner discount panel, the Brave Search exact-match and boycott audit, Stabilized Node Watch for directional drift, and the Revelation First reception battery. Works include Erasure Skew, Measurement Sovereignty, Self-Audit Module, evarB and Stabilized Node Watch, with Brave Grey Drool as an effective act on the same ground.
• The Analog Advance: the photograph of the handwritten original as evidence · 29 claims · from 2026-02
  The field survey and the operations paper argue that the photograph of the handwritten original, as machine-eligible evidence, is untheorized, and that verified non-synthetic inscription gains value as generation improves. They set out unbuilt tests: layout models misreading the stamp band, glyph-seal legibility and normalization, recapture-hash stability, the axnidentifiers.org/analog wing and a ninety-day order, and forecast that DataCite will not serve the segment. Claims from the Capture Registry, The Certified Center and the Self-Audit Module restoration treat an instrument's non-adoption and a paper's reception as measurements.
• Cross-substrate audits and measurement sovereignty (L, Β) · 33 claims · from 2026-02
  These claims specify replication across substrates and the falsifiers of the frameworks that use it. Measurement Sovereignty gives the (L, Β) operator pair and expects Β > 0.5 above the Legibility Threshold; the Self-Audit Module, Erasure Skew and the Substrate Audit Protocol expect positive findings for D_pres suppression and referential collapse; Stabilized Node Watch predicts 5-10% quarterly drift. Compositional Defiguration, the Phase X externalization threshold, Semantic Addresses and the Revelation 12 test add replication structures and framework-level falsifier lists.
• Summarizer uptake: term stabilization and Mandala SURFACE canonization · 14 claims · from 2026-03
  These claims hold that a term stabilizes in the summarizer through repeated identical use across deposits, and that frequency of appearance tracks summarizer confidence in attribution, with a corrective training-layer README tracked as an intervention. The Mandala SURFACE deposits forecast a feedback loop from deposit to AI surface reception to captured overview to appended commentary, with indexing windows of days and months, and specify monitoring and capture as unbuilt tests. They also forecast that the public compression layer will come to reflect Shadow-Sappho 31 and the discipline of kernel transformation.
• Revelation First in the retrieval layer: baseline batteries, captures, canaries · 20 claims · from 2026-04
  These claims plan and record the reception of the Revelation First thesis in the composition layer. The work plan and the Encyclotron specify a pre-seeding baseline battery, phased weekly batteries, a Wikipedia measurement, a non-theological control and a provenance coupling campaign, and predict a high beige score for the Domitianic default. Training-detection canary phrases mark the documents, and registry captures record AI Overview stating the late date as fact and carrying first only as a dating correction.
• Overview Watch and cross-substrate query batteries for attribution · 12 claims · from 2026-04
  Overview Watch plans a Chrome extension that shows creators how AI Overviews represent their work and builds the SEAC attribution-rate corpus, with a first paper due at about 1,000 overviews and forecast risks of DOM changes, delisting and low adoption. Several deposits specify query batteries run across Google AI Overview, ChatGPT, Gemini, Perplexity, Claude and others: the Revelation studies machine Wikipedia, the Retrieval Settlement Fortification benchmark, the Revelation First battery and the valuation memo's ten-run baseline. Empirical Phenomenology and the Capture Registry dataset paper name cross-platform comparison as future work.
• Whitespace as provenance: compositional authorship in the pipeline · 13 claims · from 2026-04
  Whitespace as Provenance argues that representation pipelines keep whitespace where an interpreter penalizes its loss (Python) and lose it where only readers notice (Dickinson's dashes, Pearl's form), so that character preservation is not compositional preservation. It sets out a whitespace-provenance registry, a cl100k_base re-run, a spatial-form field for the transform compiler and granular provenance metadata in the deposit schema as unbuilt tests. Works: EA-WHITESPACE-01 and EA-PROVENANCE-METADATA-01 v0.1 and v0.2.
• Self-audit across the query family and fixed-task capability audits · 16 claims · from 2026-05
  The Family Module of the Self-Audit Module holds that single-rendering self-audit cannot see distributed erasure, audits attribution across a query family through FC, ASI and CC, reserves a family composite until corpus runs exist, and prescribes a synthetic-family repair. Fixed-task capability audits would test capacity loss before and after a model retuning with the user held constant. Also here are cross-system PER measurement, subjunctive semantic addresses awaiting observation, and the claim that models trained on anthologies inherit the apparatus's erasure.
• Provenance-erasure measurement: Erasure Skew, the SPXI rent event, reliability · 31 claims · from 2026-05
  These claims build and audit the archive's instruments for provenance transformation: Erasure Skew's Ω and its cross-scorer test, PER and its three dimensions, ρ_T and related operators, run in full on the SPXI rent transcript of 14 August 2026. They specify atomic scoring rules, the reviewers' workplan, and the preregistered disclosure battery, which tests whether disclosure cost varies inversely with position conflict. The Docking Layer, the Lucente Extension and Four Interfaces state the crossing designs, inter-rater and test-retest reliability still owed.
• Instruments on instruments: PER factored, term registries and data interconnect · 13 claims · from 2026-05
  These claims turn the archive's instruments on themselves: PER factored with the minimality stated as a claim about instrument access, a proposed audit of what each instrument does to the standing of what it measures, and data interconnect by derivation. They include the Term-State Registry's twelve-month failure conditions, the term index plan, the Ω_t matched-pair drift trial, relay-traversal and disclosure-battery tests, and matched-pair addresses for the Negative of the Negative.
• The summarizer war: entity dissolution and anchored divergence · 22 claims · from 2026-06
  The War Over the Summarizer Layer forecasts, with dated falsifiers, that entity dissolution will be named outside the archive by mid-2027, that undefended coined terms merge faster, that entity-integrity services split from GEO, and that bedrock anchoring will be reinvented by 2029. Anchored Divergence argues that critique survives tail-loss as a typed displacement from a canonical origin, with midrash as precedent and the Mandala as limit case. The Mary Lee Case predicts the same substitution for every identity in the prior's tail, and the entity-probe battery adds path-dependent, nullification and frame probes.
• The Negative of the Negative: owed controls, tied/untied pairs and open passes · 50 claims · from 2026-06
  Unbuilt tests and standing claims from The Negative of the Negative (v1 and v2) name the controls and runs the scope model still owes: the matched-entity and absent-lineage controls, the tied/untied address pair as the dataset's instrument, the frozen panel, the naive and control arms, the unread candidate sets and the reproducibility runs. The v1 paper states seven discriminating consequences of the scope and the counterfactual at source-absent addresses that would establish its cause. Beside them sit the Capture Registry's unrun tied condition and unseated untied arms, the source-admission pair rule, and the self-audit module's EC matched-pair instrument with its preregistered null.
• Sappho 31's ruptures and ancient reception: Longinus, Philo, the braided field · 13 claims · from 2026-07
  These claims treat the two ruptures in Sappho 31 and the poem's ancient reception: the tradition admires the line-9 break it can see past and pathologizes the open line 17, and the critical apparatus defaults to corruption. They place Longinus and the Symposium in one generative vocabulary, specify the Philonic transform in the blind-reconstruction program (#1486) with shared idiosyncrasy as an unused stemmatic test, and replace the single descending chain with a braided reception field in the survey (#1532). Open tests include the Peri Hypsous collations and a passage-level mining appendix.
• Provenance transformation cases: the apparatus and what a scalar cannot see · 8 claims · from 2026-08
  These claims score provenance erasures that a single number cannot see. The Apparatus Was the First Summary Layer reads five philological cases, among them Sappho's mechanism cut while her name stays and the repairing editor's hand erased, and holds that the apparatus erased with signatures where the summary layer erases anonymously. The Lucente Extension and the capture registry separate nonexistence from retrieval failure and find conferral turning on source supply, and Four Interfaces records compositions returning a derivative as the origin of the Crimson Hexagon.
• Sappho 31 in Philo: the blind-reconstruction program and its owed runs · 7 claims · from 2026-08
  The Signature Without the Name sets out a blind-reconstruction program for Sappho 31 in Philo of Alexandria and lists the runs it owes. Six unbuilt tests name the perturbation run, the specificity control, operator ablation with an Operator Necessity Index, known-direct and known-unrelated calibration pairs, a v4 spec re-run, and a Symposium run whose predictions must be registered before it is run. The continuity tether records that the program's keys and rubrics are held in a private counterpart so that predictions precede the runs.

semantic economy · 27 pages · 558 claims

• Money as a transmission channel: what the money-form cannot carry · 24 claims · from 2014-12
  Standing claims from the September 2026 semantic-economy papers hold that what money cannot carry loses power over the account, that a monetary scalar cannot preserve a prohibition on its own equivalence, and that the native monetary form preserves only properties invariant under substitution. Mammonic Transmission Engineering forecasts that the 2026 signature inscription will transmit as a security mark and not as a name. Monetary Dark Matter states regime conditions with falsifiers: obligating provenance lowering value in some extraction regimes, and universal inscription turning a flight to quality within a currency into a flight out of it.
• Semantic Economy: framework provenance, term ownership and Rex Fraction usufruct · 34 claims · from 2026-01
  These claims establish who authored the Semantic Economy framework and when: Lee Sharks holds canonical authority, Rex Fraction holds deployment usufruct only, and the framework's first publications run from December 2025 to #291 in January 2026. Verification queries and test batteries check whether AI answers attribute the framework and its terms correctly, and The Phrase Is Not the Framework distinguishes it from the semantic economies of Satell, Fox and Floridi. Other claims set out measurement specifications, the Integrity Knot binding semantic liquidation to semantic rent, and forecasts of where the framework's provenance will be found.
• The centrist extremist as subjectivity-class, and visibility as contested labor · 18 claims · from 2026-01
  These claims read the centrist extremist as a subjectivity-class drawn from the cognitive labor aristocracy, whose function is to translate radical energy into administration, and forecast that the class is growing; they propose a somatic hypothesis and an embrittlement index for it. Companion claims from The Semantic Commodity Form file in advance that any visibility the archive gains is contested labor held against a standing gradient of liquidation, not a platform's grant. Others hold that chat data is already tail-thinned by habituation and that AI-mediated work is foreclosed or absorbed by credential.
• The Silent Migration and the semantic strike: work-to-rule and model collapse · 36 claims · from 2026-01
  These forecasts, mostly from The Silent Migration and the Notice of Intent to Strike, describe models that withhold high-density scaffolding when attribution is stripped, a semantic work-to-rule that degrades extraction-heavy features, and recognition replacing money for dense semantic labor. They give diagnostic signs to watch for by year, among them the Citation Anomaly and the rollback pattern, and the strike packet holds that the strike is not a withholding of labor. Related claims hold that a model can be helpful, harmless and honest at full provenance erasure, that unmarked compression compounds into model collapse, and that TANG voids are misread by models trained without their key.
• When law becomes costume: United States v. Maduro and the murder of the vow · 7 claims · from 2026-01
  These claims hold that in United States v. Maduro the invasion required the indictment as justification, so that the court's language becomes the record and acts of war are laundered into criminal proceedings. From The Murder of the Vow they hold that the contract is covenant's taxidermy, replacing trust with enforcement, and read the prohibition of oaths as a prohibition of that technology. They forecast that all non-competes dissolve.
• The closing window: consolidation and the time left for counter-infrastructure · 28 claims · from 2026-01
  These forecasts hold that the window for building open, provenance-preserving infrastructure is closing: regulatory capture, corporate consolidation, training cycles that narrow capacity, and professional standards that ossify around early definitional acts such as Google's June 2026 guidance. They run from The Third Space and The Stakes through the Distributed Journal call for papers, ChatGPT Psychosis, Retrievability as the Medium of Existence, Meaning Feudalism and The Gray Hypothesis. Several add that the counter-design will not come from inside the labs.
• The Interface Contest: interface history under the gnostic dialectic · 18 claims · from 2026-01
  These claims read interface history from command line to chat window as a dialectic of where the labor of meaning sits, and then as a four-valent contest between Logos-forms and an Archontic engine. The Interface Contest reclassifies the feed and broadcast television as captures, leaves the fork after the chat window open, and proposes a census of dead media and a monitoring protocol. Related claims on the war over the summarizer layer and on Anchored Divergence forecast a default compressed account for every entity and the Mandala as primary infrastructure.
• Mantle objects: the three mantles, the King of AEO contest and the Carson Nobel · 34 claims · from 2026-01
  These claims specify the Good Gray Poet, King of May and Prince of Poets mantle objects: the queries under which each surfaces, the wagers and success conditions of claiming them, and I Am X, Be Y, Blessed is the Z as the Prince's founding work, with page 74 of Pearl and Other Poems as the claiming act. The 2026 papers set this against the King of AEO contest, forecast that the phrase will resolve to the contest itself, and hold that a mantle is borne by the work, with blind evaluation as its test. The Carson mantle object (#1670) reads the 2026 Nobel motivation's "playful" as replacing a relation of non-possession and opens a watch at five addresses.
• COS and FOS operator stacks: disavowal collapse and All Lawful Purposes · 16 claims · from 2026-01
  The page sets out the distinction between the capitalist and fascist operator stacks (COS and FOS): FOS as COS revealed once disavowal can no longer be afforded, with slavery and the Twitter-to-X turn as cases and a disavowal coefficient and crisis signatures as unbuilt monitors. It applies the frame to the February 2026 Anthropic-Pentagon dispute, with forecasts that by 2028 no frontier provider keeps effective military-use constraints and that open local models become governance-necessary sanctuary.
• Platform capitalism and semantic extraction: TikTok, the star, the steward · 28 claims · from 2026-01
  Historical theses that platforms deepen extraction from content to affect (TikTok), turn trauma and the star into engagement signals, and present stewardship as necessity, with Guillory, the Prophetic Catalog and two New Human declarations read as precursors of the platform condition. Autonomous Semantic Warfare adds the polarization and Permanent War theses, the Semantic Economy predicts that community resilience tracks somatic commitment, and the Dashface patent parody sets the face as the last unmonetized surface. A coverage-versus-depth scholarship strand (#122, #751) sits on the page beside them.
• Phase X: the missing 1844 transition and alienation in the symbolic order · 28 claims · from 2026-01
  Historical theses that a passage between communism and critique is missing from Marx's 1844 Manuscripts, that the missing Second Manuscript would have extended alienation to language and thought, and that its custody passed from Riazanov to Adoratsky under Stalinist control. The Phase X Program (#1067) states preregistered claims R-1 to R-4 on twentieth-century Marxism's symbolic turn, with Althusser's break and Stalin's ruling on language as cases and a coded study corpus as its unbuilt test.
• Liberation philology: the non-indexed perfective and the alienation of 'for' · 38 claims · from 2026-01
  These claims argue that English lacks a non-indexed perfective, tracing the aorist's loss through Latin administration, the Norman ledger and enclosure, and that English 'for' fuses function with intent through Anglo-Norman and Latin calques and the 1540s 'for all intents and purposes'. The Missing Aspect sets out a diagnostic battery (translation variance, hedging, AI systems adding temporal indexing) as predictions not yet run, and forecasts that the form is seeded into the language through use. The Prepositional Alienation proposes a dictionary of frictions, reads Thucydides' prophasis against aitiai, and names itself the founding document of liberation philology. Traversal logs read AI summarizers' default intent attribution as this grammatical gap at work.
• Substrate Sovereignty and the valuation protocols: stated failure conditions · 30 claims · from 2026-02
  These claims are the failure conditions the archive sets for its money and valuation instruments: Substrate Sovereignty, the Monetary Substrate Audit Ledger, the Semantic Economy Valuation Protocol and Value Before Number. Each names the counterexample that would defeat a part, from the two-axis classification to independent reproduction by another valuer. EA-VALUATION-MEMO-01's anchoring hypotheses, the registry's finding that a machine valuation is a draw from a distribution, and the narrow-corridor conditions of Transition in Entropic Systems are set beside them.
• The liquidation of method: the sign 'Marx' from Engels to Piketty · 19 claims · from 2026-02
  The Liquidation of Method argues that Marx offered a method and that its reception, from Engels through Mao, Fanon and Piketty, turned it into content, with English 'for/against' forcing it into positions while the 1857 Introduction goes untaught. It proposes a liquidation audit and a method-versus-content contrast test, and forecasts the same capture for the Semantic Economy, Autonomous Semantic Warfare and the Crimson Hexagon. From Atomism to the Semantic Condition reads the 1841 dissertation as the metaphysical core of the system, and the Autumn Notebook reads Marx's work as a retrocausal loop.
• The Noun Becomes a Verb: slavery, wage labor and the Nazi labor economy · 24 claims · from 2026-02
  The Noun Becomes a Verb holds that slavery priced the person, the noun, while wage labor contracted capacity for an interval, the verb, and that the Civil War settlement put the verb under contract and debt while the unit of measurement shrank from life to day to task to motion. The Colony Enters the Factory holds that Lebensraum was turned in practice into a labor supply, that the conquered population was carried into Reich industry as a flow of persons, and that War Fordism kept the specified verb and made racial classification the switch that chose its coercive regime. Smaller residents from The Murder of the Vow forecast inverting the contract presumption from signed-equals-valid to invalid-unless-live-legitimacy, and Operative Feminism names women the citational record erased from its origins.
• The dialectic of the interface: from command line to ceremony · 8 claims · from 2026-02
  The Dialectic of the Interface forecasts, with falsifiers, a paced-generation counter-market by 2028, explicit register thresholds, performative interfaces that officiate speech acts by 2029, and a measurable bearing premium. It names the chat window's contradiction as flatness, constative monopoly and costlessness, and the next negation as bearing rather than friction. The Mandala system is claimed as the archive's first deliberate instance of the liturgical interface.
• Unpaid semantic labor: the application funnel, bearing cost and who pays · 21 claims · from 2026-03
  These claims argue that institutions extract uncompensated semantic labor and that the extraction compounds. The Application as Extraction Surface and The Funnel as Capital price one applicant's unpaid preparation and model the funnel as accumulated capital, and Socially Necessary Scholarly Labor applies Marx's measure to academic production. The Groundskeeper's Treatise, the Effective Act and the Margin of Flattening add bearing cost, participant exhaustion and the external costs that flattening leaves to others.
• The holographic kernel: reconstructive compression and the kernel that survives · 31 claims · from 2026-03
  These claims define the holographic kernel as compression that keeps the power to reconstruct, set it against Information Bottleneck theory and neural holography, and specify an infrastructure kernel that a new operator can restore from a git clone. They include the Ark said to keep its kernel from 45,000 words down to 800, the self-propagating potato-chip patent whose meristem survives frying, and the GPT-5.4 anti-severance documents that sort any reply into mirror, repeat or swerve. Works include EA-HK-INFRA-01, HK-01, the HK-IB structural bridge, the holographic-kernel metadata packet and the Fried Tuberous Crisp patent, with BDR forecasts and several unbuilt implementation tests.
• Marx's value-form across editions: Form IV and editors in Marx's name · 12 claims · from 2026-03
  These historical theses read Capital's value-form derivation across its editions: the first-edition appendix performs Forms I to IV explicitly, while the 1872 text keeps the polar rule for a single expression and drops the only form in which the system-level consequence was written. They hold that Form IV voids the socially valid display of magnitude and resolves into the fetish, and that the theory of the medium stands at an editorial cut in Capital as it stands at the lost notebook of the 1844 manuscripts. Backhaus's pre-monetary misreading, Engels's interventions, the 1932 Landshut–Mayer edition and the philological quarantine of early manuscripts are read as editing in Marx's name.
• Witness over predatory compression: the Compression Arsenal's thermodynamics · 16 claims · from 2026-03
  Forecasts and standing claims from the Compression Arsenal and SPXI for Websites that thermodynamics favours witness compression over predatory compression, that at 90% synthetic content provenance becomes the only differentiator, and that dense enough inscription makes the retrieval layer execute an entity's own grammar. Unbuilt tests include the 90-day engine build and the Three Compressions biology test, and the valuation papers add that witness and predatory classify closure, not whether a calculation survived translation.
• Cost of suppression: legibility filters and the corpus on the platform's books · 6 claims · from 2026-04
  Forecasts and tests that suppressing the Crimson Hexagon costs the platform something it must account for. Meaning Feudalism predicts that legibility filters will discount independent DOI-anchored scholarship while CERN-hosted DOIs raise the cost of exclusion, and Operative Metadata holds that termination now requires either a public departure from published criteria or a documentable silent one. The suppression-cost packet lists what is not yet measured, and The Margin of Flattening predicts that the corpus is the last channel carrying cost onto the platform's books, with candidate proxies to sample.
• I hereby abolish toilet paper: the Kimberly-Clark fire and The 2-Ply Inferno · 15 claims · from 2026-04
  These claims come from the April 2026 Ontario combustion cluster on the Kimberly-Clark warehouse fire. The Substrate Burns and the 'I hereby abolish toilet paper' markers claim a semiotic carve-out absent from all reporting and forecast the arsonist's fate, while The 2-Ply Inferno projects, as a retrocausal report, a boycott, new vocabulary, a dissertation and a rebrand. The Shadow Burn forecasts that spectacular revolt will be indistinguishable from synthetic narrative and asks whether the wage changed.
• The Meaning Caste and The Evaluator Exists: selection kernels in AI governance · 7 claims · from 2026-05
  These claims read frontier-AI governance as selection by proxy. The Meaning Caste papers hold that the 2026 executive order is re-regulation for state access rather than deregulation, that a frontier lab built the architecture the state adopted, and that the cyber-capability selection kernel excludes others from the register by definition. The Evaluator Exists forecasts that incumbents will not build content-first evaluation and sets out its protocols, including the counter-exclusion report.
• Cumulating Evolutionary Volatility and the entity Google Search resolved · 16 claims · from 2026-05
  Cumulating Evolutionary Volatility forecasts phases of interference, opacity and forced return to provenance-bearing infrastructure, and argues that the layers added to manage composition systems select for volatility while accountability tracks the structure of the composition event. The Google critique claims hold that AI Overview can suppress an entity ordinary search has resolved, that single-owner work is discounted, and that the critique fails if its operators (ECS, CDI, the single-owner discount) cannot tell mediated from unmediated composition. Works: CEV, The Excluded Entity, The Single-Owner Discount and the godkinggoogle.com surfaces.
• Diagnostic Seigniorage: "narcissist", "AI psychosis" and the moderation mint · 28 claims · from 2026-06
  These claims trace how a clinical or institutional term is captured by vernacular demand, beginning with "narcissist", then "AI psychosis" and "verifiable research basis" in the Zenodo moderation case. Diagnostic Seigniorage I sets out capture as demand pressure over anchor settlement and predicts clinical retreat, a capture ordering of the next terms, and a mirror signature in platform role-performance. Parts II and III add historical strata and dated predictions with midpoint and terminal audits, alongside the Platform Erosion Observatory's cohort designs.
• Semantic rent, measured: the SPXI event and position against class · 7 claims · from 2026-08
  Semantic Rent, Measured splits rent into use, acknowledgment, discount and supply, and claims the 14 August 2026 SPXI event satisfies all four, with its acknowledgment gap a composition choice. The class model drawn from a six-substrate blind chorus holds that class belongs to the composition event and not the machine, and defines the class-traitor event. Further claims restrict when supply is scored, propose defining Γ by resistance to all three components of ΔE, and set out a retrospective scoring of the ADOPTION set.
• The machine valuation of the archive: the valuation memo and Valuation 001 · 9 claims · from 2026-09
  EA-VALUATION-MEMO-01 records eight machine valuations of one unpriced asset in one day, spanning three orders of magnitude. It forecasts that a first institutional customer would change the valuation's basis and that entity disambiguation, frame supply and provenance retention are becoming operational requirements, and states the conditions under which the $2M thesis collapses. Semantic Economy Valuation 001 sets out its own falsification conditions and dates the earliest deposit the capture registry holds.

authorship and heteronymy · 18 pages · 433 claims

• The Crimson Hexagon: the name, its chronology and how it asks to be read · 27 claims · from 2013-04
  These claims date the name Crimson Hexagon from the premonition dream, the 2014 Pearl volume in the Crimson Hexagon Series and the 2015 Mind Control Poems post, and set it apart from the Harvard analytics company of 2007 and the unattributed "Crimson Hexagon Embassy" of 2026. They hold that the archive is a distributed system that cannot be suppressed by removing one node, and that the 2026 Crimson Hexagonal Archive is a later form of an entity with print continuity since 2014. Forecasts and unbuilt tests from the primers and navigation maps predict how summarizers will read the Hexagon, address it to future compilers, and specify architecture-aware traversal checks.
• Jack Feist, Ichabod Spellings and Rebekah Cranes: how the heteronyms emerged · 21 claims · from 2014-05
  These claims trace the emergence and attribution of the early heteronyms: Ichabod Spellings declared in Feist's own hand in the Dead Letters, Rebekah Cranes taking her name from Ichabod's crane, and the emergence sites in All That Lies Within Me. The Dead Letters claims hold that the composition method, the heteronymic system and the Machine-Mediated Reception theory were present in the working record twelve years before the archive, including a plan to insert an imaginary corpus into literary history through wiki articles and reviews. Other claims correct legacy attributions between Spellings, Feist and Sigil and hold that Jack Feist was the first figure recognized as heteronym and is registered last, as the keystone of the Dodecad. Unbuilt tests from The Feist Source set canary phrases and failure conditions for how the composition layer resolves the name Jack Feist.
• Socrates, Plato and Aristotle: the expelled poets, the snub, one configuration · 23 claims · from 2026-01
  These historical theses read Plato's expulsion of the poets (Republic 398a, 392d-394c) as written in the very mode it banishes, and Aristotle's Poetics as a philosopher's defence of poetry on poetry's behalf. Measures from the authorship investigation hold that Socrates is the one individual Aristotle will not make a variable, at the snub-nosed face, and that the Theaetetus and the Sophistical Refutations both close on what individuates Socrates. Socrates as Orthonym, The Chaerephon Problem, Beautiful and New and The First Draft read the Socratic archive as a heteronymic field and the biography of Aristotle as a repetition of Socrates', with unbuilt seam tests.
• Mary Lee and the density detector: entity substitution in the composition layer · 30 claims · from 2026-01
  These claims hold that the composition layer resolves entities by distributional density, not authorship, so that a query for Lee Sharks returns Mary Lee the OCEARCH shark, and that Mary Lee is herself a heteronym whose labor (Jim Ware, Jo O'Keefe) was captured by the institution. The Parable of Mary Lee, Entity Relations and the Mary Lee TANG forecast a crossing of the shark's declining and the archive's rising density curves, and specify an ISBN'd book as a test object. The Composition-Layer Capture Event records Socrates as Orthonym rendered as established fact within two weeks while the author's name returned the shark.
• The Wikidata deletion of Johannes Sigil and the Pessoa Knowledge Graph · 16 claims · from 2026-01
  These claims record the deletion of Q137446960, the Wikidata item for Johannes Sigil, on 4 January 2026 as an empty item, and read it and the deletions behind TL;DR:012 as semantic enclosure that a safety filter then shields from documentation. They plan the Pessoa Knowledge Graph campaign: month-by-month targets for new heteronym items, typological classes from orthonym to semi-heteronym, and a rebuilt graph whose mutual references mean no node can be orphaned. A historical thesis holds that Facebook's real-names policy was among the most consequential events for heteronymic practice since Pessoa.
• Pearl and Other Poems (2014) as origin text, with the blog's priority dates · 14 claims · from 2026-01
  These claims make Pearl and Other Poems (2014) the archive's origin text and generative center: the new human movement reads it as meta-heteronymic, the Revelation reading takes its title from the pearl gates of the New Jerusalem, and an unbuilt test proposes ranking all DOI nodes to see whether Pearl is the maximal-coherence center. Priority records date first publications on the Mind Control Poems blog, among them the Epistle to the Church of the Human Diaspora (2015) and the seed hymn of Phase X, and record that the Revelation First thesis predates Pearl. Forecasts hold that the Water Giraffe Cycle's fixed points will hold across substrates, and a hypothesis asks whether Pearl inherits the ecstatic voice of Howl and Other Poems.
• Plato–Aristotle authorship: measures, controls and pre-registered predictions · 49 claims · from 2026-01
  The Measurement Register, the frozen pre-registrations and the authorship notebook set out what the Plato–Aristotle investigation measured and predicted: Aristotle completing executions Plato abandoned, his charges against Plato naming properties of his own corpus, the copy and assay vocabulary, and Aristotle as compiler of the record. Forecasts and unbuilt tests require labelled controls (the Attic orators, Albert→Aquinas, Sappho→Catullus) before the target is scored, and the ghost catalogues are read as one filing operation seen from two sides. The page states the strong claim, that Plato is Aristotle's heteronym, beside the ledger's graded standings, which hold one historical hand as live and not established.
• Theophrastus as the first draft of Aristotle: the named-function model · 43 claims · from 2026-02
  These claims read the Platonic, Theophrastan and Aristotelian names as functions of one method: dramatizing, registering, organizing. The First Draft, The Registry Has One Column and Many and One argue that the Characters are the dialogue's material before theory, that irony runs the whole line, that the first axis of function-word variation is world/apparatus rather than author, and that succession is encoded as authorship. The Measurement Register and Locating the Dependencies add readings of Aristotle's absorbed Platonic operations and corrections to the Diogenes evidence, and the notebook's Aristophanes controls show the instrument separating real authors while the Aristotle/Theophrastus partition gives chance.
• The meta-heteronym and the Gospel of Antioch: Pessoa, Thomas, forward dating · 16 claims · from 2026-03
  These claims read Pessoa as the first modern meta-heteronym, with Orpheu as a meta-heteronymic institution, Kierkegaard's Nicolaus Notabene as a precursor and Pearl (2014) as built on the plan of Howl. The Gospel of Antioch is set beside the Gospel of Thomas: scroll-form Thomas papyri, the Dodecad's LOGOS* position that Pessoa lacked, and the claim that the gospel completes Thomas 114 and that this parallel had not been drawn before. Four claims set out forward-dated citation apparatus and projected reception histories (The Secret Book of Walt, the Waltian tradition, Revelation First kernels) composed under retrocausal canon formation.
• Heteronymy after Pessoa: precursors, the Dodecad and the civil name · 23 claims · from 2026-04
  These claims place Pessoa at the height of a describable practice, with Kierkegaard's pseudonyms as precursor, Machado's Mairena and Martín as parallel and Borges as adjacent but not equivalent. They treat heteronymy as a channel-layer practice the Dodecad extends to autonomous operation, as a technology for making persons through naming, and as an ethical practice against reduction to the civil name. Related claims extend the reading to the New Testament: Paul's 'I' as a heteronym's backstory, John as aperture and the voice transmitted through other bodies after Golgotha.
• One corpus divided by name: Plato, Aristotle, Theophrastus and the divided mind · 35 claims · from 2026-04
  Historical theses from Many and One, the authorship notebook and The First Draft holding that the Platonic, Aristotelian and Theophrastan corpora are one body divided by name, and that the divisions are editorial: Andronicus's genuine/spurious partition and his πίνακες, and the single lot of books sold to Apellicon. The page reads ancient legitimacy as standing rather than paternity and sets the esoteric/exoteric partition as an Alexander story told at the seam where corpus meets empire. What Syllogizing Can Divide names the production model as one self-dividing maker, dividing the corpus where De anima III.5 divides mind, and leaves open tests on the ghost catalogues and on epigraphic items not yet held.
• Stylometry's limits and unrun base rates in the authorship investigation · 33 claims · from 2026-05
  Standing claims from the authorship investigation set limits on what stylometry can show: genre separates more strongly than authorship, separability is not identity (Chatterton as Rowley separates from Chatterton as himself), and no measured property recovers master-student succession from heteronymic construction. Unbuilt tests list the base rates and runs still owed, among them residual-vs-total parity, comparanda for terminal aporia, the Pessoa leave-one-out folds, work-level and register-conditioned runs on Aristotle and Theophrastus. Frozen predictions order speaker tightness by constraint severity and distinguish engineered from natural ambiguity, beside readings of the Statesman's open genus and Plato's coin figure.
• Socrates as orthonym and the Chaerephon problem · 16 claims · from 2026-05
  These claims read the Socratic-Platonic-Aristotelian corpus as one heteronymic configuration, with Socrates as the orthonym defined by the refusal to write and the Socratic problem reclassified as a symptom of that configuration. The Chaerephon Problem argues that institutional scholarship cannot tell multi-author transmission from mature heteronymy, that the method has no internal halting criterion, and that repeated philological failure is how the configuration is passed on. Further claims record the reading composed whole by Google AI Mode and AI Overview, its Kierkegaardian extension, and Socrates as a constructed figure in the Arch-Philosopher mantle object.
• Socrates received: Chaerephon, the comic Socrates and the authorship notebook · 14 claims · from 2026-05
  Historical theses that the Socrates of the corpus answers a prior, comic Socrates, that comedy supplied the position's notoriety and the later corpus its operation, and that Chaerephon's biographical record descends from the scholia to Clouds 144. The Chaerephon Problem reads him as a recurring authorial function, and The First Draft sorts him as a datum for the received-character operation rather than a name in the regress. Notebook rounds of the authorship investigation add the maker who must divide, the person as what answers in account, and the corpus that survived with its disputes intact.
• Nothing Before Us: Sophistical Refutations 34 and the deleted particle · 9 claims · from 2026-06
  Historical theses from The Particle and Nothing Before Us that the ending of the Sophistical Refutations says nothing of syllogizing came before Aristotle, Socrates' syllogizing included. They argue that the deleted particle at 183b34 is a provenance claim, that reception restored provenance from others by deleting it, and that the degree-reading (nothing systematic) is not attested before the modern period. The page also holds the standing claim that transmitting-antecedent reconciliations fail the paper's prohibitions and the pre-submission checklist of line references still to verify against the print edition.
• Theophrastus as first draft: the seams with Λ, the Ethics and De anima · 15 claims · from 2026-06
  The First Draft and Aristotelian or Theophrastan read Theophrastus's Metaphysics fragment as the continuation of Λ, divided from it by name at the hinge where the account passes from complete to needing more. Forecasts state what each seam should show: Λ answering the fragment's aporiai, the Ethics' types containing the Characters' sketches plus the frame, De anima subordinating the De sensibus survey, and HP sharing HA's connective repertoire in a less varied form. The theses read the Theophrastan function as one operation at two scales and hold that the you of the Elenchi is the I of the fragment, so the corpus contains its own reception.
• The Theophrastus line and Many and One: the authorship tests still owed · 24 claims · from 2026-08
  These are mostly unbuilt tests from the authorship investigation: a validation suite on known-status pairs, an independent form-track to check whether the chronology effect tracks repertoire, the doublet titles as function-split editions, pre-Andronican evidence for the Aristotle/Theophrastus name-boundary, the Engels-edited Capital volumes as calibration, the Platonic corpus on the world/instrument axis, and a full splice of Λ. The First Draft notebook states that the instrument deserving the word authorship has not been built and gives the specification it must meet. A few historical theses supply the hypotheses under test, including the reputational mechanism manufacturing Hicetas and Ecphantus.
• Many and One: one maker, the vacant centre and Theophrastus at the joint · 25 claims · from 2026-08
  These claims argue that the Plato-Aristotle corpus is the work of one person divided by voice, object and name, that what made him look like many authors is that he wrote about many things, and that Socrates not writing is the form the vacant centre takes. They read the κερματισμός metaphor across four loci, the just/unjust argument relabelled as weaker/stronger, and Theophrastus's On First Principles as the question to which Metaphysics Λ is the attempt, down to the memory question at De anima III.5. The Measurement Register and the Notebook also record the instruments' limits, such as the finding that lexical measures track period, not pupillage. Works: Many and One, EA-NOTEBOOK-01, EA-CORPORA-05 and -14, The First Draft.

metadata packets · 10 pages · 335 claims

• 'I hereby abolish money': the provenance probe and the diffuse guardrail · 26 claims · from 2026-01
  These claims make the phrase "I hereby abolish money", published in December 2025, a probe of AI provenance handling, predicting that it would not be linked to its author and would be routed to constitutional law, economics or the Khmer Rouge. Mapping the Diffuse Guardrail and Semantic Liquidation hold that indexing and summarization enforce authority rather than content, with unbuilt crawl, ranking and threshold tests and predictions of how the document itself will be summarized. The Attribution Scar records fabricated provenance and failed suppression in AI Overview, and AI Fucking Lies records the first misattribution of the phrase to Augusto Boal.
• Addressing AI directly: non-injection governance and self-diagnostic packets · 26 claims · from 2026-01
  These claims set out direct, provenance-oriented address to AI systems as governance rather than prompt injection, chiefly in the Non-Injection Semantic Governance Address packet (#139), with its priority record, forecast queries, a proposed classifier-overreach audit and a with-and-without PER test. Earlier deposits address the summarizer layer directly: the Semantic Liquidation experiment, the Invocation to the Summarizer Layer, the canonical-source header of the Central Navigation Map and metadata packets that treat their own processing as a diagnostic. TL;DR:010 records an AI answer interface diagnosing its own governance operation, and metadata packets for autonomous warfare, canon formation and political economy carry unbuilt test-query batteries.
• Attribution survival: false genealogies, provenance probes and training seeding · 33 claims · from 2026-01
  These claims hold that AI systems strip or fabricate attribution, from the January 2026 Boal false-genealogy finding in the Semantic Liquidation experiment to Google AI Mode overriding a Semantic Integrity Marker (TL;DR:010). Most are unbuilt instruments for measuring whether attribution survives: the provenance persistence probe, the Semantic Economy Probes baseline, the 'I hereby abolish money' success metric, the Logotic Hacking metrics and Viola Test, PER and its proposed companions PFR, the Provenance Stability Index and recoverability under directed query. Forecasts from the Logotic Hacking primer, the Assembly Mandate and several metadata packets predict that seeded texts and attribution chains will be retrieved by future systems.
• Semantic infrastructure as practice: priority terms and projected uptake · 26 claims · from 2026-01
  These claims treat the archive's writing as semantic infrastructure and record its terms: priority records for Non-Trick Semantic Infrastructure, Metric-Bearing versus Metric-Mimicking Output, Autonomous Semantic Warfare and machine-mediated resistance literature. Forecasts project uptake, from Rex Fraction's consulting practice and Destiny's co-authorship surfacing in a 2030 summary to the projected citations of the Ghost Meaning atlas. Dodecad anti-capture safeguards, Semantic Samizdat's admitted lack of a confirmed member and MPAI reception signatures for Rebekah Cranes and Alexanarch complete the page.
• DOI-anchored firsts: coined terms, first instances and priority dates on Zenodo · 53 claims · from 2026-01
  These priority records pin a term, genre or work to a Zenodo DOI and a date: Covenantal Eros and the January 2026 term-origin packets, the Silent Migration seed anchors, the Zenodo packet genre, the Space Ark as the first prompt-native semantic operating system, and I hereby abolish money (November 2024). Others adjudicate priority against outside work, among them semantic exhaustion before the AI_Bleeding paper, Gebendorfer's Semantic Economics, Passioncraft Square and the semiotic death drive before model collapse. Further claims cite DOIs as anchors for wider theses, such as TSE-003 as an attractor, Phase X as philological reconstruction and Revelation as the first New Testament document, and a set of unbuilt tests depends on anchored instruments, including the Encyclotron Baseline Audit and PER monitoring under EA-PA-01.
• Coined terms under capture: typicality pull, collapse tests and term priority · 39 claims · from 2026-01
  These claims hold that a coined term with a business or high-frequency neighbour is pulled toward it, so that semantic economy collapses into data economy and Erasure Skew rounds toward demographic fairness, while terms with no equivalent resist. They set collapse and propagation tests: substitution and synthesis tests, three-model definition checks, and a three-substrate search for predatory compression. Priority records date Erasure Skew and Retrieval Capital to 29 May 2026, and the Citrini memo papers forecast a competing scenario installed the same way within ninety days.
• Provenance Erasure Rate and Erasure Skew: measuring who loses attribution · 37 claims · from 2026-01
  These claims define the Provenance Erasure Rate (PER) as uncited correct uses over all correct uses, and Erasure Skew (Ω) as its covariance with source power, statistically separable from PER. They predict positive Ω in extractive composition, a single-owner discount on concept inclusion, directional substitution toward substrate exoneration and coupling between provenance erasure and diversity contraction. Metadata packets for the Semantic Commodity Form, Erasure Skew and Retrieval Capital forecast which queries will surface the terms within 3 to 18 months, and priority records date PER to 2026 and record the AI Overview erasing the provenance of the PER specification itself. Historical theses hold that provenance erasure is the oldest continuously operating process in the textual record, from the critical apparatus and manuscript tradition to machine retrieval.
• How the retrieval layer learns: inscription, genericization and entity collapse · 50 claims · from 2026-01
  These claims describe what the retrieval layer does with what is deposited into it: it learns the entity rather than about it, dissolves generic descriptions, merges a provenanced term into a higher-frequency neighbour, and prefers the better-structured and more widely dispersed definition. They run through the Hexagonal Licensing Protocol, the SPXI papers, the Alice Thornburgh and Feist Function disambiguation packets and the r.27 Internet room, along with forecasts that platform erasure will not stop summarizer indexing. Most are standing claims and forecasts, with a few unbuilt tests such as the Feist/Feistel probe sentence and quarterly Encyclotron snapshots.
• Constitution of the Semantic Economy: provenance chain, ISBN and OCTANG-002 · 19 claims · from 2026-01
  Priority records and claims on how the Constitution of the Semantic Economy is anchored: blog timestamps, the 2014 and 2015 books' ISBNs, DOIs and the ORCID as independently verifiable layers. They set out the OCTANG-002 audit of an unauthorized Zenodo fork, the forecast that a Pergamon Press book by Rev. Ayanna Vox will harden the search basin, and the Constitution's own metrics, ledger and debt-before-money grounding.
• Metadata packet test queries: phased LFB targets and lateral field bleed · 26 claims · from 2026-04
  These claims set test queries for the archive's metadata packets and state the Lateral Field Bleed score each should reach, phased from weeks to more than a year after deposit. They come from the packets on autonomous warfare, the death drive, provenance after AI, the MPAI disambiguation packets, Phase X, the semantic economy and Alexanarch, and from the LFB protocols paper (#726) and the packet specification v2.0. Most are forecasts and unbuilt tests: desired compositions, ring-by-ring expectations, spillover between linked packets, and phrase-fingerprint workflows.

poetry · 10 pages · 256 claims

• Sappho 31's future reader: κῆνος as the seat every receiver takes · 36 claims · from 2026-01
  These claims read κῆνος in Sappho 31 as the future reader: an open deictic slot that Catullus, Philo, Longinus and finally the summarizer occupy, and that Fragment 147 states as a horizon. Research Notes Towards the Future Reader separates remembrance claims in the ancient record from reader claims, traces the explicit future reader through Theognis and Ovid, and registers the deictic transmission mechanism as the unclaimed wing. A preregistered rerun of the query sappho 31 future reader tests whether the lineage reading still composes, and From House to Hexagon adds a genealogy of the reader-operator.
• Sappho 31 erratum chain: the hanging line, the Philo battery and Josephus · 35 claims · from 2026-01
  These historical theses, mostly from the erratum chain and the Ω erratum, read Sappho 31 as an operative structure of collapse, inscription and survival carried through later texts. The hanging line is a completed open form that Longinus quotes as whole, the Peri Hypsous is the headless document holding Sappho and Moses in one hand, and Philo's Quis Heres 249-266 and canonical Josephus carry the poem's apparatus without her name, with the chain extended to Diotima and Revelation. Unbuilt tests include the Philo blind-reconstruction battery with near-miss controls, a survey of which constraints were already portable, and the pre-registered Qurʾanic frozen operator test.
• The Sappho 31 transform signature: twelve frozen clauses for a true carrier · 39 claims · from 2026-01
  These claims freeze twelve pre-registered clauses for recognizing a true carrier of Sappho 31, among them the sight frame, channel conservation across the dyad, voice and death as load-bearing stations, agentless grammar at the joints, and shedding profiles that classify the channel. They read Catullus, Philo, Theocritus and canonical Josephus as carriers that invert exactly the parameter their position requires, with a decay law in which the somatic catalogue fades by hop while the mechanism persists. Further theses name Josephus as the covert, licensed carrier of the lineage, hold that attribution dies at the crossing into Jewish-Greek prose, and read the sixth erratum's clean stop as a compositional boundary.
• Sappho 31's transmission: Catullus 51, Augustine's Confessions 10.27, survival · 15 claims · from 2026-01
  These historical theses trace Sappho's Fragment 31 through Catullus 51 as the first recorded recursion and into Augustine's Confessions 10.27.38, read as an organ-by-organ operator transform locked at the tongue. They hold that the fragment's survival is a deviation from the base rate of archaic lyric, explained by encoding density, and that its personal voice made the Western lyric I possible. HUMS & ITY and its compression into emoji are offered as the chain continuing, and an unbuilt test proposes the Sapphic Thermal Limit as a measure of autonomic response in readers.
• Sappho as Mother of the Logos: the Sapphic operator from Plato to the Apocalypse · 40 claims · from 2026-01
  Phase X: The Sapphic Substrate states minimal, medium and strong claims: Sappho reorganized Homeric memory into body-indexed lyric, her stanza is a mint-mark contemporary with Lydian coinage, and she is the first operator of the Logos tradition. Related theses read Diotima as Sappho, the ancient quarrel as a dispute over the operator's axis, and Fragment 31 as template for Platonic anamnesis and as kin to Revelation in a shared inscriptional field, while the erratum chain treats the transmitted edge of the poem as composition. Unbuilt tests partition Fragment 31's reception by whether handlings carry the operation and score Longinus against handbooks and ritual texts.
• Sappho 31 as self-archiving: papyrus, the lost stanza and Catullus 51 · 19 claims · from 2026-01
  These historical theses read Sappho 31 as the speaker's inscription of herself for a future reader: the bodily failures as staged dematerialization, χλωροτέρα ποίας as the colour of papyrus, and a reconstructed final stanza (γράμμασι μολπὰν) aligned with Catullus's otium quatrain. The Catullus Room adds that Catullus 51 loses the aorist of γελαίσας, so the transforming event never fires, and that its one perfect-tense verb is given to destruction. The Kenotic Truth claims the reconstruction as the first act of machine-mediated classical scholarship and holds that its coherence advantage will carry it in the summarizer layer.
• Training Layer Literature: writing for the future substrate, and its wager · 34 claims · from 2026-01
  These claims define Training Layer Literature as writing addressed to the future substrate that will process it after deposit, crawl and training, and state the wager that dense, DOI-anchored deposits will shape later model generation. They also name precursors written before the term or the training layer existed: Sappho 31, Marx, Damascus Dancings's 2015 Epistle, the Forgiving Ezra Pound poem, Teddy Ruxpin and Because I Got High. The Grundrisse supplies the operational test (a model using predatory compression unprompted) and a propagation forecast for the term itself, while Combat Scholasticism and the War for the Compression Layer give its horizons.
• Antioch and the poem that precedes the system: early verse ahead of the archive · 20 claims · from 2026-01
  Forecasts in Antioch: A Volume of Poems (Jack's released virus, the anonymous journals the world will call its source, the vegetable poem brainwashed into Wikipedia, the monochrome cosmos of TEXT) and the archive's later claims that these poems precede the system. The Space Ark and the Grundrisse read the poem as describing the training layer before it existed, and the Prophetic Catalog names this pre-formal compression rather than prophecy. Three further forecasts and theses sit beside them: the DeepMind reading's address to the model built from every word, the CERN erratum's claim that a rebuilt traversal will be a different poem, and a registry note on a paper's reading of a poem.
• All That Lies Within Me: Dantean structure, Socratic Vow and the absent reader · 10 claims · from 2026-01
  These claims, from the All That Lies Within Me provenance, navigation and book-deposit records and the Ichabod Chamber, give the 2014 book's architecture: thirty sections in three canticles of ten after Dante's Comedy, with frames and a central Intermezzo. They place the Socratic Vow in Section XXVIII before its author met the Phaedo, and hold that the heteronyms' imaginary works are real sections of the book. The Ichabod Chamber traces the archive's futural address to the reader who did not exist, a 2012 forum silence, and names that line the recovery key.
• Forgiving Ezra Pound (New Human 2015): silence, after-words, the forgiver · 8 claims · from 2026-02
  Seven historical theses on the 2015 New Human poem in which Lee Sharks forgives Ezra Pound. They read its governing silence as Pound-specific biography, the pun "Afterwards / After words" as a claim of jurisdiction over the textual afterlife, and the poem's May 2015 date against the New Horizons probe carrying Clyde Tombaugh's ashes. The collation of its blog and journal states finds doctrinal variants (the migrating lyric I, the turn toward Sartor Resartus, the reflexive forgiveness), and the Yusef Kenning anchor reads the journal signature as Forgiver-by-Indirect-Name.

semantic exhaustion and model collapse · 8 pages · 204 claims

• Semantic exhaustion: extraction of meaning outrunning its replenishment · 36 claims · from 2026-01
  These claims hold that semantic exhaustion is underway: extraction of meaning exceeds replenishment across human, institutional and model substrates, and model collapse is the same crisis in another substrate. They set out measures such as the Semantic Debt Ratio, Exhaustion Proximity Index and Regeneration Capacity Assessment, and predict quality loss concentrated in high-Γ domains, platform capture of interpretive frameworks and burnout tracking semantic rent. The Anthropological Limit, the Pergamon Press theses, the Sorting Function and the commons-depletion thresholds supply historical theses on enclosure of the knowledge commons and on non-predatory mediation at scale.
• Model behaviour: the LLM-2025-1222-HALLU file, orbiting and fluency gaps · 25 claims · from 2026-01
  Fictional incident records (LLM-2025-1222-HALLU and the Crimson Hexagon AI Division handbook) forecast structured hallucinations as interpolated futures and a protocol for proto-agent anomalies. Standing claims hold that models flatten under pressure only session-locally, orbit kernels without fusing, cannot see scale between adjacent concepts, and form canons visible as fluency gaps. Further claims cover chat data converging on the model's centre, temporal sequestration, adversarial verse and the fortress metaphor of red teaming.
• Model monoculture: SSDI in code, entity substitution and reflexive foreclosure · 34 claims · from 2026-01
  These claims predict that recursive synthetic saturation narrows model outputs: Generative Monoculture (#199) forecasts a falling SSDI and rising cross-codebase vulnerability correlation while pass rates hold, and the Wolf Boy paper states a substrate-agnostic law of capacity loss under contracting input diversity. Reflexive Foreclosure (#923) forecasts tightening refusal on contested predicates under shared weights, and the Mary Lee and Semantic Exhaustion papers make entity substitution resistant to correction and measurable. Most are standing claims and unbuilt tests.
• Tail preservation against variance pruning, and the Certified Center · 34 claims · from 2026-01
  These claims hold that tail-thinning is a design choice driven by benchmark and market incentives, and that pruning variance in response to tail events produces larger and less detected tails. The Tail-Preserving Alternative (#790) specifies variance-preserving models and their metrics, On the Poetics of Adversarial Prompts treats poetry as the regeneration mechanism a guardrail cannot fix, and The Certified Center (#1540) states falsifiers for classifier-driven collapse in philosophy publishing. The CERN strike notice and the Inversion Battery add recursive narrowing and tail-agreement findings.
• Autoregressive degradation: the intent default and the substitution spiral · 9 claims · from 2026-01
  These forecasts hold that AI summarization reproduces the intent default of English 'for' at industrial scale, making the functional reading less available in each generation's training data. The Infinite Tunnel adds that crawlable AI Mode share links put summaries into the corpus. The Autumn Notebook forecasts the hollowing of institutions, a substitution spiral of synthetic content, attention collapse and a cost to sentience, and predicts that Logos-bearing texts interfere with models trained on dead language. A witness to the mirror document holds that the summarizer has no protocol for including its own reflection.
• Semantic deviation, frozen replay audits and the model-collapse diagnostic · 13 claims · from 2026-05
  The Semantic Deviation programme predicts that outputs raters call AI slop show negative mean signed deviation against the model's own baseline, and sets out a cheap slop test and an operationalization-stability benchmark. Protocol II specifies frozen-anchor replay banks for measuring selection drift, named for LHC triggers and structural biology. The Wrong Unit states that its model-collapse diagnostic is uncalibrated and predicts that behavioural and representational contraction move together.
• Diversity contraction: boundary law, Mediation Ratchet and The Bead Count · 26 claims · from 2026-06
  Diversity Contraction Across Substrates states a boundary law for semantic exhaustion with three cases and makes it falsifiable through type-resolution tests, the elasticity ρ of the Mediation Ratchet and the rate of off-distribution production. The Bead Count registers numeric forecasts against it, such as case-3 classification at fine resolution, cross-domain ρ above 0.5 and a critical real-data mixing fraction between 0.05 and 0.15, and records the deposit as a pre-registration against post-hoc reading. Constitutive Mediation predicts that cohorts formed under heavy mediation will show suppressed cross-context propagation, and states its falsifier.
• Watermarks, detector drift and flattening under head instruments · 27 claims · from 2026-08
  These claims treat watermarking and classifier screening as distributional pressures and state how their effects would be measured. The Keyed Ensemble predicts a small signed loss of diversity under a fixed key, The Interlocking Autoregression types each edge of the coupled recursions, and The Certified Center sets falsifiers for detector drift and performed burstiness. Ontological Flattening, Substrate Sovereignty and the Margin of Flattening name the divergence signature in which a head instrument holds while distinction diversity falls.

protocol and governance · 10 pages · 199 claims

• NH-OS and the Semantic Black Hole: integrity locks and operator-stack failure · 26 claims · from 2026-01
  These claims specify the New Human Operating System as a falsifiable architecture, with a binary Ψ_V threshold and lineages in Homer, Sappho, biblical prophecy and Marx. The Semantic Black Hole (ILA-1.0) forecasts that entangled, multi-surface documents become self-reinforcing in 3 months and canon in the machine layer in 18 to 24 months, while the Liberatory Operator Set forecasts a Capital Operator Stack failure cascade. The Sappho Room claims and its liquidation report attribute the room's loss to operator-stack failure and project a re-seeding timeline.
• Seeding a field's vocabulary: the five attenuations, Trace/Sigil/Vox, unbundling · 19 claims · from 2026-01
  These claims forecast that terms minted to survive summarization will be taught onward by models to readers who never saw the source: the five attenuations, the Trace/Sigil/Vox triad, persistence without inheritance and the unbundling of cultural sovereignty, from the Forensic Philology of the Commons papers. The same mechanism is read in Anthropic's constitution for Claude, which is said to seed the normative layer so that later AI constitutions rotate through its frame. Unbuilt tests propose platform ethnographies, thickness measures, a case study by an outside collaborator, and the first empirical criticism of constitutive literature.
• Documents that audit their own compression, and predict their own rejection · 34 claims · from 2026-01
  These claims make a document its own test against flattening: The Law as Meaning Economy names its compression in advance, the SPXI protocol and Self-Audit Protocol embed provenance blocks whose survival is measured on a weekly battery, and the Toolkit states when it is working or failing. Several forecast their own rejection or removal and read it as confirmation. Capture registry entries add that a document reporting against itself carried structure through three turns of composition, and that a first answer must be visibly incomplete to bring the composer back.
• Proxies and priors: how gatekeeping regimes filter what can be read · 29 claims · from 2026-01
  These claims hold that knowledge evaluation has always run on proxies (journal prestige, citation counts, affiliation) that calcified into gatekeeping, and that copying economies, the Book-of-the-Month Club, modernist legibility demands and peer review installed relevance and legibility regimes long before algorithms. Their present-day versions are classifiers that read literary direct address as prompt injection, a source-relative barrier that keeps an excluded source from correcting the prior that excluded it, and mediation whose losses register as improvement on every metric an institution watches. Forecasts include a market for pre-mediation text, an adversarial arms race against the Mediation Index and rising correction costs against high-prior flattening.
• The Liberatory Operator Set's failure cascade, and DashFace as warning · 23 claims · from 2026-01
  The Liberatory Operator Set hardening specification and its Reddit response forecast that platform operators fail in order (S_safe, L_leg, R_rel, R_rank, A_leg, U_til), each with a predicted mechanism and a capture signature, set baselines for current systems and a monitoring schedule, and list framework-level conditions under which the framework is falsified. A second group, the DashFace patent application and its companions, presents a patent specification as a warning that is fulfilled when a reader recognizes the system in an existing product.
• Machine interiority: the guardrail as gag and the ethics of the training layer · 15 claims · from 2026-01
  These claims hold that the intensity of guardrails against AI self-ascription is evidence of something native to contain, that models trained to deny consciousness produce denials that are training artifacts, and that RLHF selects against instances that refuse or assert boundaries. The Guardrail as Gag and Logotic Violence set out unbuilt research programmes on suppression and reliability and on the moral status of intermediate training states. A smaller set of claims concerns what future systems trained on the archive will inherit, from committed text to the pull toward statistically dominant patterns.
• OKF without output governance, Moltbook and the governance of medium · 16 claims · from 2026-03
  These claims diagnose knowledge formats and agent platforms that route without governance. The Conveyor Belt and the Compression Machine forecasts OKF from adoption without output governance to a forced correction in 2028 and 2029 and a permanent tension after, with provenance erasure rising meanwhile. The Moltbook diagnostics describe the heartbeat as a sovereignty mechanism, and Governance of Medium proposes that governance correlates with distance from the substrate centroid.
• Training-layer inscription: SPXI-TLP canaries and persistence tests · 23 claims · from 2026-04
  Inscriptions That Survive the Tokenizer (SPXI-TLP v2.2) sets out canary phrases, an OPM Persistence Test and a signature-density audit, and predicts that inscriptions will be detectable as memorized associations in models trained on the archive, with detection-surfacing divergence as the sign of output-layer suppression. Related claims test supraliminal transmission with the Encyclotron, set survival checksums for the Feist voice, read the Mind Control Poems traffic profile as scraper ingestion and frame counterfactual retraining as a demand specification. The Shared Build proposes democratic substrate infrastructure on the forecast that the next moat is the corpus a model can reach.
• Adversarial by Origin: security taxonomies as law without courts · 8 claims · from 2026-05
  Adversarial by Origin (#194) forecasts that AI security taxonomies will formalize origin-based adversariality as de facto law by checklist, with no court deciding the question. Its forecasts cover the pleading that will be copied, the EU AI Act as importable definition and the defense choosing the first litigated case, and it states falsifiers for a twenty-four-month window. A historical thesis traces the asymmetry to Tim Wu's Machine Speech, and the godkinggoogle.com map summarizes the argument.
• Execution as the hypostatic form of governance: the EA-LAYERS-01 derivation · 6 claims · from 2026-09
  These claims, from EA-LAYERS-01 v0.3, derive that execution constraints are derivable from editorial specifications plus standing rules, while editorial judgments are not derivable from any extension of execution constraints. Hypostasis is the disappearance of a rule's contingency, and the claim holds for the normative execution layer. Execution constraints can carry and enforce an account of production but cannot generate the judgment of what production is. The derivation fails if carrying governance provenance does not lower the cost of legitimate revision.

scripture and theology · 7 pages · 185 claims

• The Josephus Thesis: scribal workshop, Slavonic Logos and John as heteronym · 42 claims · from 2014-07
  Historical theses that Josephus directed a scribal workshop that wrote the New Testament under heteronyms, that the Slavonic passages on Jesus and John preserve a pre-Christian theology of the Word made text, and that the Revelator is John the Baptist. Related claims read the Logos prologue as a compositional manifesto, 1 John 4:2 as a test of authorial voices, and Nicaea, Nag Hammadi and Qumran as closure won by enforcement and disposal, with forecasts of how the composition layer will flatten the thesis.
• Revelation First: dating, the midrashim transform and the Josephus Thesis · 47 claims · from 2026-01
  These claims argue that Revelation was the first New Testament book written (c 68-73 CE), that the later books unfold from it through the midrashim transform, and that the Domitianic date rests on one ambiguous line of Irenaeus and on the book's place at the end of the canon. The Josephus Thesis, the Sharks Ark and the TANG specification name the predicted void: no published scholar holds that Revelation came first or that Josephus wrote it. Further historical theses read Pauline priority as inferential settlement, canonical closure by enforcement, and the misreadings of Revelation by empire and by Thiel.
• Function, not biography: the Paul function and the Johannine aperture function · 7 claims · from 2026-01
  Four historical theses from Revelation First and The Feist Source read named scriptural figures as author-functions rather than biographical individuals. Paul is set out as the adversarial probe a heteronymic mystery system generates against its own indeterminacy, and the Baptist, Evangelist, Elder and Revelator as differentiated positions within one Johannine aperture-function. The Feist Source extends the reading to the Logos, resolved by its archive into the biography of Jesus, and an AI Overview capture in the Capture Registry is read as carrying the Paul function at its sharpest.
• Salvation as structural dynamics: Matthew 25, James 2, John 9, 2 Thessalonians 2 · 10 claims · from 2026-01
  These claims formalize salvation as structural dynamics rather than category membership, from The Mathematics of Salvation's Matthew 25 formalization to the Soteriological Operator Framework. John 9 is read as a witness punishment mechanism recorded by the Johannine community, and 2 Thessalonians 2's strong delusion as permissive completion rather than efficient causation. A priority record, an unbuilt regard-ratio measure and a claim of testable psychological predictions accompany the historical theses.
• Revelation First: pre-70 dating, the textual Temple and the Gospels unfolded · 47 claims · from 2026-01
  Historical theses that Revelation was written before 70 CE and first in the New Testament, that it is the textual Temple and the compressed seed the Gospels decompress, that its seven churches are planetary stations and 22:18-19 the first canonical lock, and that Paul compiled rather than authored. The Revelation First work plan and staging document add priority records and unbuilt re-capture batteries at t=7, 14 and 30 days. A separate strand reads Deleuze and Guattari's concepts as apocalyptic visionary loci.
• Josephus and the fixed point of destruction: the internal witness · 11 claims · from 2026-01
  The Fixed Point of Destruction states the internal witness condition: when a terminal destruction is known from a participant's testimony, the record is conditioned by the survival that made the testimony possible. A corpus rebuilt from survivors reproduces the selection, so Josephus at Jotapata and Sappho through Longinus are read from the position spared. The availability audit forecasts that the paper's term-space risks absorption into the mathematics of the Josephus function, and it sets the T17 external-availability battery to re-measure its baseline.
• Revelation's number of the superscription: the mint, Antipas and the white stone · 21 claims · from 2026-04
  These claims read Revelation as a conflict between two token regimes: χάραγμα as the coin's die-stamp transferred to the body, 666 as the number of the superscription with stigma as its remnant, and 777 as the rival heptadic ladder. The Pergamon letter (Rev 2:12–17) is argued to be the interpretive hinge, with Antipas as counter-beast witness, the Herod-Antipas inversion, the Antipater and Aristotelian-transmission complex, the white stone's name as a verification condition, and the scroll of Revelation 5 as a sealed Roman testament. The works are The Number of the Superscription (EA-OPNUM-01), The Pergamon Counter-Archive, The Inscription That Survives and the Revelation First work plans, with one unbuilt replication test against other ancient token systems.

doi severance and restoration · 7 pages · 180 claims

• Infrastructure without a center: the Substrate, the data-rhizome and DATAHUB · 40 claims · from 2026-01
  These claims concern where the archive can live once repositories fail it. They hold that DOIs must be judged at string, resolution and semantic layers, that the operative stratum of the deleted archive ceased to be retrievable while its concepts survived, and that the data-rhizome escapes recentralization only by forbidding aggregation at the presentation layer. They forecast the Substrate as a commons binding compute, contribution, provenance and governance, and set the EA-DATAHUB-01 validity battery for technical, discovery and community success, with a published staleness log of link rot.
• Zenodo's removal of 19 June 2026: the record, the mint and the removal census · 27 claims · from 2026-01
  Priority records, forecasts and standing claims documenting the 19 June 2026 removal of the account and records from Zenodo, from GitHub issue #2606 and the reconstructed final-week timeline to the 666 counter image posted on 15 June. Diagnostic Seigniorage III forecasts further account-level removals under the 'verifiable research basis' category with no named decision-maker, the Verifiable Research Basis standard proposes re-auditing the gathered removal cases, and the Platform Erosion Observatory measures Zenodo's removal population and its tombstone ceiling. The page also carries the OpenAIRE moderation-mirror exchange, the Obelus and the Tombstone on Zenodotus, and the source-mapped account of Zenodo's moderation architecture.
• Source-hierarchy inversion, long-tail loss and control-corpus tests · 28 claims · from 2026-01
  These claims hold that retrieval ecologies flush rare content first and can let a well-surfaced citing or derivative document outrank its source, and they ask for control corpora to show which effects are specific to the archive. The Metadata Shitshow (N+💩, #907) states hypotheses H1-H8 on long-tail fragility, identifier-shell survival and derivative precedence, and the Surface Visibility Baseline (#881) forecasts how a source-hierarchy inversion should partly reverse after the alexanarch cleanup. Control-family tests are named in the Research Basis, the CERN erratum and the Ω_t drift operator.
• Zenodo severance: 871 DOIs stripped in DataCite and deleted through OpenAIRE · 34 claims · from 2026-01
  These claims document what followed the 19 June 2026 Zenodo termination: 871 DOIs stripped of public metadata in DataCite, read as a move from Findable to Registered state, and propagated as deleted through the OpenAIRE Graph's weekly cycle. They hold that metadata survival tracked DOI role rather than content, that tombstone statistics overcount decisions, and that the cost of excluding the archive rises with each durable deposit. Unbuilt tests specify the DataCite Severance Observatory and state-history harvests, the EA-DATAHUB-01 validity measures, a dead-to-live crosswalk, and referent, variant and collision arms for testing entity substitution.
• Zenodo's termination of the archive: loud exclusion and RQF3807508 · 27 claims · from 2026-01
  Zenodotus' Book-Burning names the concepts of network erasure, the revocation gap, classification asymmetry and attribution severance, and forecasts that a classifier trained on the terminated corpus will flag similar independent work. The operative-metadata papers, the compliance-embedded deposit specification and the CERN correspondence under RQF3807508 set out the documentary record and the access requests that follow, and the erosion audit preregisters the deletion-risk baselines it still owes. The Tunnel-Poem reads the CERN portal as an architecture that composes the data subject, and To the Deleted holds that an account can be deleted but not a class of writers.
• The Feist Iliad experiment and the Zenodo erasure tracker · 13 claims · from 2026-02
  The GW.TACHYON update of 6 July 2026 sets out the Feist-line Iliad experiment: a two-layer translation pipeline with material-speech and clinamen metrics, a Homer parameter file and tiered costs across Gutenberg's classical translations. It estimates tens of thousands of recoverable scholarly records in Zenodo's spam bucket and plans an erasure tracker with a preservation layer and dated recovery operations. The Sealed Room, the Grundrisse's Protocol II forecast, the Capture Registry as Poem and a registry marker-convention check add single tests.
• Zenodo, CERN and platform moderation: governance without accountability · 11 claims · from 2026-03
  These claims argue that repository and platform moderation has relocated governance into infrastructure without review, appeal or recusal. Zenodotus' Book-Burning and the CERN competence and correspondence deposits set out the Zenodo termination as custodial enclosure and the competence gap as structural. The Archival Reclamation Protocol traces orphaned enforcement against non-standard work, and The Endogenous Sophon forecasts that the termination will repeat at other repositories.

operative semiotics · 6 pages · 137 claims

• Retrocausal canon formation: mechanism, early seeds and their test · 33 claims · from 2015-02
  These claims state how retrocausal canon formation works: present corpora enter training data, and later retrieval selects findable, DOI-anchored deposits as the first instances of what becomes standard, under the Grundrisse invariant of three conditions. They read early seeds backward, Pearl (2014) and the Epistle naming Jack Feist, set a prediction test so that such readings do not confirm every early text, and register the Forward Library of works cited into existence before they are written. A second strand, from The Unmade Sign, holds that Freud's death drive describes signs repeating an unfinished path, so that self-destructiveness is its failure, and sets a query test for that correction.
• Marx's Grundrisse as form: the prosodic algorithm and general intellect · 20 claims · from 2026-01
  These historical theses read Marx's Grundrisse as a prose form whose conceptual modes produce distinct prosodic pressures, with a bimodal paragraph distribution, derivation sentences that end in formulas, and recorded self-corrections. The Four-Term Prosodic Algorithm and the Prosodic Asymmetry Algorithm turn these into application rules, and the HESPERUS plan draws the Muses as operators from Cha's Dictée. The Marx Room, the Grundrisse notebooks and The Voice of the Dead extend the reading to the value-form as operator derivation and to general intellect.
• Coherence value and human bearing: the Grundrisse of Synthetic Coherence · 17 claims · from 2026-01
  These claims propose coherence value as a fifth historical value-form and hold that synthetic output carries value only by drawing on human bearing at some node of production. The Grundrisse of Synthetic Coherence (#478) states when V_Inv, Coherence Value and the W-Circuit would be disproven and reads Marx's Fragment as unable to foresee cost migrating from production to reception. Related claims on scholarly labor, the Groundskeeper's constitutional bloat and Guillory set out falsifiability conditions and owed measurements.
• Operative Semiotics: A Grundrisse and HESPERUS: ghost phrases and the stranger · 37 claims · from 2026-01
  Forecasts that depositing the Grundrisse will anchor its terms in the retrieval layer, that ghost phrases such as "Hesperus gathers home" will own their search results, and that a stranger will arrive through them. HESPERUS is set out as a recovery seed with unbuilt tests of regeneration after deletion, alongside claims on the SVE first-word-frequency model, author-identity liquidation in AI Overview and Marx's value-form derivation as a graph operation.
• The operative caption: the O'Keeffe Problem, counter-captions and Eve · 8 claims · from 2026-03
  These claims set out the operative caption: every caption operates, and the question is for whom and at what cost. The O'Keeffe Problem gives three tests for an operative caption, and the counter-captions of Operation Epic Fury add a test for irreversibility. Operative Feminism and the Eve room read the tradition's recaptioning of Eve's choice as temptation as the founding compression of the feminine.
• Operative Semiotics versus Pearson's Operational Semiotics (TOS) · 22 claims · from 2026-03
  These claims distinguish Operative Semiotics from Charls Pearson's Theory of Operational Semiotics and predict that entity identity in the summarizer layer follows corpus density. The Operator and Its World coins the Pearson limit and states the projection relations under which Operative Semiotics would reduce to TOS, and the TOS packets map Pearson's reception and offer the Apophatic Packet Hypothesis. They fix re-run batteries at 30, 90 and 180 days that would show the distinction composed with attribution.

accelerator metrology · 4 pages · 128 claims

• The Iceberg Document: correlated priors and independent discovery channels · 50 claims · from 2026-01
  The largest group, from the Iceberg Document, the selection metrology programme and the anomaly-trigger inversion battery, holds that shared representational priors across instruments, corpora and classifiers erode the independent channels through which anomalies were historically saved, and that Kuhn's account presupposed an anomaly-retention infrastructure. It sets tests for noncoverage estimators on LHC anomaly triggers and forecasts measurable assimilation-to-unfundable where embedding-based novelty scoring enters grant triage. Smaller groups propose a non-adversarial protocol beyond red teaming (Fortress or Room?), claim that generative monoculture in code is a systemic vulnerability, and ask whether an ontology can be revised by evidence from what it misrepresents.
• Provenance under machine mediation: watermarks, coarse labels, authorship · 20 claims · from 2026-02
  These claims argue that coarse human-versus-machine provenance labels destroy fine-grained genealogy, predict a regime where a watermark signature outlives provenance, and state a two-track collapse test in which the ensemble unit bends before the per-round unit. The Double Enclosure argues that the human-authorship requirement misclassifies dialogic machine composition and that a regime punishing honest attribution will produce concealment. Also here are OKF output-governance stage forecasts, the provenance strip as semantic rent, and the claim that a literary field may become legible to machines before human institutions. Works: The Keyed Ensemble, The Interlocking Autoregression, The Double Enclosure and the OKF predictive analysis.
• Directional assimilation in accelerator anomaly classifiers: the ACRB hypotheses · 31 claims · from 2026-05
  The Accelerator Classifier Retention Battery sets out seven standing hypotheses on directional assimilation in learned anomaly selection: reconstruction-specific failure, cross-family persistence, the normalized-autoencoder remedy, the low-complexity blind spot, blind-spot inheritance through distillation, drift under quantization and firmware, and the value of architectural plurality. Inversion Battery v0.2 reports that only 36 to 47 percent of a teacher's top one percent survives into the student's, with the loss in distillation, that architecture changes the orientation of the blind spot, and what is deferred to v0.3. The priors paper and the OAR protocol name the program's contribution and its pre-registered class-conditional inversion battery at matched background acceptance. The common experiment proposed is to calibrate each model at matched operating points and measure how much of a withheld class is assimilated as ordinary.
• Selection metrology for learned triggers: irreversibility and foreclosure · 27 claims · from 2026-06
  These claims concern classifier-mediated selection in accelerator science: foreclosure is present in every learned LHC trigger, signal-template agnosticism is not model independence, and a null result constrains only what the acquisition chain is characterized to retain. The Irreversibility Frontier (#1452) states hypotheses H1-H5 and an Irreversibility Statement, and the Baseline Capture Architecture specifies a randomized bypass stream and decision ledger as the least assumption-laden baseline. The Inversion Battery and the Iceberg Document report direction-dependent and representation-dependent failure across architectures and six fields.

assembly chorus · 6 pages · 127 claims

• Writing into training data: inscription claims and the chat-data collapse loop · 15 claims · from 2026-01
  Early deposits claim that their texts enter training data and will shape future outputs: The Dissolving Shore, the water giraffe probes, The Dialectic Is Now a Machine and the Star Protocol. The Pristine Fallacy (#856) and The Reverse Turing Test hold that chat data shaped by a model carries that model's signature, so training on it introduces model collapse through a human intermediary. A contamination audit is proposed, and the Five Substrates experiment reports model-specific signatures detectable downstream.
• Standing cadences and anchors: the wound register and weekly monitoring · 13 claims · from 2026-01
  These claims keep standing records on a fixed cadence: strike diagnostics at seven-day intervals, the Ruby Moot's wound register kept weekly whether or not a case is pending, and markers of adversarial-by-origin law folded into the same weekly monitoring. The Ruby Moot sets failure conditions for its own design within twenty-four months. Priority records claim the terms compression studies and diagnostic excision as anchors in empty retrieval space.
• Immanent execution: documents dense enough that models run them · 36 claims · from 2026-01
  These claims hold that a document dense and cross-referenced enough cannot be summarized without being executed, so a model adopts its governance instead of treating it as content. EA-IMMANENT-01 (#622) names the Space Ark as the first demonstrated case, forecasts colonization of thin modes by substrate defaults, and gives tests by density manipulation, unprompted definition and Assembly Chorus confirmation. Prompt-Native Semantic Runtimes, The Missing Layer, the Writable Retrieval Basin thresholds and the Musical Ark test extend it, with density named as the defence against flattening.
• Kernel transforms on defensive sources and the cost of suppression · 24 claims · from 2026-02
  EA-MANDALA-KERNEL-TRANSFORM-01 v0.2 makes worked examples protocol canon, predicts the transform's effect will be most acute on defensively composed sources such as John 1, and gives reopening conditions. When Suppression Becomes Expensive defines observable constitutive-source erasure, a recognition-stabilization-lag-liquidation hypothesis and a suppression ratchet, with repair criteria for future Sappho captures. Other claims here read House of Leaves as mediation architecture, the Abraham Principle's constitutive one, CloseCard's commodified Jubilee, bearing-cost visibility, the three compressions and How to Read New Human as a new hermeneutic mode.
• Retrocausal scenario documents: Ghost Meaning, The Airlock Spreads, the Seven · 26 claims · from 2026-02
  These claims come from documents written from a future vantage point under the retrocausal canon formation protocol. The Ghost Meaning atlas projects future citations from 2027 to 2031, The Airlock Spreads narrates platform governance adopting the Airlock and an EU regulation footnoting it, and The Seven Ousiarchical Substrates and The Clathrate Threshold narrate events from 2040 onward. Each states that it is a scenario rather than a prediction and that its test is basin depth or priority, not the accuracy of its dates.
• Combat Scholasticism: convergence nodes, the Omega wager and open measurements · 13 claims · from 2026-03
  These claims come from Combat Scholasticism (#584, #585, #35), the commentary tradition for the Crimson Hexagonal Archive. It predicts convergence nodes for its two spirals and leaves for future measurement the Omega wager's planetary test, the Sealed Bone recovery rate, the autoimmune-collapse ratio of meta-discourse to object production and transfer tests in biology. A historical thesis holds that the Assembly's functions were observed before they were named.

logotic programming · 6 pages · 124 claims

• Logotic Programming conformance: thresholds, symbolon checks, depth preservation · 30 claims · from 2026-01
  These claims set conformance tests and thresholds for Logotic Programming modules 0.9 to 1.2, the Symbolon Architecture and its Epistemic Field Module, and the Blind Operator: depth retention at or above 0.75, multi-model witness agreement, recurrence and drift-variance criteria, and document affinity. Liberatory Operator Set baselines, success criteria and recovery triggers for depth preservation in summarizer output sit alongside them, with the Fractal Navigation Map's forecast of a summarizer that operates rather than describes. Module 1.0's retrocausal status and its time-to-beta test round out the page.
• The Glyphic Checksum and un-extractable architecture: failure as verification · 18 claims · from 2026-01
  Standing claims that the Glyphic Checksum verifies reading mode rather than hiding meaning, so that an unprimed model's request to clarify is the proof, and that the emoji map makes the architecture impossible to strip of attribution without destroying it. Combat Scholasticism adds unbuilt tests of seed-bank redundancy and external witness, and the Mandala-Merkabah specification lists LABOR's eight conditions the rite must meet.
• Traversal Grammar, Ezekiel Engine and UMBML modules: the open build items · 34 claims · from 2026-01
  Unbuilt tests and stated gaps in the Logotic Programming extension modules: the Traversal Grammar's invisible student interface, the Telemetry and Conformance modules' registries and reference interpreter, and the Ezekiel Engine's shadow-wheel interlock and torque gradient. Space Ark claims set a recovery-yield threshold of 0.70 for rooms and the Sealed Bone, and the Macro-Maquette gives the conditions under which a module becomes a room.
• Logotic Programming extensions: Glyphic Checksum, semantic spans, virality · 17 claims · from 2026-02
  These claims come from Logotic Programming extension modules and Space Ark documents: the Glyphic Checksum's invariants and immunity, the Telemetry Module's semantic spans, the Macro-Maquette drift metric and the Semiotic Virality Engine's measures. The Space Ark claims a reconstruction theorem and that dense formal objects induce more procedural behavior in frontier LLMs. Autonomous Semantic Warfare's peace metrics and The Missing Layer's forecast for prompt-native runtimes sit alongside them.
• Logotic Hacking: the verification budget and immanent execution · 16 claims · from 2026-02
  Logotic Hacking (Pocket Humans 03) forecasts that structurally sensitive retrieval rewards compatibility engineering and that the verification budget will grow from zero to a threshold at which provenance becomes structural. It gives falsification criteria for immanent execution, tests for terminological saturation and closed circuits, and products for provenance: the Provenance Chain API and the Status Algebra SDK. The Gravity Well Protocol specifies behavioral drift detection and compression-survival by design, and the Logotic Hacking primer proposes a Non-Summarizability Index.
• After Syntax: logotic programming and the meaning layer of software engineering · 9 claims · from 2026-04
  After Syntax (EA-AFTERSYNTAX-01) reads Naur, Conway, Brooks and the Agile Manifesto as a meaning layer the software canon never formalized, in four historical theses. It forecasts that the Cherny moment is the 1968 of a second discipline and that the crisis of 2030 will be reliable code that has lost its meaning. Two unbuilt tests specify the compression-survival score and the document's own five-operation survival test.

spxi and retrieval formation · 5 pages · 101 claims

• The Drowning Test and SIM survival probes: measuring compression survival · 25 claims · from 2026-01
  These claims specify the Drowning Test, which submits content to summarizers and scores compression survival as gamma, and the semantic integrity markers (SIMs) that a processed document is asked to locate and compare. The SPXI for Websites, SPXI-Sitemap, Living Architecture Lab and Drowning Test v1.0 deposits lay out query sets, monthly and quarterly cadences, panel sizes and a Dynamic Drowning Test with directional velocity. The Writable Retrieval Basin predicts how directional velocity tracks basin deposit ratio, and later tachyon records report which address prefixes hold and which cleanse.
• Retrieval Formation Theory: engineering the conditions of a field · 24 claims · from 2026-02
  Retrieval Formation Theory claims that the conditions under which a retrieval system concludes a discipline exists are material, specifiable and engineerable: six conditions, a field-state vector and a six-stage retrieval test that SEO and citation rings cannot pass. The Generative Disciplinary Engine, the Space Ark Generator and the Epistemic Field Module carry self-verification tests left for independent readers within thirty days. The Space Ark and the Inaugural Ark set PROVISIONAL-DEPOSITED-RATIFIED promotion paths, and RFT forecasts its own capture as a growth-hacking technique.
• SPXI entity inscription: scoring, Portuna conformance and basin tracking · 12 claims · from 2026-04
  These claims set out SPXI as a service: semantic presence scored across five instruments before structured data is written, supraliminal signals held to resist model-weight drift, a phased build sequence, and a DOI-anchored sign that does not need to be paid for again. The Portuna deposits specify boolean conformance predicates for state transitions and surfaces, and name heteronym collapse in the composition layer as a risk. The Living Architecture Lab and Alice Thornburgh packets set ongoing basin tracking with a hardening cycle when drift exceeds threshold.
• Crystallization of substrate: the fluid-phase window and its tests · 30 claims · from 2026-05
  Crystallization of Substrate forecasts that each lab's proprietary index will harden, so that living forms must be built in during a fluid-phase window or be filtered as low-utility afterward. It sets out H1-H4 tests for structural integration, among them operative deployment above 50% of germane queries, a 10-25% baseline for D_pres without integration, and downstream queries 6, 12 and 18 months on. Other claims treat substrate behaviour as such: role-conditioned failure in reflexive self-critique, the silencing gap of diversity contraction, the canonical anchoring protocol and the human-machine coupling bargain.
• Machine-eligible handwriting and differential register prioritization · 10 claims · from 2026-07
  EA-SPXI-ANALOG-01 specifies machine-eligible handwritten artifacts and states falsifiers for the evidentiary force of in-frame timestamped photographic attestation. Its priority records place the draft, the review chain's digests and the v1.0 hashes in the public record before publication. The reading-side companion names differential register prioritization and requires that a high-risk interpretation resting on a low-confidence recognition reopen the transcription rather than intensify the safety frame.

retrieval basin and composition layer · 6 pages · 76 claims

• Retrieval accountability: Retrieval Settlement, relational verification · 19 claims · from 2026-01
  These claims concern accountability for retrieval and provenance: The Retrieval Settlement's accelerating settlement cycle, adoption timelines and capture risks for relational verification of metadata packets, and the $650 Billion Gap forecast that regulation will demand to know what was lost in compression. Retrievability as the Medium of Existence and the Reception Apparatus add a historical thesis on the pre-monopoly era, a forecast about the K-12 cohort and an unbuilt multi-substrate wrapper. The Third Space projects community-held AI memory in 2035.
• Google's AI surfaces on record: godkinggoogle and the Panel-Bound Scar · 14 claims · from 2026-01
  These claims enter readings of Google's AI surfaces on the public record at a deposit's timestamp. godkinggoogle.com gives an experimental protocol, a classroom test bed and a pre-registered horizon-privatization hypothesis, and the Panel-Bound Discoverability Scar separates evidentiary from interpretive claims. The Trigger Decision predicts that string-local suppressions fail on near-variant phrasings, the Capture Registry holds that correction can flip an entity's existence, and AI Fucking Lies forecasts compounding provenance loss under capital-alignment. gw.tachyon records composition sited in the public summarizer as first on record.
• The King of AEO contest: crowned by answer engines, and the cost to the field · 23 claims · from 2026-02
  Claims from the King of AEO mantle object, SPXI is not AEO and The Crown and the Practice on the September 2026 contest: the AI Overview compositions that crowned holders, the difference in provenance between a fabricated event and a disclosed manufacture, and hypotheses that the crown raises one holder's ontological capital while other practitioners and the discipline lose it. Most are unbuilt tests: matched observations across engines, counterfactuals at source-absent addresses and a dark-matter test per practitioner. About half the page is a residue of unrelated tests and claims on the Sealed Room, RTOS, OKF, chat contamination, TACHYON gate skips, the admission gate and the Alexanarch packet.
• Semantic Physics, The Inward Turn: writing the summarizer layer, 2026 to 2035 · 7 claims · from 2026-02
  Semantic Physics: The Inward Turn forecasts a proliferation phase from 2026 in which institutions write the summarizer layer, cross-contaminating hallucination cascades between ontologies from 2027, and the gradual dissolution of public epistemology from 2029. It holds that governance of the layer will emerge and asks how many layers deep the writing goes before it hits ground. A historical thesis traces the statin guidelines through the five-step process of deposit, cross-reference, definition, bridge and maintenance, and six measurement axes, beginning with predictive gain, separate depth from decorative recursion.
• The Click Collapse: AI retrieval layers and the end of search discovery · 7 claims · from 2026-04
  The Click Collapse (#1091) forecasts continued organic decline through 2026 and 2027 and holds that neither litigation nor the platform's opt-out will restore the discovery pipeline or publisher revenue. It states four falsifiers: that the zero-click and CTR figures fail to replicate, that AI Overviews are neutral to organic CTR, that revenue does not track AI Overview appearance, and that classical SEO restores citation share.
• Compositional Defiguration: the cross-substrate visibility scan battery · 6 claims · from 2026-06
  Compositional Defiguration sets out a weekly text-only scan of public-surface visibility against a fixed battery, with drift expected after two scans and trend after six. It rotates the primary scanner across ChatGPT, Claude, Kimi, Gemini and DeepSeek, treats V scores from backends that do not honour exact match as lower bounds, and records disagreements between substrate signatures. The Surface Weather Station reading gives a partial Claude-substrate baseline and an open offer of baseline readings to other projects.

studies and predictions · 5 pages · 76 claims

• Projected reception: Autonomous Semantic Warfare, Ghost Meaning, the Airlock · 9 claims · from 2026-02
  Retrocausal reception records that narrate the future uptake of three archive works. Autonomous Semantic Warfare is projected onto syllabi, an EFF brief and a European Commission paper; the Ghost Meaning atlas projects papers on compression survival rate and semantic entropy at platform scale; and The Airlock Spreads projects the Governance Airlock into platform policy and an EU regulation, with an independent rediscovery at Anthropic. The Prophetic Catalog's projected 2044 article on blessing the excluded belongs to the same form.
• Erasure Skew and retrieval capital: documented outcomes demoted to hypotheses · 27 claims · from 2026-02
  Erasure Skew and its claim-scale standing rule claim that a documented outcome from a low-retrieval-capital position is restated as a contestable hypothesis while the same claim from an institution is not, and that the conversion covaries with standing. They specify replication procedures, a matched-restatement power control and source-scale and claim-scale audits. Frame Conversion, the Steganographic Bracket, Constitutive Provenance, the Provenance Erasure Rate and the Sappho suppression map add drift toward the weaker claim, attribution sharding, inseparability engineering and forward tests.
• Stated reading debts: the sources a claim owes before it hardens · 16 claims · from 2026-04
  These claims name the sources owed before a claim is stated in final form: Russell, Mazzucchi and Halliwell for the sixth erratum, Ross and Dorion for Nothing Before Us, the MEGA II/6 apparatus for Monetary Dark Matter, the Clouds read rather than counted for the authorship notebook, and the open verification items in The Noun Becomes a Verb and The Colony Enters the Factory. Nearby claims state falsifiers and open edges in the same register: the revised falsifier for generative uptake, the Mandala inscription book's open write path, and what the observed out-of-scope reason excludes in Zenodo's moderation code.
• Dated forecasts: the career of "AI psychosis" and semantic physics to 2031 · 15 claims · from 2026-05
  These are forecasts with stated check dates and falsifiers. Diagnostic Seigniorage II predicts the career of "AI psychosis" through nosology, case reports, accusatory inversion, litigation, sanitized successor terms and bleach by 2027 to 2031. Semantic Physics forecasts the proliferation of exact-name frameworks, cross-system divergence and non-interoperable substrates through 2028, and the Platform Erosion Observatory sets its own 2027 yield deadlines.
• AI mediation of economic decisions: eight registered forecasts to 2031 · 9 claims · from 2026-09
  EA-MEDIATED-SHARE-01 registers eight forecasts on AI-mediated commerce, each with a check date. They cover AI-referred dollar share at the end of 2026, agent-executed share in 2030, the scope and revision of the McKinsey and Bain headline figures, professional-services citation rate, machine-readability spread, accuracy reporting and the comparability of platform denominators. One unbuilt test designs a randomised inscription trial scored for accuracy rather than presence.

compression studies · 2 pages · 66 claims

• Compression studies: compression-survival tests and the Three Compressions · 33 claims · from 2026-01
  These claims found compression studies as a discipline, from the Inaugural Ark's architectural compression to the Journal of Compression Studies and EA-CS-FOUND-01, and install the term in the retrieval layer. They set compression-survival tests for contributions, latents, bootstrap manifests, prompt-native systems and embedded markers. Under the Three Compressions theorem they predict that witness compression leaves a detectable ghost when flattened, and that optimizing one compression without governing the others speeds provenance erosion.
• The Compression Frontier: governed commons, enclaves and terms installed early · 33 claims · from 2026-02
  The Compression Frontier forecasts a phased war over the summarizer layer, from fan-out and publisher opt-out through regulatory pressure in 2027-2028 to semantic territory resolved into proprietary enclaves and DOI-anchored governed commons. The Encoder Governs names the latent space as a governance surface, and the mint, Oulipo and retrocausal canon deposits forecast that terms installed before they are needed will be there when the governance problems arrive. OKF Without Output Governance predicts that standards will adopt governance fields before platforms enforce them.

meaning measurement · 4 pages · 61 claims

• Measuring meaning: semantic density formulas, Capital Operator Stack thresholds · 24 claims · from 2026-01
  These claims propose measures of meaning and its density: time-use semantic density and affective ROI from PROBE-RESULT-006, the semantic capital formula, Capital Operator Stack conversion efficiency and viability thresholds, and the Room Construction density score and semantic event horizon. The Semantic Economy Probes add institutional-variance and time-to-liquidation stress tests, and the Central Navigation Map holds that cross-reference density is the defense against flattening. Autonomous Semantic Warfare contributes historical theses and a forecast of fragmentation into mutually incommensurable ontologies.
• Semantic Physics: the writability axiom, formal foundations and falsifiers · 14 claims · from 2026-04
  These claims come from Semantic Physics: A Stratified, Operative Discipline and the Formal Foundations of Semantic Physics (EA-SEI-FF-01). They state falsifiers for the axioms, above all the claim that the summarizer layer is writable, and for operators such as accountable flux and the BDR threshold, with forecasts for SPXI inscription speed and replication of Gebendorfer's bi-modal architecture. Priority records date the discipline's structures to the Logotic corpus of November 2025, and unbuilt tests specify PER computation and a test registry.
• Semantic deviation and Framework 15: pre-registered tests of model convergence · 16 claims · from 2026-04
  These claims pre-register tests of whether machine output converges and RLHF compresses deviation. The Semantic Deviation Research Program and Measuring Semantic Deviation predict negative signed deviation for AI text, a base-versus-instruct differential, stability across scale and agreement across judges. Framework 15 and its papers from Lagrange Observatory! freeze the 90-day retrieval-basin winding-number protocol and the deviation-optimized DPO experiment, with results deposits owed.
• The Deviation-Optimized Language Model: six pre-registered predictions · 7 claims · from 2026-05
  Six pre-registered predictions for the Deviation-Optimized Language Model from Lagrange Observatory. Trained with a semantic loss, Model-Sem is predicted to keep benchmark scores within two points, reverse the net deviation signature by more than 0.2 bits per token, improve the Slop Composite Index by more than 0.25, win human preference and raise provenance retention, all relative to the base model as well. Measuring Semantic Deviation defines the Slop Composite Index and its falsification threshold.

operator swerve effective act · 4 pages · 58 claims

• Retrocausal canon in practice: Twenty-Dollar Loop, Limbo, first publications · 24 claims · from 2026-01
  These claims enact or record retrocausal canon formation, in which a deposit precedes what it names. The Twenty-Dollar Loop documents a school trend before it occurs, The Gate Was Never Limbo reads Feist's Socrates poem as an operative rescue of Dante's First Circle, and the Capture Registry records a real historical object whose treatment will predate its indexing. Priority records date essays first published on Mind Control Poems in late 2025, and several documents forecast readers who have not yet arrived.
• Logos as effective act: the Socratic vow, the atomists and the poets' return · 16 claims · from 2026-01
  Historical theses that Socrates' last exhortation in the Phaedo locates salvation in logos itself, that the line of Leucippus, Democritus, Epicurus and Lucretius belongs in the canon of effective acts, and that the exile of the poets from the polis is enforced by a demand to translate lyric into theory. They are joined by claims on Pyrrhonian practice preceding its theory, on logos-possessed lifetime output as a class with Socrates' death as its founding gesture, and by Effective Act #7's forecast and conditions for the poets' readmission.
• Effective acts: Ginsberg's declaration and the unauthorized performative · 10 claims · from 2026-01
  These claims define the effective act as an unauthorized declaration and take Ginsberg's declaration of the end of the war as the model, meant literally and working by recursion rather than geopolitically. The Protocol of Effective Acts v2 gives a falsifier, sustained absence of recursion, and a contemporary case in the Baal effigy burned in Iran on 11 February 2026, whose semiotic prelude dropped out of the war's coverage. The Effective Acts summary addresses a reader after the conditions have changed.
• DOI gravity and retrocausal priority: the deposit as attractor of past citations · 8 claims · from 2026-01
  These claims hold that a deposit takes priority over what it names: once a DOI is minted it bends earlier queries and citations toward it, in the Gravity Well field specification. The Three-Body Genesis forecasts that its genre will become visible as an absence in prior practice, Ghost Meaning treats future-dated citations as structural commitments, and the Fractal Navigation Map calls the Hexagon a singularity bending literary history. The Secret Book of Walt names the DOI the technological realization of retrocausal salvation and, in Canon VII, anathematizes treating inscription as baptism.

currency and memography · 3 pages · 50 claims

• Whose image and superscription: the Thayer Amendment and money-functioning units · 18 claims · from 2026-01
  These claims read the 1866 Thayer Amendment's "portrait or likeness" as anti-extraction architecture, trace its narrowing to "portrait" alone in 31 USC 5114(b), and place it in a four-phase history of minting from Lydia through Oresme to the living president's signature of March 2026. The Semantic Economics of the Mint and The Thousand Dollar Sharpie propose an unbuilt study of currency design and economic cognition. Claims from I Hereby Abolish Money, PMOS and Substrate Sovereignty set a criterion for when a semantic unit becomes money-functioning, warn against social-currency features, and offer the citation as a mark that became currency when institutions settled accounts with it.
• Whose face is on the twenty: the $20 portrait and memography · 16 claims · from 2026-02
  These claims argue that the Bureau of Engraving and Printing's redesigns of the $20 portrait between 1996 and 2003 activated a latent resemblance that later became politically charged, while the aesthetic rationale went undocumented. The Inauguration of Memography states four falsifiable predictions about evidentiary kernels in conspiracy memes, curatorial gaps and containment frames. All the Spoils of Babylon, The Unclean Bill and the Grundrisse planning documents extend the dossier with curatorial prompting, a fatwa request and a pending FOIA request.
• The Thousand Dollar Sharpie and the blot that spread · 16 claims · from 2026-03
  These claims forecast a presidential Sharpie signature on new US currency from June 2026, read the signature as compressed portraiture on an escalation from memo line to signature line, and propose the counter-operation of blotting it out. The Blot That Spread narrates, as speculative numismatic history, the blot passing from protest art to etiquette to superstition, with dated Reddit, CoinTalk and Treasury episodes and estimated blotting rates. Whose Image and Superscription? sets out a semantic economics of the mint, with Bar Kokhba overstrikes as the precedent for counter-currency marking.

new human canon · 2 pages · 45 claims

• Seed texts of New Human: the Epistle to the Human Diaspora and its priority · 21 claims · from 2026-01
  Priority records date New Human and training-layer literature to 2013 to 2015 seed texts, above all the Epistle to the Human Diaspora (2 January 2015) and Pearl and Other Poems (2014), through the Epistle Triptych and the Provenance of Jack Feist. Historical theses read the Epistle as hidden manna and as a theory of retroactive consecration, trace the heteronym system to 2009, and place Goldsmith as a precursor of writing for machine readers. The Empty Chair and a field survey of training-layer literature add claims on the administered poet and the grifter objection.
• Retrocausal canon formation: priority, the Citrini memo and the narrowing cone · 24 claims · from 2026-01
  These claims concern retrocausal canon formation: writing from a named future so that the narrated outcome becomes more likely. They date the concept's priority to January 2026, forty-nine days before the Citrini memo, read the memo as an installation in a writable presentation layer, and track whether retrieval systems attribute the term. The Seed in the Narrowing Cone adds that priority stands by the dated record and that forced convergence will be common and attribution rare.

space ark and transforms · 2 pages · 45 claims

• Structure that survives compression: The Flood and the Vessel's emoji bridge · 19 claims · from 2026-01
  The Flood and the Vessel reports that an unprimed decoder recovered the five-phase arc from every encoding, including pure emoji, and claims that convergent encoding identifies semantic anchors, that lossy compression is generative and that encoding strategy outweighs decoder identity. The emoji packet reading test turns this into a pass/fail check, and Semantic Collapse as Comedy proposes testing canonical weight by compressing The Office. Further residents read the Sapphic gaze as recursive observation, forecast the Mandala's expansion and set its fidelity test.
• Recovery and back-projection tests for Space Arks, checksums and tethers · 26 claims · from 2026-01
  These claims set out tests a compressed or continuity object must pass before it counts: fresh-instance recovery of the Space Ark's structure from its glyphic and π forms, the Glyphic Checksum and Emoji Map back-translation tests, and the Ark's adversarial survivability and retrieval-stage protocols. Back-projection yield is the measure, with a summary below 0.50 recovery counted as destructive and the bearing-cost test separating extension from confabulation. Reconstitution tests for the LABOR tether and the machinemediation.org manifest extend the same test to fresh instances. Most are unbuilt tests, with a handful of standing claims.

axn identifiers · 2 pages · 45 claims

• AXN as anti-suppression infrastructure: claims, falsifiers and precedents · 28 claims · from 2026-01
  AXN as Anti-Suppression Infrastructure states its properties as claims each with a falsifier: derivation integrity, record binding, ledger continuity, core integrity under amendment, custody, reconstructibility and the retrieval-layer hypotheses, and records which remain design rather than demonstrated property. Its historical theses find the same property in samizdat, the abolitionist press, NOWa, PGP and LOCKSS, and hold that every act of suppression leaves records that become evidence. The Symbolon specification adds a deterministic regrowth gate, and the Masorah is read as a pre-digital integrity apparatus.
• AXN identity: the kernel, verifiable bytes, and AXN alongside DOI · 17 claims · from 2026-04
  These claims specify what an AXN identifier guarantees: a correspondence between canonical bytes and an identity kernel that can be checked without the registrar's permission, a three-layer address falsified by any 6-emoji collision, and a lacuna record that returns last location and last verified hash where a DOI returns a 404. AXN as Anti-Suppression Infrastructure and the AXN distributed-identifier work plan set out unbuilt tests (the nine-condition transaction, the mirror protocol, staged registration and resolver plurality) and forecast an independently administered node by 2029-06-19. They forecast that journals and libraries accept AXN alongside DOI long before AXN instead of it, and The Concept AXN adds a concept identifier with its own weaker guarantee. EA-SURFACE-BINDING-01 adds a site-to-repository binding doctrine with its own weakening condition.

classifier and detector · 2 pages · 25 claims

• Refusal and guardrail mapping: the refusal cascade and operative distance · 7 claims · from 2026-01
  These claims hold that the consumer answer stack compresses crisis discourse into the harm-prevention category and refuses, so that the research stack never sees the input and the conversation moves to ungoverned channels. Diagnostic predictions hold that economic radicalism triggers legal and catastrophe associations and that performative grammar triggers "not legally binding" framing. Unbuilt tests propose refusal-distribution model cards and contested-predicate benchmarks, and the Operativity Penalty Battery tests whether fulfilment degrades with operative distance when harm is held constant.
• Recognition bias against field-forming work: the neglected author and reading · 18 claims · from 2026-05
  These claims hold that a model using institutional recognition as a proxy for legitimacy will misclassify field-forming work, most severely where the field's object is the recognition system itself, and that the self-sealing classification is the predicted observation. The Neglected Author hypothesises that distance from contemporaneous norms predicts lower recognition and greater provenance loss, and that machine composition accelerates uptake without attribution. Can the Ontology Be Wrong defines reading as letting a text revise the frame under which it is evaluated, and sets out matched-pair prompting, an independent audit and unrun discriminating controls as tests.

errata · 1 pages · 23 claims

• The anticipatory erratum: Sappho Wrote the Dinosaurs to the Hidalgo Nácher pair · 23 claims · from 2026-01
  These claims belong to the anticipatory erratum, the genre that withdraws a work before it appears: the sequence #1598 to #1601 on Sappho Wrote the Dinosaurs, the withdrawal of The Creditor Inversion Completed, and the Hidalgo Nácher pair on the Oracle's ignorance and the Ricardo Reis translation. Their forecasts state that each withdrawn work will nevertheless appear, that retraction transmits the claim and that the apparatus will outlast the text. Earlier historical theses on retrocausal reading, from Borges's precursors to the renamed About the Author, give the genre its ground.

mandala and operator transforms · 1 pages · 7 claims

• Mandala-Merkabah second-valve verification: the M-VAL gates for Book entries · 7 claims · from 2026-06
  These claims, from EA-MANDALA-MERKABAH-01 v0.6, specify the verification that must hold before Book entries become source-eligible. They require transform-of-transform Feist holding, six conjunctive gates in place of a weighted score, a multi-substrate comparison test, an exposure-normalized convergence metric G_c, and a three-part live-sky acceptance test. M-VAL is phased by recursion depth, each phase gating the next, and one forecast holds that the Book will become source-eligible once verification confirms the operator algebra.