Findings — Cumulative Synthesis

Status: Living synthesis — iteration 1. Read this first; it is the executive summary of the whole wiki. Last updated: 2026-06-10

This page distills the current state of the investigation into six questions: what is solid; where do the frameworks genuinely clash; which deep principles look most fundamental; what candidate unifying directions survive scrutiny; is a universal theory even possible; and what should we work on next. Every claim is tagged per EPISTEMICS.md. Detail and derivations live on the linked pages; this page is the map, not the territory.

Headline (iteration 1). The two pillars of physics — relativistic quantum field theory (the Standard Model) and general relativity — are each confirmed to extraordinary precision and are mutually consistent everywhere we can currently test. Of the 17 catalogued inter-framework "contradictions" (GAPS_AND_CONTRADICTIONS.md), none is a demonstrated logical inconsistency: under adversarial refereeing they resolve into domain-of-validity mismatches, unsolved-but-consistent problems, and conceptual tensions — and they cluster precisely where both theories are extrapolated past their tested regimes (the Planck scale, black-hole interiors, the initial singularity). Independently, several of the strongest results in mathematical physics converge on one wager: that causal / algebraic / entanglement structure is more fundamental than metric geometry (HYPOTHESES.md, UNIFYING_PRINCIPLES.md). That wager is currently confined to the anti-de-Sitter corner and has no distinguishing experimental test — which is exactly why it is not yet physics.

1. What is firmly established (the load-bearing core)

These are the parts any future theory must reproduce, not overturn.

Quantum mechanics works and is weird in a precise way. The Hilbert-space formalism, unitary evolution, and the Born rule (whose form is forced by Gleason's theorem for $\dim\ge 3$ ) are confirmed to high precision. Bell/CHSH inequalities are violated in loophole-free experiments, and no-signaling holds — so nature is non-locally correlated but not superluminally communicating. ESTABLISHED See Quantum Mechanics.
The Standard Model is an anomaly-free chiral gauge theory $SU(3)_c\times SU(2)_L\times U(1)_Y$ ; renormalized perturbation theory predicts e.g. the electron anomalous moment to ~10+ significant figures. ESTABLISHED See Particle Physics, Quantum Field Theory.
General relativity is confirmed as a classical geometric theory across perihelion precession, light bending, Shapiro delay, binary-pulsar decay, direct gravitational-wave detection, and black-hole shadows; below the Planck scale it is a predictive effective field theory. ESTABLISHED See General Relativity.
Thermodynamics, statistical mechanics, and their bridge are secure, now sharpened by exact fluctuation theorems (Jarzynski, Crooks) and the renormalization-group account of universality. Black-hole mechanics maps exactly onto the laws of thermodynamics ( $S_{\rm BH}=k_Bc^3A/4G\hbar$ , $T_H=\hbar c^3/8\pi Gk_BM$ ). ESTABLISHED See Thermodynamics, Statistical Mechanics.
$\Lambda$ CDM is a 6-parameter concordance cosmology fitting the CMB, BBN, large-scale structure, and the ESTABLISHED fact of cosmic acceleration. See Cosmology.
Structural theorems are bedrock: Noether (symmetry $\Rightarrow$ conservation), Wigner's classification (a relativistic particle is a unitary irrep of the Poincaré group), microcausality, unitarity ( $S^\dagger S=\mathbb 1$ ), and Lovelock's rigidity (Einstein + $\Lambda$ is essentially the unique 4D second-order metric theory). ESTABLISHED

The honest takeaway: there is far more established, mutually consistent physics than is usually implied by "GR and QM are incompatible." The incompatibility is real but narrow and located.

2. The sharpest genuine tensions

Catalogued and refereed in GAPS_AND_CONTRADICTIONS.md (IDs GC-n). The single most important finding of the audit: classify before alarming. Highlights, with their refereed kind:

[GC-1] Background-dependence vs background-independence — domain mismatch + OPEN foundational. Minkowski QFT presupposes a fixed metric; GR makes the metric dynamical. The referee downgraded the popular "mutually exclusive in principle" framing: QFT on fixed curved backgrounds is rigorous, and perturbative graviton EFT places a dynamical metric inside QFT below $M_{\rm Pl}$ . The genuine obstruction is a superposed/fluctuating causal structure, which deprives microcausality and the particle/vacuum concepts of their usual definitions. ESTABLISHED contrast / OPEN resolution
[GC-2] Perturbative non-renormalizability of gravity — an ESTABLISHED theorem ( $G$ has mass dimension $-2$ ; two-loop $C^3$ divergence, Goroff–Sagnotti 1985), with an OPEN UV completion. This is not an inconsistency: an EFT with a cutoff is fully consistent below it.
[GC-3] Cosmological-constant problem — OPEN fine-tuning, not a falsified prediction. In strict EFT logic $\Lambda$ is a renormalized free parameter; what "fails" is naturalness. The famous " $10^{120}$ " is a regularization-dependent ratio (an electroweak cutoff gives ~ $10^{60}$ ); but finite QCD/electroweak condensates already overshoot the observed value by 40–60 orders. See CONSTANTS_AND_SCALES.md.
[GC-4] Black-hole information paradox — domain-of-validity tension, not (per referee) a logical inconsistency. Unitarity is now widely favored: AdS/CFT is manifestly unitary, and island/replica-wormhole computations reproduce the unitary Page curve in controlled (largely AdS, large- $c$ ) settings INFERENCE; no first-principles mechanism exists for a realistic 4D black hole OPEN.
[GC-6] Measurement problem — structural incompleteness, not a bare inconsistency: the unitary axiom and the Born/projection rule have disjoint stipulated domains, and contradiction appears only under the contested premise that unitary evolution is universal. Decoherence explains the pointer basis but yields an improper mixture; it does not select an outcome. ESTABLISHED tension / OPEN resolution
[GC-7] Arrow of time — ESTABLISHED structure + OPEN cosmological origin. Time-symmetric microlaws cannot yield monotone entropy increase without a time-asymmetric input (Loschmidt, Zermelo); the residual mystery is why the early universe had such low gravitational entropy, exported to cosmology and quantum gravity.

Net: the "war" between QM and GR is, on inspection, a handful of located domain-mismatches plus deep-but-consistent unsolved problems. That reframing is itself a load-bearing result for the feasibility question (§5).

3. The most promising candidate deep principles

From the ranked synthesis in UNIFYING_PRINCIPLES.md:

Most likely genuinely fundamental: (i) Noether's theorem + representation theory — theorem-level, regime-universal INFERENCE, high; (ii) unitarity / information conservation [INFERENCE, high, conditional on QM being final]; (iii) microcausality / causal order — least negotiable for relativistic QFT, with the sharp conjecture that causal order outlives spatial locality INFERENCE; (iv) the Past Hypothesis — uniquely a boundary condition rather than a law [OPEN-if-irreducible].
Deepest meta-principle: the renormalization group / effective field theory. ESTABLISHED Its lesson may be anti-foundationalist — that "fundamental" and "emergent" are scale-relative, which directly bears on whether a final theory exists (§5).
The central live convergence. Multiple independent results — the holographic area bound, Ryu–Takayanagi, Jacobson's derivation of the Einstein equation as an equation of state, Van Raamsdonk's disentangling argument, and the Type III $_1$ structure of local QFT algebras — point the same way: information/entanglement/entropy may be deeper than geometry. SPECULATIVE overall; rising to [INFERENCE for the RT sub-claim within AdS]. The honest caveat (see §4) is that essentially every rigorous result here lives in the negatively-curved AdS corner, which has the wrong sign of $\Lambda$ for our accelerating universe.

4. Candidate framework directions and their honest status

Nine directions are developed and refereed in HYPOTHESES.md (H0–H8). They are research directions, not results. Strikingly, they converge on one wager — causal/algebraic/entanglement structure precedes metric geometry — approached through four lenses:

Modular / algebraic (H0, H4): take the Type III $_1$ von Neumann algebra and its Tomita–Takesaki modular flow as primitive; the crossed product with an observer/clock yields a Type II trace, an emergent time, and an area-leading generalized entropy; the entanglement first law $\delta S_A=\delta\langle K_A\rangle$ gives the linearized Einstein equation.
Causal order (H1, H3, H6): a partial order is primitive; a causal poset already fixes the conformal metric (Malament/HKM), so only one scalar (scale/volume, or a relational clock) need be supplied; sharpened by causal-set theory and celestial/BMS boundary symmetry.
Amplitudes / positive geometry (H5): amplitudes are the canonical form of a positive geometry (amplituhedron); locality and unitarity become derived boundary properties — established only for planar $\mathcal N=4$ SYM.
Operational / quantum-error-correction / post-quantum (H7, H8): bulk geometry as the continuum limit of a QECC; QM as the rigid limit of an operational information theory.

The most epistemically defensible entry is the negative one. H2 argues that "spacetime from entanglement," judged as a universal theory, currently fails two bars — (1) all rigorous results are AdS/large- $N$ , linearized, wrong- $\Lambda$ -sign; (2) Type III $_1$ algebras have no tensor factorization, so the "qubits sewing space" picture is ill-defined (though not logically inconsistent — modular theory is the rigorous reformulation). It supplies concrete graduation criteria (a de Sitter RT analog; full nonlinear Einstein from entanglement; a dynamically-formed asymptotically-flat black hole). INFERENCE

Shared, possibly fatal obstacles every direction must survive: the Weinberg–Witten theorem (an emergent massless spin-2 needs an emergent extra dimension), the Type III $_1$ obstruction, AdS-confinement of the rigorous results, and the absence of any distinguishing near-term experimental prediction. On the wiki's own anti-crank standard (AGENTS.md §4), a direction with no distinguishing test is not yet physics. Confidence across H0–H8 ranges from very-low to medium; none is asserted as established.

5. Is a single universal theory of physics possible?

The feasibility assessment (refereed) turns on disambiguating three senses of "universal/final":

(A) A consistent unified quantum-gravity framework reconciling GR and QFT — POSSIBLE, high confidence. The clashes of §2 are domain-mismatches and unsolved-but-consistent problems, not proven contradictions; consistency conditions on a quantum theory of gravity are extraordinarily rigid; and AdS/CFT is a working existence proof INFERENCE that some consistent quantum gravity exists. Whether the framework for our universe is found and empirically confirmed is genuinely OPEN — plausibly attainable mathematically, but possibly never decided by experiment given the ~15-orders-of-magnitude "desert" between accessible energies and $M_{\rm Pl}$ .
(B) A final theory — a true bottom to the tower of effective theories — possible but NOT demonstrably so. The Wilsonian "tower with no bottom" reading (reinforced by $\phi^4_4$ triviality) and the emergence programs (geometry from thermodynamics/entanglement) make "no fundamental bottom layer" a serious live alternative; "fundamental" may be partly scale-relative. Lean: a final framework is more plausible than a final substrate. SPECULATIVE
(C) A complete theory that fixes all contingent parameters — the LEAST likely. The string landscape + anthropic selection SPECULATIVE/CONTESTED, the cosmological-constant problem OPEN, and the contingent low-entropy Past Hypothesis OPEN each threaten parameter-fixing completeness even if the underlying framework is unique.

Net assessment. Modestly likely as a consistent unified framework (whose existence in principle is essentially uncontested); substantially less likely as a parameter-fixing, complete-and-final bottom; with empirical confirmation, rather than logical possibility, as the most probable point of permanent failure. We explicitly decline the common overreach that Gödel's incompleteness theorems forbid a final theory — undecidability of a theory's consequences is not incompleteness of its laws. INFERENCE

6. Highest-leverage next questions

Prioritized in ROADMAP.md; registry in OPEN_PROBLEMS.md.

Get one real-universe ( $\Lambda>0$ ) result out of the emergence program — a de Sitter RT/entanglement analog, or nonlinear Einstein from entanglement. This single deliverable would do more than anything else to move the central wager (§3–4) from AdS folklore toward physics. It is H2's graduation criterion.
Black-hole information as the decisive test bed — push the Page curve beyond controlled AdS toy models toward a dynamically-formed, asymptotically-flat black hole.
Operationalize the measurement problem — tighten collapse-model (GRW/CSL) bounds via interferometry/optomechanics; this is the one corner of [GC-6] that is experimentally decidable.
Sharpen the GR↔QFT interface — for each item in GAPS_AND_CONTRADICTIONS.md, state precisely which incompatibility is logical vs domain-of-validity (continue the refereeing already begun).
Distinguishing-prediction audit — for every program in UNIFICATION_LANDSCAPE.md, record the single most plausible near-term distinguishing observation; flag the (many) programs that have none.
Assumption-demotion program — attack the highest-value "fundamental-suspects" in ASSUMPTIONS_LEDGER.md: the smooth spacetime continuum, exact linearity of QM, the Born rule as a separate axiom, locality, and 3+1 dimensions.
State the cosmological-constant problem without folklore — a fully regularization-aware account separating the genuine puzzle from cutoff artifacts.

Iteration 2 update (2026-06-08)

Five refereed deep-research tracks (see notes/) pushed on the highest-leverage questions. Net effect: the central wager is sharper and more constrained, not confirmed. The referee flagged 0/25 new claims unsound; several were corrected in place.

What moved.

The Λ>0 graduation barrier is structural, not temporary. INFERENCE, high There is no de Sitter RT surface, no uncontested unitary dS dual, and — decisively — no dS analog of the AdS entanglement-first-law ⟹ Einstein chain. The genuine positive: the CLPW Type II₁ static-patch algebra gives a derived finite Λ>0 entropy with a maximal-entropy state (= empty dS), a real foothold for H4 — but it is observer-relative, background-fixed entropy bookkeeping, not dynamics. H2's negative meta-conclusion stands and is reinforced. [→ notes; OP-41, OP-42, A-20]
The algebraic/holographic program encodes geometry; it does not yet generate it. INFERENCE, high — conceded by its own architects The modular/crossed-product results (Witten; CLPW; KLS 2024–25, extending Type II to any Killing horizon) are a rigorous reorganization of semiclassical gravity on a fixed background with a Killing/conformal symmetry; "background independence" in Witten 2308.03663 means state-independence on a fixed worldline algebra. Whether any von Neumann-algebraic subregion description survives nonperturbatively is OPEN (Giddings 2505.22708, "algebraic spacetime disruption"). [→ notes; OP-44, OP-46, A-23, H4-R2/R3]
The Born rule is forced in form, open in meaning. ESTABLISHED form / OPEN meaning Gleason (dim≥3) + Busch/CFRS (dim=2) force $p=\mathrm{Tr}(\rho E)$ given non-contextuality of probabilities; but no no-collapse derivation of the quantitative (unequal-amplitude) rule avoids re-importing a Born-equivalent premise. [→ notes; OP-43, A-22]
Exact linearity is plausibly derivative of causality — but untested where it matters. INFERENCE strong / CONTESTED that it is fundamental Generic deterministic nonlinearity permits superluminal signalling (Gisin; Polchinski; Bielińska–Eckstein–Horodecki 2024); stochastic objective collapse (GRW/CSL, Diósi–Penrose) evades the no-go and is empirically open. [→ A-3 refinement]
Tabletop quantum gravity (BMV) is a class-falsifier, not a quantization proof. INFERENCE for the exclusions; loophole status CONTESTED A positive result excludes mean-field semiclassical and classical-channel gravity and constrains stochastic-CQ — establishing non-classicality of the Newtonian sector — but not graviton quantization. The mediation-vs-locality premise is logically stronger than no-signaling and is actively disputed (Di Biagio; Aziz–Howl vs rebuttals, 2025). [→ notes; OP-40, HYP-GIE-DISCRIM]
Entropic gravity, disaggregated. [mixed] Jacobson (1995) is a genuine theorem given the area law + Unruh $T$ ; Padmanabhan a structural rewriting; Sakharov a mechanism (which outputs the area law); Verlinde-2011 a heuristic; Verlinde-2016 dark matter testable but disfavored at cluster/CMB scales. Iteration-1 correction: galaxy weak-lensing (Brouwer 2021) agrees with EG — it does not falsify it. [→ notes; OP-45, OP-47, A-21, HYP-ENTROPIC-GRAVITY-SPLIT]

Does it cohere into one program? — Partial. INFERENCE The synthesis (notes) finds two internally-coherent programs: (A) geometry-from-algebra (de Sitter, algebraic, entropic tracks — one substrate, one method, one decisive deliverable: the dS entanglement first law, OP-41) and (B) is-the-quantum-framework-final-and-does-it-cover-gravity (Born/linearity + BMV — reconstruction theorems, no-go arguments, experiments). They are joined only by thin bridges: no-signaling/causal-order as a shared load-bearing constraint, and BMV non-classicality as a logical prerequisite for the whole emergence program. The tempting "it's all information" unifier is refused — "information" names non-interchangeable objects across the two clusters (von Neumann-algebraic relative entropy vs operational probability/LOCC monotones).

The most defensible posture is therefore a program of demotion-and-derivation: hold the established cores fixed (local QFT Type III₁ for any Λ (undressed, $G_N=0$ ); the area law; no-signaling/causal-order; the Gleason–Busch-forced Born form; unitarity — but see the Iteration 3 update below: these "cores" are heterogeneous and three are conditional, Type III₁ chief among them) and prosecute as suspected-derivative the status of metric geometry, the local-subsystem picture, exact linearity, and the Born rule's axiom-status — each to a decisive yes/no test. No track has yet graduated from encoding to generating geometry, and Programs A and B do not (yet) merge. That is the honest iteration-2 verdict: a sharp, coherent strategy with a single highest-leverage target (a Λ>0 entanglement first law), not a unification.

Iteration 3 update (2026-06-08)

Iteration 3 shifted from surveying to attempting (three attemptive tracks + a red-team of our own posture; see notes/ iter3-*). Net: the coherence verdict is unchanged at PARTIAL — and iteration 3 confirms that is correct. Progress was real but almost entirely negative/structural: obstructions got sharper and the posture got more honestly hedged, while the central wager got no closer to confirmation or to a distinguishing test.

Did Program A move? Sharpened, not advanced. INFERENCE, high The de Sitter first-law attempt (notes, OP-41) converted iteration-2's bare "no dS first-law chain exists" into a located obstruction: AdS's FGHMVR derivation inverts a continuum of boundary-anchored regions to a pointwise $\delta G_{\mu\nu}$ , but the dS static patch supplies one region and one Killing vector, so $\delta S_{\rm gen}=\delta\langle K\rangle$ yields only the single $\xi$ -integrated Hamiltonian constraint (= the dS generalized second law), not the tensor equation; and the variational principle is sign-reversed — empty dS maximizes $S_{\rm gen}$ (Banihashemi–Jacobson–Svesko–Visser, $dE=-T\,dS$ ), structurally opposite to AdS RT minimization. Separately, Jacobson 2015 is already a local entanglement-equilibrium derivation of the full Einstein equation valid at any Λ — but non-holographic and presupposing the area-entropy coefficient and a fixed-volume maximal-entanglement hypothesis, so it answers a different question. One un-foreclosed escape remains (an $SO(d+1,1)$ family of static-patch modular Hamiltonians; Chen–Xu 2511.00622); the obstruction is tagged OPEN, not a no-go.

Encode vs generate, made rigorous (notes, OP-46). INFERENCE, high A decision-usable criterion (two graded axes — geometric-input $\Delta_{\rm geo}$ and post-processing $\kappa_\Phi$ — plus a five-item "smuggling" checklist) grades six emergence constructions. Finding: no surveyed construction generates Lorentzian spacetime from factorization-free, region-free, symmetric-state-free data; the encode/generate frontier coincides exactly with a hand-chosen tensor factorization or graph — i.e. with the Type III₁ no-factorization fact in disguise. Cao–Carroll "space from Hilbert space" comes closest but outputs only a Riemannian spatial metric after positing a factorization.

The A+B merge: narrow success, deep failure (notes). INFERENCE, medium On a shared von Neumann algebra with faithful normal state $(\mathcal M,\omega)$ one can non-circularly co-derive Pillar B's Born form (lifted to infinite dimensions by the 2026 causal-rigidity theorem, arXiv:2602.09056) + linearity and Pillar A's KMS/thermal time + crossed-product Type II entropy — with the "observer" demoted to an internal quantum reference frame, genuinely dissolving the agent/no-agent clash here. But the GPT axioms that fix the Hilbert-space framework (tensor composition, purification, continuous reversibility) have no Type III₁ formulation and fail in infinite dimensions (candidate-no-go OP-48) — and fail precisely in the closed-universe, no-external-observer regime. So iteration-2's "thin bridge" upgrades to a substantive partial merge at one sublayer, a located no-go at the deeper one — not the "it's all information" fusion.

Red-team of our own posture — four forced weakenings (notes; 0 fatal errors):

The "fixed cores" are heterogeneous, and three are conditional. Most importantly, local QFT Type III₁ "for any Λ" is a fixed core only on a fixed background; once gravity is dynamical it is OPEN/CONTESTED (gravitational dressing alters the algebra at leading order in $G_N$ — Giddings 2505.22708 / 2510.24833). This corrects the iteration-2 framing above, which listed it flatly as fixed. The area law is an input only for the Jacobson–Padmanabhan–Verlinde lineage (it is output by Sakharov, derived in AdS by RT); no-signaling is shared but does not single out QM (PR-boxes); unitarity is conditional on standard QM and is deliberately violated by GRW/CSL.
"Demotion-and-derivation" is a research heuristic, not a falsifiable program — its falsifiable content lives entirely in the object-level deliverables (OP-41, OP-44, the BMV $m^2$ -vs- $m^3$ test, collapse bounds) and in the bracketed SPECULATIVE wager it quarantines.
The six-lens "convergence" is partly a streetlight effect — the lenses agree partly because they share load-bearing inputs (Killing-symmetric vacua, AdS/large- $N$ , the unproven causal-set Hauptvermutung), which is weaker evidence than genuine independent convergence.
CLPW Type II₁ is a feature of the static patch's bounded area, not a Λ-diagnostic — Schwarzschild–de Sitter (Λ>0) is II∞×II∞ (KLS), so the II₁-vs-II∞ distinction tracks the horizon/observer-Hamiltonian spectrum, not the sign of Λ (refines the iteration-2 H2-R1 correction further).

What survived every attack and was strengthened: the deflationary core (the catalogued GR/QM "contradictions" are not logical inconsistencies); the central diagnosis "the algebraic/holographic program encodes geometry, it does not generate it" (reinforced by the verified Giddings results); the Born form/meaning split; linearity-as-derivative-of-no-signaling; the refusal of the "it's all information" fusion; the dS negative meta-result; and BMV-as-class-falsifier-not-quantization-proof.

Empirical falsifiability portfolio (the honest bottom line). INFERENCE, high Every live channel tests the program's logical gate (is gravity non-classical? — BMV/QGEM, ~10–20+ yr), the status of its cores (is linearity/Lorentz exact? — objective-collapse bounds decisive ~2035; LIV/EP null anchors), or its correct target (is the universe really eternal dS? — DESI/Euclid $w(z)$ , 2–7 yr; currently a 2.8–4.2σ dynamical-dark-energy hint). None tests the geometry-from-algebra wager itself. Matter-wave / levitated-optomechanics mass scale is the shared enabling technology gating both live channels. By the wiki's own anti-crank standard, the wager is therefore still not yet physics — a sharp, coherent, well-hedged research strategy, not a result.

Iteration 4 update (2026-06-08)

Three attack tracks + a web-verified literature refresh + a convergence audit (see notes/ iter4-*) closed the last open escape route from iteration 3 and confirmed convergence. The standing verdict is unchanged for the third consecutive iteration — PARTIAL coherence; geometry-from-algebra encodes but does not generate geometry; the wager is not-yet-physics — but the central diagnosis now rests on a much harder argument.

The dS first-law escape was executed and failed — a scoped near-no-go. INFERENCE, high Iteration 3 left one un-foreclosed route to a Λ>0 Einstein-from-entanglement derivation: enlarge the single static-patch boost to the full $SO(d+1,1)$ family of observer/static-patch modular Hamiltonians. Executed, it fails on one interlocking obstruction whose decisive component is a dimension count: the isometry orbit is finite-dimensional ( $\dim=2d$ ), supplying at most $\dim SO(d+1,1)=10$ global Killing-charge constraints, which cannot Radon-invert to the infinite-dimensional local field $\delta G_{\mu\nu}(x)$ — it lacks the "radius modulus" / continuum of anchored regions that AdS's FGHMVR inverts; compounded by the family being geometric in a common state only at Bunch–Davies (where the probe vanishes) and a second-order, sign-definite ( $\delta^2 S_{\rm gen}<0$ ) concave-GSL character. The published execution of exactly this route (Chen–Xu, arXiv:2511.00622) lands precisely where the count predicts: entropy-matching and a second-order quantum first law, no $\delta G_{\mu\nu}$ . This promotes HYP-dS-NOMIN toward a near-no-go (HYPOTHESES.md HYP-dS-NOMIN-R3) and yields a reusable principle (HYP-dS-CARDINALITY): Einstein-from-first-law needs an infinite-dimensional local region family probing every bulk point; the finite isometry orbit of a maximally-symmetric background gives only global charges. Two routes survive: the non-unitary dS/CFT pseudo-entropy route, and Jacobson's small-ball continuum realized inside a single CLPW Type II₁ algebra. [→ notes]

OP-48 settled as a sharp disjunction. ESTABLISHED/INFERENCE Can the GPT framework-fixing axioms live on a Type III₁ algebra? (a) Hard no-go for exact local tomography / tensor composition on a single Type III₁ factor; (b) but the split property (nested $\mathcal O_1\subset\mathcal O_2$ , nonzero collar) supplies an intermediate Type I factor with a genuine factorization and local density matrices — defeating the would-be hard no-go; (c) yet this is not a clean reconstruction: the canonical intermediate factor is singled out only by the vacuum's modular conjugation $J$ , relocating the Bisognano–Wichmann/CHM geometric-reference-state circularity to the factorization stage (HYP-SPLIT-GEOM), and it fails for adjacent (zero-collar) regions. A full A+B merge still requires a state-/collar-free composition surrogate — OPEN. [→ notes]

The encode→generate frontier is near-vacuous under the standard readout. INFERENCE The OP-46 clause-(iii) escape (recover geometry from a non-symmetric modular flow) is squeezed by two chained, web-verified theorems: Borchers/Wiesbrock (half-sidedness ⟺ positivity — the emergent translation group comes from a relocated positivity premise, not neutral data) and Caminiti et al. (arXiv:2502.02633: recovered boost generators are conformal Killing vectors, presupposing a symmetric background). Connes' reconstruction, moreover, yields only a Riemannian manifold, not a Lorentzian one (GAP-NCG-LORENTZIAN-SIGNATURE). The genuinely-live, unrealized target is a non-CKV readout (modular-Berry / Connes spectral distance) yielding a Lorentzian metric. [→ notes]

Literature refresh (web-verified). ESTABLISHED (i) Muon $g-2$ is now SM-consistent: the 2025 Theory-Initiative white paper (arXiv:2505.21476) adopts a lattice-QCD HVP average, giving $a_\mu^{\rm SM}=116\,592\,033(62)\times10^{-11}$ and $\Delta a_\mu=38(63)\times10^{-11}$ — no tension (a residual internal lattice-vs- $R$ -ratio / CMD-3 HVP tension persists). The historic ~4–5σ anomaly is largely resolved; OP-21 and particle-physics.md are updated. Peripheral to the central wager. (ii) The three previously-[unverified] 2026 preprints are all real / web-confirmed (no fabrication): 2603.06211 (Zhang), 2601.18856 (Pronskikh), 2601.13012 (Pati — fixes the inner product, not linearity). (iii) Collapse-model bounds tightened but not closed; first Lorentzian-signature asymptotic-safety fixed-point evidence (incremental, CONTESTED); DESI still DR2.

Convergence verdict — how far to a conclusion. INFERENCE, high The audit (notes) finds the new-insight trend diminishing (three iterations at the same verdict; marginal yield now a sharper obstruction or a self-correction, not new positive structure) and distinguishes two distances:

(A) Analytical saturation — close (~1–3 more focused iterations). "Is a universal theory possible?" is answered in disambiguated form (senses A/B/C); the central wager is named; why it's not-yet-physics has survived every attack. What remains is a finite, named list: OP-48 point (c) (collar-free composition surrogate); OP-44/OP-CP1 (nonperturbative crossed-product fate); the OP-46 non-CKV Lorentzian readout (the only move that would flip the headline verdict).
(B) Object-level confirmed theory — not forecastable. Gated not just by the ~15-order-of-magnitude empirical desert but, more bindingly, by the fact that the wager has no distinguishing experimental test even in principle — every live channel tests the program's gate, cores, or target, none the wager itself.
Recommendation: HYBRID — attack the three named decidable problems over the final ~1–3 iterations, do the cheap documentation passes (done), and stand up lightweight tracking of the four empirical channels (BMV; collapse bounds ~2035; DESI/Euclid $w(z)$ ; LIV/EP), since those — not further analysis — gate the object-level verdict.

This is the honest standing conclusion: a rigorous, stable map of exactly why a universal theory is open and what would close it — converging on analytical saturation, with the object-level question handed to experiment.

Iteration 5 update — analytical saturation reached (2026-06-08)

The closing analytical push (four web-verified attack/route tracks + synthesis; see notes/ iter5-*) executed the one move iteration 4 designated as the only verdict-flipper — and it did not flip. Two of three named decidable problems closed as no-gos/bounds, the third sharpened to a located boundary; the program reached the analytical-saturation condition. Fourth consecutive iteration at PARTIAL / encodes-not-generates / not-yet-physics. The standing verdict is now consolidated in the capstone CONCLUSION.md.

The verdict-flipper was executed and did not flip. INFERENCE, high The OP-46 non-CKV Lorentzian readout (modular-Berry / kinematic-space) produced a genuine first — the first emergent Lorentzian ( $\mathrm{dS}_2$ ) metric in the survey, crossing the signature barrier the Riemannian Cao–Carroll and Connes-spectral routes fail (Huang–Ma, arXiv:2003.12252) — but its Lorentzian signature is inherited from the symmetric CFT vacuum, not generated from non-symmetric data. "Encodes-not-generates" is now readout-independent: it closes at the same signature-installation step across all three known readouts (stress-tensor-flux/Sorce–CKV, modular-Berry/zero-mode, Lorentzian-NCG/η-foliation-twist). [→ notes]
OP-48(c) settled as a clean conditional no-go. ESTABLISHED/INFERENCE On a single Type III₁ factor, "collar-free + state-independent" composition and "singles out QM" are mutually exclusive: the genuinely collar-free surrogate (Haagerup standard form + Connes fusion) is type-agnostic and a universal embezzler ( $\kappa_{\max}=0$ , van Luijk et al.); any QM-selecting axiom forces a Type I factor (the split construction), re-importing the collar + reference state. The QM-selecting composition axiom coincides with the geometry-importing input. [→ notes]
OP-44/OP-CP1 sharpened to a regime boundary. INFERENCE, high The crossed-product Type II structure survives as an approximate, QRF-relational leading-order truncation below $N_\ast\sim R/2GE_f$ ; at/above it, Giddings' algebraic spacetime disruption degenerates the inclusion structure. So "Type III₁ for any Λ" is a robust kinematic/leading-order input but not a fixed nonperturbative core — with no theorem on either side (the sole residual formally-open lever). [→ notes]
Both surviving dS routes fail. INFERENCE, high The dS/CFT pseudo-entropy first law evades the cardinality facet but inherits non-unitarity (complex central charge $c=i\,3R/2G$ ); Jacobson-in-CLPW merely re-imports $\eta=1/4G$ (the CLPW II₁ trace encodes one area datum, not a local continuum). Unifying root (HYP-dS-CARDINALITY-R1): the area-entropy coefficient is an INPUT, not an output, in every known route. New decidable sub-problem: OP-dS-PSEUDO-REALITY. [→ notes]

Saturation declared. INFERENCE, high Iteration 5 produced zero new positive structure and zero verdict movement — only sharpening, closure, and one symmetry-bound positive. Per the iteration-4 criterion the analytical program is saturated. The standing conclusion, its two explicit hedges (OP-46 residual MEDIUM; OP-48c horn-C INFERENCE), and the residual watch-list now live in CONCLUSION.md; the empirical channels gating the object-level question are tracked in EXPERIMENT_WATCHLIST.md. The object-level question remains not forecastable — the wager has no distinguishing test even in principle.

Iteration 6 update (2026-06-10) — new-input iteration

The first post-saturation iteration: not further reasoning over the saturated map, but a deliberate injection of new input — 6 web-verified research tracks + 4 adversarial attack tracks (22 agents, all refereed; ~90 findings kept, zero fatal; see notes/ iter6-*). The standing verdict is UNCHANGED at the headline level for the fifth consecutive iteration — PARTIAL coherence; Program A encodes-not-generates; the wager not-yet-physics; object-level not forecastable — but the hedge structure underneath it was genuinely restructured: one hedge shrank, one residual lever closed negative, one new named hedge was added, and the central obstruction broadened from three readout families to four.

The no-test claim is now a corollary, not an observation (ENCODING-SCREEN) — and a third hedge is added. INFERENCE A direct, literature-wide attempt to overturn CONCLUSION §7's sharpest sentence ("no distinguishing experimental test even in principle") failed and strengthened it: if Program A is encode-only in the OP-46 sense, its observational content equals that of the EFT whose background it consumed, so every candidate signal (VZ/GQuEST geontropic noise, DSW horizon decoherence, everpresent-Λ w(z), trapped-ion Bisognano–Wichmann) factors through physics equally available to geometry-first descriptions. A class-separating test would require a "reverse Weinberg–Witten" theorem ("observable X is realizable only without fundamental metric degrees of freedom") — no theorem of that shape exists. The §8 analytical reopening condition and the missing experiment are thereby the same obstruction (one-way: a generative bridge is a prerequisite for, not a guarantee of, a class-separating observable — the referee weakened the Corollary's "iff" to "only if"). Because the screen inherits the OP-46 encode-only classification, the no-test claim now carries the same MEDIUM hedge — recorded as HYP-ENCODING-SCREEN, the wiki's third named hedge. [→ notes]
The obstruction broadens to a FOURTH readout family, and a new smuggling pattern is named. INFERENCE, high Causal fermion systems — the most serious candidate previously absent from the wiki — install the Lorentzian datum at their first axiom: the $(\le n,\le n)$ -eigenvalue condition on the operator class is a necessary carrier of the signature (an elementary spectral lemma, derived and numerically re-checked by finder and referee, shows the lightlike/null-cone relation is empty for positive-semidefinite operators) — the Krein η relocated into the definition; all worked examples are Dirac-sea encodes on a pre-given Minkowski background, with minimizer existence proven only for finite-dimensional $\mathcal H$ . Separately, a genuinely new selection-type pattern surfaced (Sung-Sik Lee's matrix model, JHEP 06 (2020) 070, arXiv:1912.12291; a 2025 classical signature-potential toy model): signature appears as a state-dependent dynamical output, but on a posited carrier with a hand-installed signature-valued constraint-algebra template — these generate only the choice, not the possibility space. Result: HYP-CKV-VACUITY-R3 — every known Lorentzian-output route is installation-type (symmetric state; indefinite-pairing axiom) or selection-type (signature-valued template); the smuggle list gains two named items, and the OP-46 open target compresses to its minimal form: derive the indefinite pairing (Krein η / (n,n) operator class) from modular data. CONCLUSION §8's "none exists or has been sketched" gets a precision repair (Lee 2020 is the nearest sketch; it fails the full spec). [→ notes]
OP-44's interior is restructured — still no theorem at $N_\ast$ , but both sides hardened and a fourth option appeared. ESTABLISHED (cited theorems) / INFERENCE (bearing) Jensen–Raju–Speranza (arXiv:2412.21185) prove the AdS boundary time-band algebra strictly has no commutant nonperturbatively (a theorem-grade boundary-anchored instance of the failure side) and a rigidity theorem (the modular crossed product is the only crossed product of a III₁ algebra with a trace — the survival side has a unique landing spot within the crossed-product class). De Vuyst–Eccles–Höhn–Kirklin (arXiv:2411.19931) locate the order- $G_N^2$ proof obligation at the quadratic (boost) linearization-instability constraints — cleanest for spatially closed (dS-like) universes. KLS II (arXiv:2601.07910) and Chandrasekaran–Flanagan (arXiv:2601.07915) make the survival side theorem-grade at linearized order for the complete constraint family, including a GSL for non-stationary perturbations and a proof of the quantum focusing conjecture in the perturbative regime. Kudler-Flam–Witten (arXiv:2510.06376) add a published fourth verdict option: the nonperturbative algebra may exist only after averaging over $N$ . Herderschee–Kudler-Flam (arXiv:2510.01556) realize the iter-5 "split-property downside" — but by stringy (Hagedorn) physics at sub-string-length separations, not by $O(\kappa)$ dressing; every published post-breakdown candidate remains algebraic, weakening (not refuting) Giddings' "beyond algebras" horn. The $N_\ast$ boundary itself is unmoved. [→ notes, notes]
OP-48(c) horn C is PARTIALLY DISCHARGED — the hedge shrinks but does not vanish. INFERENCE, high — theorem-shaped under the full-pair reading Under the GPT-faithful (full-pair, normal-product-state) reading of "recovers QM," the "must interpose Type I" step is now theorem-shaped: a normal product state across a commuting pair forces a Type I interpolating factor (GNS factorization + the established split-inclusion equivalences, anchored to Doplicher–Longo 1984 / D'Antoni–Longo 1983 / Summers 1990), so on the Haag-dual pair $(\mathcal M,\mathcal M')$ exact QM-type composition holds iff $\mathcal M$ is Type I; in the approximate reading, every normal state of a III₁ pair sits at uniform norm distance $\ge 1-1/\sqrt2$ from every product/separable state (Summers–Werner + Tsirelson). The residue relocates from operator algebra to (a) the GPT formalization choice (full-pair vs subsystem reading) and (b) the now-sharpened alternative: commuting III₁ factors support their own consistent multipartite composition theory via universal embezzlement (Quantum 9, 1818 (2025)) — a proven spectrum of composition regimes with two physically motivated endpoints, which algebra classifies but cannot choose. The choice is a physical/axiomatic input — encodes-not-generates recurring at the composition layer. [→ notes]
OP-49 closes to NEAR-NO-GO (conditional exclusion, resolved-negative). INFERENCE, high, with named residuals In the dS/CFT pseudo-entropy first law (arXiv:2511.07915), the entire region- and state-dependent — hence the entire linearized-Einstein — content is carried by the imaginary part (the real part is the region-independent constant $S_{\rm dS}/2$ ; Doi et al., PRL 130, 031601). Reality therefore forces triviality: any protocol making the pseudo-entropy real either nulls the probe ( $\delta\langle T_{tt}\rangle\equiv0$ , no Einstein constraint) or deforms the states/dictionary back to a real-Euclidean/AdS-type (encodes) computation — grounded in the verified Guo–He–Zhang pseudo-Hermiticity reality theorem (JHEP 05 (2023) 021) and mirrored by the unitary-AdS timelike-EE first law (arXiv:2511.17098), where the roles of Re/Im invert exactly under $c\to ic$ . Named residuals: exceptional-point (non-diagonalizable) transition matrices; an SVD-entropy first law (with a holomorphy-breaking argument against it — conjecture, decidable by a stated computation); openness of the horn-(ii) protocol class. House language mirrors OP-41: near-no-go. Correction absorbed: the iter-5 parenthetical had the dS Re/Im split backwards — the divergent, region-dependent part is imaginary; the finite universal part is real (R6A4-F2). [→ notes]
The closed-universe Hilbert-space debate is mapped — new problem OP-50 CONTESTED. ESTABLISHED that the debate exists; the premise CONTESTED A 2024–2026 literature (Usatyuk–Wang–Zhao, SciPost Phys. 17, 051 (2024); Harlow–Usatyuk–Zhao, arXiv:2501.02359; Nomura–Ugajin) argues dim $\mathcal H_{\rm closed}=1$ per α-sector, with standard QM recovered only relative to an internal observer of effective dimension $\sim e^{S_{Ob}}$ — an independent, path-integral route to the same "QM-selection requires an observer-anchored factorization" structure OP-48(c) derived algebraically. The one-dimensionality premise is actively contested (Antonini–Rath–Sasieta–Swingle–Vilar López; Hong Liu's observer-free subregion-subalgebra framing; Engelhardt–Gesteau) — a structural analogy that reinforces, but cannot harden, horn (C). QRF results (DEHK PW=CLPW) show even entropy values are observer-relative; Maldacena's observer fixes the dS sphere partition function's phase but not its coefficient; Zigdon's perturbative-string result suggests the candidate microscopic theory produces no $A/4G$ saddle at all — η input, or absent, never output (HYP-dS-CARDINALITY-R1 extended). [→ notes]
Amplitudes scope update — H5 confidence raised, the gravity bound confirmed structural. ESTABLISHED/INFERENCE Surfaceology extends amplitudes-from-combinatorics beyond planarity (all genus), beyond SUSY (non-SUSY gluons in any D, pions, colored fermions), and partially beyond integrands — for colored, massless theories; Rodina (PRL 134, 031601 (2025)) proves hidden zeros uniquely fix Tr(φ³) trees within a local ansatz (locality is a hypothesis, not derived — referee correction); cosmohedra package the full Tr(φ³) cosmological wavefunction. The decisive bound stands: no positive geometry / surface formalism for gravity (no color, no surface), and every construction consumes flat-space/FRW kinematic data as input — encode-direction evidence. H5: SPECULATIVE, confidence low → low-to-medium, scope rewritten. [→ notes]
Landscape: two additions and six stale-fixes. ESTABLISHED as published programs New entries: Oppenheim's postquantum classical–quantum gravity (PRX 13, 041040 (2023); theorem-level decoherence–diffusion trade-off, Nat. Commun. 14, 7910 (2023); already squeezed ≥15 orders by tabletop data, PRResearch 6, 033076 (2024)) and the swampland dark-dimension scenario (micron-scale inverse-square-law falsifier; a claimed O(1)-parameter DESI fit CONTESTED). Both do carry distinguishing tests — demonstrating the wiki's anti-crank standard is passable in this field, which sharpens by contrast the central wager's not-yet-physics demotion. Stale-fixes: causal-set Hauptvermutung (Müller 2025 preprint splits it into variants with claimed truth values); everpresent Λ CMB-disfavored in current implementations + no published DESI confrontation (an identified gap); GFT condensate mean-field-justified (PRL 130, 141501 (2023)); first CDT–FRG quantitative bridge; spin foams numerically Lorentzian; LQG quantum-OS phenomenology. [→ notes]
Phenomenology: GQuEST is the first funded Program-A-adjacent experiment — and still not a wager test. ESTABLISHED (chain) / INFERENCE (taxonomy) The Verlinde–Zurek geontropic program and GQuEST (PRX 15, 011034 (2025); under construction at Caltech) is the first experiment probing a conjecture derived in the modular/causal-diamond vocabulary the wager privileges — but its chain re-imports $\eta=1/4G$ at the modular→metric step (extending HYP-dS-CARDINALITY-R1's pattern to a phenomenological chain), and it is decision-grade only for [minimal graviton-EFT IR noise $\delta L^2\sim\ell_p^2$ , Carney et al. PRD 113, 106002 (2026)] vs [exotic Planck-normalized UV-IR mixing $\delta L^2\sim\ell_p L/4\pi$ ]. Meanwhile the CQ gate may close from the classical side: ultralocal-continuous dead, ultralocal-discrete below the experimental upper bound (with stated caveats), non-local continuous surviving by ~1 order — possibly before BMV ever runs (no funded BMV roadmap exists). DSW horizon decoherence is real but deflated (Biggs–Maldacena, arXiv:2405.02227: ordinary thermal matter does the same — thermality, not horizon-ness). [→ notes]
Experiment refresh: every standing number re-verified; four rows change. ESTABLISHED g−2 final + WP2025 hold; DESI DR2 (2.8–4.2σ) still current and the w(z) decision point compresses to 2027 (DESI full 5-yr analysis; Euclid DR1 Oct 2026; Roman launch 2026-08-30; LSST underway); S8 substantially eased on the KiDS axis (KiDS-Legacy 0.73σ from Planck — DES Y6 still in tension CONTESTED) while H0 persists; a new ~4σ LHCb $B^0\to K^{*0}\mu^+\mu^-$ anomaly (OP-21 updated); Σm_ν presses below the oscillation floor (~3σ effective-negative-mass tension, entangled with the $w_0w_a$ degeneracy — one ΛCDM stress, not two); matter-wave record 1.7×10⁵ Da at μ=15.5 (strongest macrorealism exclusion to date); GW250114 + GWTC-4.0/5.0 deliver the strongest strong-field GR nulls (area law confirmed, arXiv:2509.08054, PRL 135, 111403; no echoes; Kerr-consistent ringdown); LHAASO pushes linear LIV to $E_{\rm QG,1}>10\,E_{\rm Pl}$ . None of it touches the wager — exactly as the watchlist caveat predicts. [→ notes]

Saturation status after new input. INFERENCE, high The saturation diagnosis survives its first genuine stress test: ~90 refereed findings of new 2023–2026 input produced zero verdict movement at the headline level (fifth consecutive iteration) — only hedge-level restructuring. The ledger now reads: three named hedges (OP-46 residual, MEDIUM, unchanged; OP-48c horn C, shrunk — theorem-shaped under the full-pair reading, residual = GPT-formalization choice + the III₁-embezzlement-composition alternative; HYP-ENCODING-SCREEN, new — the §7 no-test claim inherits the OP-46 MEDIUM hedge), one residual lever closed negative (OP-49 → near-no-go), one restructured but open (OP-44, with a located order- $G_N^2$ obligation and a four-option verdict space), one new contested problem (OP-50), and one compressed verdict-flipper target (derive the indefinite pairing from modular data). The object-level question remains handed to experiment, with the 2027 w(z) verdict the nearest-dated event that could reframe the target. See CONCLUSION.md (amended), EXPERIMENT_WATCHLIST.md (rewritten caveat + four new rows), and the iteration-6 synthesis.

Iteration 7 update (2026-06-10) — the decidable levers, attacked

Iteration 7 executed the four named decidable levers iteration 6 left on the table (the compressed OP-46 target; OP-44's order- $G_N^2$ obligation; OP-49's SVD residual; OP-48c's formalization-choice residue + the OP-50 watch). 9 agents; 4 attack tracks, all adversarially refereed with independent re-derivation, numerical replication, and live citation audits; referee corrections and outcome checks are binding over the original claims. The standing verdict is UNCHANGED at the headline level for the SIXTH consecutive iteration — PARTIAL coherence; Program A encodes-not-generates; the wager not-yet-physics; object-level not forecastable. What moved is the hedge ledger: the principal hedge hardened (OP-46 MEDIUM → MEDIUM-HIGH, conditional), the last substantive OP-49 residual closed negative (conditional), OP-48c's definitional residue resolved, and OP-44's obligation was split into three named theorem-shaped parts — including the first identified perturbative foothold for the failure side.

The designated verdict-flipper was executed — and resolved NEGATIVE in the most informative way possible: the compressed target is achievable but provably geometry-void. ESTABLISHED (η₀ construction) / INFERENCE (vacuity + trichotomy) A canonical Krein fundamental symmetry is derivable from pure $(\mathcal M,\Omega)$ modular data: $\eta_0=\mathrm{sgn}(\ln\Delta)$ is a smuggle-free ( $g_1$ – $g_6$ F, $g_7$ T; $\kappa_\Phi=0$ ) $J$ -odd self-adjoint unitary in $\{\Delta\}''$ , indefinite iff $\omega$ is non-tracial, with Krein signature $(\infty,\infty)$ on any type III factor — one member of an infinite family of $J$ -odd sign functions $\varepsilon(\Delta)$ (referee construction: $\mathrm{sgn}(\sin\ln\Delta)$ ; uniqueness requires an added modular-rescaling-invariance axiom — a correction that pushes negative). But three vacuity results show it carries zero geometric information: (i) in the Borchers/HSMI class — exactly where modular data has causal content — $(H,\Delta,J,\varepsilon(\Delta))$ is unitarily universal (one multiplicity cardinal; $\Delta_{\rm geo}=0$ proven, not graded; substantially the known uniqueness of non-degenerate standard pairs, extended by "any fixed function of $\Delta$ rides along"); (ii) generic algebra-incompatibility ( $\mathrm{Ad}\,\varepsilon(\Delta)\notin\mathrm{Aut}(\mathcal M)$ , numerically demonstrated against an $\mathrm{Ad}\,\Delta^{it}$ control); (iii) non-localizability without a posited carrier. A new modular-form trichotomy lemma (ansatz-scoped on its sesquilinear leg, enumerative on its non-sesquilinear leg) closes the J-route generally: every canonical pairing from modular data is positive, signature-free, or sign-of-Δ Krein (indefinite but universal). Bonus structural result (the strongest finding per referee): the $g_3$ trap is structural — $\langle a^*\Omega,Ja\Omega\rangle\ge0$ by self-duality of the natural cone, so the modular conjugation $J$ carries reflection positivity abstractly (the Euclidean datum), not merely in Bisognano–Wichmann vacua; no $J$ -induced pairing can be secretly Lorentzian. Literature direction-map (negative search, referee-corroborated): all prior art runs Krein → modular (Gottschalk, JMP 43 (2002) 4753; Jakóbczyk 1985, OSTI 5135323; the NCG twist line — Devastato–Farnsworth–Lizzi–Martinetti, JHEP 03 (2018) 089, arXiv:1710.04965), never modular → Krein. Net: the OP-46 hedge hardens MEDIUM → MEDIUM-HIGH (conditional on the modular-constructibility ansatz); HYP-ENCODING-SCREEN inherits; the iteration-6 compressed sentence gets a mandatory precision repair — its literal reading is achieved yet flips nothing, so the faithful reopening target is "derive an algebra-compatible, LOCALIZED (fiberwise) $(n,n)$ pairing from modular data" — the carrier problem (new named item HYP-FACTORIZATION-IS-GEOMETRY), the fifth independent route to converge there. Slogan: the modular sign is to signature what thermal time is to time. [→ notes]
OP-44's order- $G_N^2$ obligation splits three ways (P1a/P1b/P1c) — one part discharged, two located residues, and the first perturbative failure-side foothold. [SHARPENED-OPEN] (P1a) Quadratic-order descent of the complete global-dS constraint set to the CLPW static-patch crossed product is discharged at inference grade (conditional on the inherited quantum-imposition hedge, with fixed-point factoriality not separately proven): decompose $\mathfrak{so}(d+1,1)=(\mathbb R_B\oplus\mathfrak{so}(d))\oplus\mathfrak m$ relative to the pode observers — the boost constraint is the CLPW constraint, the compact $\mathfrak{so}(d)$ twirl preserves Type II (AKL centrality criterion), and the $2d$ coset constraints are solved by the observer worldline acting as a complete dynamical frame; confirmed at the literature level by DEHK §2.5.1 verbatim (citing FJLRW arXiv:2403.11973 and AKL arXiv:2407.01695). This strengthens the survival side below $N_\ast$ — but only by consuming pre-installed $\mathfrak{so}(d+1,1)$ structure, itself an encoding step (HYP-ENCODING-SCREEN mildly reinforced). (P1b = R1) DEHK footnote 12: classical sufficiency of the Taub conditions with worldline clock sources is unchecked, and distributional stress-energy exits the FMM/AMM Sobolev framework — located, decidable OPEN. (P1c = R2) The cubic constraint does not split across the horizon: the one-sided boost-weighted cubic charge ( $\Delta_3=(3d+1)/2$ ) has vacuum fluctuations $\sim A/\delta^{2(d-1)}$ (vs $A/\delta^{d-1}$ quadratic — the latter reproducing the known area law as a consistency check), so subdominance requires a Planck-length dressing-collar floor $\delta\gtrsim\ell_P$ in every dimension — the gravitational parallel of OP-48's stringy $\ell_s$ floor (the iter-4/iter-6 collar-tax leg sharpened from a second independent direction); the relative cocycle is not manifestly affiliated to the CLPW algebra (strictly weaker than provably outer; free-field scaling caveats binding — the exponent could soften). The obstruction horn of R2 is the first identified perturbative foothold for the failure-side reading of OP-44, at order $\kappa^3$ rather than at $N_\ast$ ; both horns are theorem-shaped (discharge = cubic edge modes, KLS-II-style; obstruction = no affiliated renormalization ⟹ outer derivation). The FMM/AMM classical-exhaustion input (quadratic Taub conditions are the only classical obstruction; Moncrief I/II, FMM I, AMM II) carries INFERENCE, medium-high for its dS consequence — the $\Lambda>0$ transfer of the cone theorem is unproven. Literature delta 2024–2026: the nearest discharge machinery exists (Kirklin's all-orders GSL, arXiv:2412.01903, JHEP 07 (2025) 192; Ali Ahmad–Jefferson 2501.01487) but nobody has executed the dS closed-slice cubic step. The bulk $N_\ast$ statement remains theorem-free on both sides; nothing here moves it. [→ notes]
OP-49's last substantive residual closed: the SVD first law on the dS₃ hemisphere FAILS (resolved-negative, conditional, on referee-narrowed grounds). INFERENCE The iteration-6 stated decidable computation was performed. Unconditional pillar: $S_{\rm SVD}$ is real by construction while the verified arXiv:2511.07915 carrier ( $\delta\langle K_A\rangle$ , imaginary $T_{tt}$ ) is purely imaginary — the literal first law $\delta S_{\rm SVD}=\delta\langle K_A\rangle$ fails with no assumptions beyond the paper's own verified content; that alone closes the residual as the iter-6 holomorphy conjecture predicted, now upgraded to an exact operator identity ( $\delta\,\mathrm{Tr}(\mathcal T^\dagger\mathcal T)^m=2m\,\mathrm{Re}\,\mathrm{Tr}[(\mathcal T^\dagger\mathcal T)^{m-1}\mathcal T^\dagger\delta\mathcal T]$ — the SVD response is the Re-projection of a $\mathbb C$ -linear functional). Conditional mechanism (genuinely novel, not proven): the SVD replica object is an alternating $c/c^*$ branched cover whose sheetwise-additive twist dimension carries $c+c^*=0$ — cap-size blindness at the universal level — but exact blindness additionally requires an unproven heterogeneous-cover junction additivity at the branch points, where $2m$ interfaces between conjugate theories meet; interface-CFT precedent (Sakai–Satoh-type transmission-dependent effective central charges) shows naive additivity generically fails (four referee majors, all kept with corrections). Either way no local Einstein constraint survives in the SVD channel: a θ₀-independent response fails the wave equation and CHM inversion outright, and even a cap-dependent real response cannot equal the imaginary carrier. OP-49 stays NEAR-NO-GO with the residual list rewritten: (a) exceptional points; (b) beyond-2511.07915-family generalization; (c′) heterogeneous-cover junction additivity — three items, not two. Byproduct $S_{\rm SVD}=S_{\rm dS}/2$ (region-independent horizon constant) drops to SPECULATIVE (the $q=1$ self-normalization is in tension with the trace-norm inequality given the verified complex pseudo-spectrum); region-independence INFERENCE, conditional on (c′) — consistent with the encodes pattern: the only intrinsically real two-state entropy retains exactly the horizon constant and zero local geometry. Absence re-confirmed (timestamped search-bounded, 2026-06-10): no SVD-in-dS computation or SVD first law exists in the literature; this computation is new. [→ notes]
OP-48(c)'s definitional residue is RESOLVED: the full-pair reading of "recovers QM" is forced (forced-modulo-rejecting-OP-48's-question, per referee) — and the failure locus relocates to the product-preparation primitive. ESTABLISHED (verbatim axiom texts) / INFERENCE, high (the forcing) All three GPT reconstructions — Hardy quant-ph/0101012, CDP arXiv:1011.6451, Masanes–Müller arXiv:1004.1483, axioms checked verbatim against the primary PDFs — quantify composition universally over THE designated pair (no existential quantifier over embedded subfactor pairs exists) and install product states at framework level, upstream of local tomography; the subsystem reading is inexpressible as a weakening of any reconstruction axiom and vacuous where expressible — it can only be implemented by deleting the full pair from the system roster, i.e. abandoning OP-48's question. The sharpest new mathematics of the iteration (referee re-derived, correct): the separation half of local tomography HOLDS on the Type III₁ Haag-dual pair ( $\mathrm{span}\{ab\colon a\in\mathcal M,b\in\mathcal M'\}$ is σ-weakly dense, so distinct normal states differ on some product experiment) — the failure locus is not tomography but the normal-product-preparation primitive (PS), violated uniformly with the Summers–Werner gap $1-1/\sqrt2$ . Horn C upgrades to theorem-conditional-on-framework (a condition shared by Hardy/CDP/MM at framework level, weaker than any single paper's axiom set), conditional exactly on (i) the carried Summers–Werner primary spot-check, (ii) the carried split-equivalence technicalities, (iii) the normal-state postulate. The residue relocates to two named exits: (X1) singular (non-normal) product states — existence holds, but by the referee-corrected Schlieder/Roos + min-norm-domination + Hahn–Banach route (the submitted max-tensor universal-property route is mathematically invalid); admitting them abandons the program's own normal-state postulate; (X2) the III₁/embezzlement composition endpoint, which exits the OPT framework entirely. Both exits leave the framework rather than reinterpret its axioms — the state postulate is an installation choice; encodes-not-generates recurring at the state-postulate layer. OP-50 watch: three verified post-2026-01 developments — Harlow–Usatyuk–Zhao published (JHEP 02 (2026) 108); Nomura–Ugajin arXiv:2602.13387; Higginbotham arXiv:2512.17993 (JHEP 03 (2026) 183) — plus a third position: Miyata–Horikoshi–Tajima–Ugajin (arXiv:2606.04575), symmetric-orbifold closed-universe sectors whose pre-gauge-constraint dominant sector is exponentially large with hyperfinite Type II₁ structure and whose post-constraint physical Hilbert space grows only polynomially in $N$ (neither strictly 1-dim nor $e^{S_{Ob}}$ ; cross-feeds OP-44's survival side as a model-specific construction, not a theorem). OP-50 stays CONTESTED, now three-cornered. Negative sweep (dated process note): no EGH/ARSSV/Liu direct responses post-2026-01. Still NOT-YET-PHYSICS: no experimental channel. [→ notes]

Saturation status and the hedge ledger after iteration 7. INFERENCE, high The saturation diagnosis survives its second stress test — this time against the wiki's own designated decidable levers rather than new field input: four refereed tracks, every outcome label upheld by the referee (one conditionally, on narrower grounds), and zero verdict movement at the headline level — the sixth consecutive iteration. The ledger now reads: hedge 1 (OP-46 residual) HARDENED, MEDIUM → MEDIUM-HIGH conditional on the modular-constructibility ansatz — the J-route is closed by construction-plus-vacuity rather than by absence, and the reopening target re-compresses to the carrier problem, where five independent routes now close (the Type III₁ no-factorization frontier; the split/collar relocation; the CFS carrier posit; the OP-49 protocol deformations; and now the η₀ localization step — recorded as HYP-FACTORIZATION-IS-GEOMETRY); hedge 2 (OP-48c horn C) shrinks again and changes character — no definitional freedom remains in the reading; the residue is the normal-state postulate plus the composition-primitive choice (X1/X2); hedge 3 (HYP-ENCODING-SCREEN) inherits the hardened MEDIUM-HIGH grade with the same ansatz condition attached. One residual lever closed negative-conditional (OP-49's SVD channel, residual list now a/b/c′); one restructured into three theorem-shaped parts with the first failure-side perturbative foothold at order $\kappa^3$ (OP-44, P1a discharged / R1, R2 located); $N_\ast$ untouched. No experimental channel emerged from any track — NOT-YET-PHYSICS unchanged; the experiment watchlist is untouched and the 2027 w(z) decision point remains the nearest-dated external event. See CONCLUSION.md (amended), OPEN_PROBLEMS.md, HYPOTHESES.md, and the iteration-7 synthesis.

References

See BIBLIOGRAPHY.md. This synthesis rests on the sources cited across the domain and analysis pages; no claim here is asserted beyond what those pages support.