{"id":"40ca58e5-1582-4517-b8c1-6b407d49a2b9","arxiv_id":"2607.14042","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Out-of-time-order correlators along a de Sitter observer worldline, computed with the shockwave eikonal formalism including recoil and backreaction, violate the cyclicity and positivity required of a trace, falsifying a weak static-patch holography conjecture.","lead":"This paper tests a popular idea about de Sitter space—that a gravitational path integral with an observer computes a quantum trace—and finds it fails. Out-of-time-order correlators violate two basic properties of traces, so the conjecture as stated is likely false.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Contradiction in §3.1 hinges on unproven factorization of time-ordered four-point functions; edge bounds |f|≤1 could fail if connected corrections grow like Ge^t or (e^t/m)^2.","rationale":"The reader's weakest_assumption correctly identifies the factorization assumption in Eqs. (3.8) and (3.10) as the most load-bearing unsupported step. The central positivity argument is otherwise well developed: the first-order perturbative coefficient is positive and is shared by both F12 and F14, so the two-resummation ambiguity does not remove the sign issue. The unification of recoil and backreaction via lifted ℓ=1 zero modes is a genuine technical advance, and the eikonal formulas are cross-checked against the tetrahedron saddle in overlapping regimes. However, the contradiction is proven only if the exact trace satisfies the boundary bounds |f|≤1, and those bounds use a factorization that is asserted rather than derived. The footnote calls the factorization 'safe' based on OPE intuition, but no estimate of the connected correction is provided; if that correction is not subleading by 1/m at the relevant edge, the maximum-modulus step does not close. This does not move the verdict: the paper is already CONDITIONAL, and the same caveat was flagged by the reader. A direct computation of the connected TOC in the same eikonal framework would settle the issue.","tokens_in":40134,"tokens_out":11484,"duration_ms":113636,"concrete_test":"Compute the connected part of the time-ordered four-point function in Eq. (3.8) at the edge τ=π/2, α=π/2 using the same eikonal shockwave framework of §2.4 (or, more simply, in the large-ν' recoil limit of §2.6), without assuming factorization. If the connected correction is bounded by const/m uniformly in Lorentzian time s, the PL bound closes and the contradiction stands; if it contains a term scaling as G e^s or e^{2s}/m^2, the edge bound fails and the central falsification claim is not established. This is a finite one-loop/eikonal check, not a new conjecture.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The falsification of the weak static-patch conjecture rests on the Phragmén-Lindelöf step in §3.1: f is analytic in the strip and is shown to satisfy |f|≤1 on the two edges. The edge bounds use Cauchy-Schwarz on A12 and A14, but the last lines of Eqs. (3.8) and (3.10) replace the exact four-point trace by the factorized product Gν(...)Gν'(...) with the footnote 'we assumed that time-ordered correlators approximately factorize.' This is load-bearing: if the connected part of the time-ordered four-point function is not O(1/m) but instead contains a contribution of order Ge^s or (e^s/m)^2 at the edge, the upper bound becomes 1+O(Ge^s), so |f|≤1 is not established, and the conclusion that |f|>1 inside the strip is not a contradiction. The authors call this a 'safe assumption' and argue OPE corrections are small, but no computation is supplied. The positivity violation is only as strong as this unproven edge bound. The F12/F14 ambiguity and contour issues are acknowledged in the text and do not weaken the first-order positivity argument as much as the unsupported factorization.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper tests the 'static patch holography' conjecture, which equates Euclidean gravitational correlators along an observer worldline with traces in a Hilbert space. After reviewing encouraging evidence from the sphere partition function and two-point function, the authors compute four-point OTOCs in an eikonal/shockwave approximation, incorporating both gravitational backreaction and observer recoil. The central technical result is the sign of the first perturbative correction to the regularized OTOC: Eq. (2.42) shows a positive, exponentially growing correction, opposite to the alternating series familiar from black holes. This leads to two different resummations, F12 and F14, which are exchanged by cyclicity. The paper then argues in Section 3.1 that the positive first-order correction conflicts with the combination of analyticity, cyclicity, and positivity required of a finite-dimensional trace, by applying the Phragmén–Lindelöf/chaos-bound argument. It concludes that a weak version of the static patch holography conjecture is false. Appendices provide supporting calculations in a dS JT toy model, higher-dimensional shock waves, and the pure-recoil tetrahedron saddle.","tokens_in":40406,"tokens_out":3396,"duration_ms":37362,"significance":"If the central claim is correct, the paper establishes a concrete obstruction to a natural and actively studied conjecture about de Sitter quantum gravity. The technical work is substantial: the shockwave formalism is extended to include observer recoil via lifted ℓ=1 zero modes, and the dS JT model gives a clean analytic comparison. The explicit perturbative sign in Eq. (2.42) is a parameter-free prediction of the formalism, and the distinction between F12 and F14 provides a useful organizing framework. The main result is falsifiable in the sense that a specific sign of the OTOC growth is computed and then shown to be incompatible with trace positivity. However, the strength of the final contradiction depends on an unproven factorization assumption in the Phragmén–Lindelöf argument, which is load-bearing and not merely a presentation issue.","major_comments":[{"comment":"The edge bounds |f|≤1 are essential for the Phragmén–Lindelöf step, but the final equalities in Eqs. (3.8) and (3.10) replace the exact time-ordered four-point function by a factorized product of two-point functions. The footnote calls this a 'safe assumption' based on OPE dominance of the identity, but no estimate is supplied for the connected part. If the connected time-ordered correlator at the strip edges contains contributions of order G e^s or (e^s/m)^2, the Cauchy–Schwarz bound becomes |f|≤1+O(G e^s), and the conclusion |f|≤1 inside the strip does not follow. Since the entire falsification of the weak static-patch conjecture rests on this bound, the factorization must be proved (or bounded by a quantity smaller than the O(1/m) error budget) before the contradiction is established.","section":"§3.1, Eqs. (3.8), (3.10), footnote 20"},{"comment":"The matching between the eikonal OTOCs and the large-ν′ recoil limit is stated as a result but the derivation is explicitly omitted: 'We will not show the calculation here, but in the antipodal configuration... it exactly matches (2.78)'. This identification is then used to locate where F12 and F14 are related by analytic continuation through the small-time region, and it underpins the discussion in §2.5.2 and Fig. 8. The connection is a nontrivial technical claim and should be shown at least in outline; without it, the cyclicity discussion is partly based on an unverified match.","section":"§2.6, Eqs. (2.78)–(2.80)"},{"comment":"The statement that 'the entire perturbative series is positive' is used to motivate the Borel-resummation ambiguity, but only the first two orders are displayed. Since the Section 3 argument uses only the first-order sign, this is not a blocking issue, but if the stronger statement is intended literally, a proof or reference is needed.","section":"§2.4.2, after Eq. (2.42)"}],"minor_comments":[{"comment":"The sentence 'Here, the ± refers to the ≶' is confusing because the equation also has a ± sign in the exponential; please clarify the notation.","section":"§2.4.2, Eq. (2.37)"},{"comment":"The normalization constant c is described only verbally. Please give its explicit definition or explain how it is fixed by the early-time value of f.","section":"§3.1, Eq. (3.1)"},{"comment":"There are minor typographical issues, e.g. 'gravitional' in §2.1 and the axis label 't log|C|' in Figs. 3 and 11, which should be cleaned up.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of a hep-th journal and the coordinated submission with refs. [32–34] is transparently disclosed. The main concern is the unproven factorization in §3.1; if the authors can supply a bound on the connected time-ordered four-point function at the strip edges, the central claim would be much stronger. The omitted derivation in §2.6 should also be added before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is worth engaging. The genuinely new piece is the eikonal OTOC computation that treats observer recoil and gravitational backreaction on the same footing, and it produces a clean result: the first perturbative correction to the regularized OTOC is positive and grows like G e^t, the opposite of the chaos-bound expectation for a trace. The sign calculation around (2.42) is explicit and looks solid, and the lifted ℓ=1 shockwave zero modes are a real technical advance over treating recoil and backreaction separately. The tetrahedron saddle analysis is also interesting, and the paper is candid about the F12/F14 ambiguity, including the possibility that an averaged contour could restore cyclicity.\n\nThe soft spot is exactly where the stress test puts it. The contradiction in §3.1 requires |f|≤1 on the edges of the analyticity strip, and the edge bounds in (3.8) and (3.10) are obtained by replacing a time-ordered four-point function with the factorized product of two-point functions. The footnote says 'we assumed that time-ordered correlators approximately factorize.' That is load-bearing, not a side remark. The authors may well be right that OPE corrections are small, but they do not supply an estimate showing the connected part cannot be O(G e^t) or O((e^t/m)^2) at the edge. Without that, the Phragmén-Lindelöf step does not close, and the paper does not strictly falsify the weak conjecture. The positivity violation from the sign itself is more robust, but the conclusion that it conflicts with a Hilbert-space trace depends on the unproven bound.\n\nThe paper is also honest about other soft spots: the large ν' matching calculation in §2.6 is stated without derivation, and the contour questions for F12/F14 are acknowledged rather than resolved. Those are minor relative to §3.1. The citation pattern is fine; relying on earlier work by overlapping groups is appropriate when the shockwave technology and the conjecture under test come from those papers.\n\nWho gets value: people working on de Sitter holography, especially static patch and observer worldlines. It is one of the sharper recent attacks on the conjecture. I would bring it to a reading group and cite it, but I would not treat the falsification as established until the factorization assumption is justified or replaced. The paper deserves a serious referee; the report should concentrate on §3.1 and the size of the connected time-ordered correlator at the edge.","headline":"A serious technical challenge to weak static-patch holography, with the falsification claim resting on one unproven factorization assumption in §3.1.","tokens_in":40914,"tokens_out":3055,"would_cite":true,"duration_ms":29350,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The static-patch holography conjecture, in its weak trace form, is false: the out-of-time-ordered correlator computed from the Euclidean gravitational path integral with an observer worldline has a leading time-growing correction that incre","keywords":["static patch holography","de Sitter space","out-of-time-ordered correlator","shockwave","observer recoil","chaos bound","trace positivity","eikonal approximation"],"falsifier":"Compute the next-order correction to the factorized time-ordered products in Eqs. (3.8) and (3.10) and check whether it grows like e^t or is comparable to the O(1/m) terms; if it grows exponentially, the upper bound |f|≤1 may fail and the positivity contradiction collapses. Alternatively, find a single cyclic positive function that matches the eikonal series to all orders, which would refute the paper's conclusion.","tokens_in":39993,"feed_emoji":"🌌","tokens_out":4938,"duration_ms":47037,"temperature":0.7,"pith_summary":"The paper tests the static-patch holography conjecture—that correlation functions along a de Sitter observer's worldline, computed from the Euclidean gravitational path integral, equal a trace over a Hilbert space—by computing a four-point out-of-time-ordered correlator (OTOC) in the eikonal approximation. Its central finding is that a weak version of this conjecture fails: the leading time-growing correction to the OTOC has the wrong sign, increasing rather than decreasing the regularized correlator, which is incompatible with positivity of a trace and with the chaos bound. Shock-wave scattering on the de Sitter horizon produces a time advance, not the time delay familiar from black holes, and this sign flips the entire perturbative series. The calculation also reveals two different resummations of the eikonal series, exchanged by the cyclicity property of a trace, so no single cyclic answer reconciles both. A sympathetic reader comes away with a concrete obstruction: if the paper is right, the dictionary between worldline correlators and traces needs modification.","feed_headline":"De Sitter OTOC breaks trace positivity, sinking weak holography","feed_subtitle":"A shockwave time advance makes the out-of-time-ordered correlator grow, contradicting cyclicity and positivity of any Hilbert-space trace.","key_machinery":"The central object is the gravitational shock-wave solution on the de Sitter horizon sourced by a particle with large null momentum. In pure de Sitter the shock-wave equation has a pure-gauge zero mode at angular momentum ℓ=1; the observer's presence lifts this mode, turning it into a physical recoil effect, so a single shock-wave calculation includes both gravitational backreaction and observer recoil. In three dimensions the shock profile h(ϕ) has a negative average, encoding time advance, and the eikonal OTOC is an integral over shock strengths with a phase exp{i p+ p− h(ϕ)}. The 't Hooft scattering phase and its two contour choices produce the two resummations F12 and F14.","core_discovery":"Working with a round-sphere Euclidean de Sitter geometry plus an observer worldline, the authors show that the OTOC of two pairs of light fields, computed to leading order in Newton's constant G and inverse observer mass 1/m, behaves as 1 + positive (G e^t + ...) + positive ((G^2 + G/m + 1/m^2) e^{2t} + ...). Because the coefficients are positive rather than alternating, the regularized OTOC grows above its early-time value—the opposite of what would happen for a positive trace with a finite Hilbert space. They then turn this sign into a rigorous contradiction: together with the trace ansatz and analyticity, the growth forces a function |f| to exceed 1 inside a strip while being bounded by 1","pith_inferences":["(editorial) A plausible way to rescue holography is to abandon the strict trace interpretation and instead treat the worldline observables as generating a non-tracial state or a complex integration contour; the paper leaves this open.","(editorial) The same time-advance mechanism should show up in other worldline observables, such as a speedup of causal contact between antipodal points; searching for those effects in lower-dimensional toy models could test whether the sign is universal.","(editorial) If the factorization assumption at the edges of the analyticity strip fails exponentially in time, the contradiction would disappear; a direct next-order check of the factorized time-ordered products is the cleanest way to decide."],"forward_implications":["If the central claim is right, the Euclidean path integral with an observer worldline does not compute a cyclic positive trace for four-point OTOCs, so the static-patch holography conjecture must be weakened or replaced.","De Sitter shock waves produce time advance rather than time delay, and the OTOC grows; all coefficients in the eikonal series are positive, making the Borel resummation ambiguous.","Cyclicity is broken by a choice of integration contour: F12 and F14 are different, and no single contour picks a cyclic answer consistent with both Cauchy-Schwarz bounds.","The static-patch dS JT toy model captures the wrong-sign effect: its eikonal action has a negative coupling C, giving the same kind of OTOC growth as the higher-dimensional shock wave.","Even the tracial (infinite-temperature) OTOC grows initially, though it remains bounded by the time-ordered correlator."],"fun_headline_variants":["OTOC growth kills de Sitter trace conjecture","Shockwave positivity violation dooms holographic trace","De Sitter OTOC defies cyclicity and positivity","Negative shocks flip sign, sink static patch trace"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The contradiction assumes that time-ordered correlators on the observer's worldline approximately factorize at the edges of the analyticity strip, with errors small compared to the O(1/m) effects; if those errors grow exponentially in time, the Phragmén-Lindelöf bound does not close and the claimed contradiction with positivity does not follow.","fun_headline_variants_meta":{"raw":{"variants":["OTOC growth kills de Sitter trace conjecture","Shockwave positivity violation dooms holographic trace","De Sitter OTOC defies cyclicity and positivity","Negative shocks flip sign, sink static patch trace"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000187,"raw_usage":{"total_tokens":1152,"prompt_tokens":717,"completion_tokens":435,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":461,"completion_tokens_details":{"reasoning_tokens":373}},"tokens_in":461,"tokens_out":435,"duration_ms":4532,"temperature":1.0,"reasoning_tokens":373,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T02:55:00.869125+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the next-order correction to the factorized time-ordered products in Eqs. (3.8) and (3.10) and check whether it grows like e^t or is comparable to the O(1/m) terms; if it grows exponentially, the upper bound |f|≤1 may fail and the positivity contradiction collapses. Alternatively, find a single cyclic positive function that matches the eikonal series to all orders, which would refute the paper's conclusion.","supporting_citations":[],"review_version":1}