{"id":"01da4eee-d4fa-478a-89dc-3ebad92bbd08","arxiv_id":"2607.14215","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Semiclassical gravity in de Sitter space produces four-point observer correlators whose sign of the scrambling correction is opposite to the black-hole case (anti-scrambling); the authors propose bounded-energy quantum systems with folded Euclidean time as a realization.","lead":"Gravity calculations in a simple model of de Sitter space show that its horizon behaves backwards from a black hole: a perturbation makes other particles arrive sooner, and a strong enough perturbation can destroy the whole universe. The authors propose that quantum systems with both a highest and lowest energy could mimic this 'anti-scrambling', setting a concrete target for any quantum theory of de Sitter space.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"dS3 anti-scrambling sign rests on unproven s-wave dominance; inelastic channels could enter at the same order.","rationale":"The reader's strongest concern — s-wave/inelastic truncation for dS3 — is also the most load-bearing issue I can identify. The dS2 calculation has independent support from the worldline theory, so the sign claim is not globally fragile. But the paper explicitly says it has delegated the s-wave justification and the inelastic scattering to other papers; without that justification, the dS3 extension is a conjecture. This directly affects the central claim of anti-scrambling in 'de Sitter space' and any implication that all dS static-patch quantum descriptions must reproduce the computed correlators. The Euclidean-fold proposal (Sec. 6) is admittedly incomplete and could fail (Stokes phenomenon, n ambiguity), but it is presented as a provisional mechanism, not as the core derivation of the sign. The factor-of-two Lyapunov discrepancy with [24–26] is a concerning inconsistency, but it does not by itself invalidate the anti-scrambling sign; it deserves a separate quantitative reconciliation. Therefore the correct verdict remains CONDITIONAL: the dS2 result is convincing, but the dS3 generalization and the 'every quantum description' statement require the s-wave dominance (or an explicit restriction to dS2) to be supplied. My recommendation is no change to the reader's verdict.","tokens_in":51017,"tokens_out":4208,"duration_ms":50404,"concrete_test":"Compute the leading 1/b0 correction to the dS3 OTOC without imposing s-wave truncation: use the full non-symmetric shockwave shift h(φ) of Eq. (3.51), insert a probe wavepacket with the angular profile of a heavy χ^n operator, and evaluate the connected part of G4 averaged over φ. If the resulting sign of the 1/b0 term differs from Eq. (4.36)/ (B.9), or if the s-wave contribution is not parametrically dominant in the limit n→∞, the dS3 anti-scrambling claim fails. Equivalently, compare the partial-wave expansion of the two-shell S-matrix with the s-wave-only eikonal phase using the coupled-channel framework of [27,29].","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central observable is the sign of the 1/b0 correction in the observer OTOC, Eq. (4.36), which is opposite to the black-hole case and is attributed to the negative scattering phase δ (4.23). This sign is robustly derived in dS2/JT, where the worldline computation (5.35)–(5.37) independently reproduces it. The advertised generalization to dS3, however, hinges on an unproven truncation: the closing paragraph of Sec. 4 concedes that in dS3 'quantum mechanically there can be inelastic scattering out of the s-wave sector' and then restricts to heavy operators χ^n so that 'the s-wave sector dominates'. This is asserted, not derived; the systematic treatment is explicitly delegated to [27,29]. Eq. (B.9) is claimed to be correct in dS3 for such operators, but no partial-wave expansion or estimate of the inelastic matrix element is provided. The recoil effect of Sec. 3.5 is a concrete non-spherically-symmetric contribution, and the paper states it is 'systematically included in [27,29]' — i.e., not included here. If the inelastic channel contributes at the same order 1/b0, the phase δ — and therefore the sign of the correction — could be altered or canceled. Since the abstract and title make a statement about 'de Sitter space' generally, the dS3 claim is load-bearing for the paper's headline. The dS2 result is not affected by this concern, but the dS3 generalization, which is needed for the strong conclusion that 'every quantum description of the dS static patch' must reproduce anti-scrambling, remains conditional.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper gathers gravitational 'data' on observer correlation functions in de Sitter space, with the goal of constraining any fundamental quantum description of the dS static patch. The authors reduce spherically-symmetric dS3 gravity with dust to JT gravity with positive cosmological constant (Sec. 3.1), construct one- and two-shell solutions, and use them to compute the time advance of a probe (Secs. 3.3-3.4). They then derive the observer two-point function and the out-of-time-order four-point function, Eq. (4.36), whose first gravitational correction has a sign opposite to the black-hole result. This 'anti-scrambling' correction is traced to the negative eikonal phase (4.23). The result is cross-checked in Sec. 5 using a worldline theory, with the four-point correction (5.37) reproducing (4.36) for operators of dimension one. The paper also notes that for time separations beyond 2t_scr the back-reaction destroys the universe (Sec. 3.4). In Sec. 6, the authors propose that anti-scrambling correlators could arise from a quantum system whose Hamiltonian is bounded above and below, using correlators folded in Euclidean time. The proposal is explicitly incomplete: the fold integer n is undetermined, and the analytic continuation is assumed to commute with the semiclassical limit. The dS3 generalization is also not fully established because it relies on s-wave dominance that is asserted rather than derived. The dS2/JT results, by contrast, appear internally coherent and are ba","tokens_in":51295,"tokens_out":6885,"duration_ms":78724,"significance":"If the central claim holds, the paper provides concrete, non-perturbative constraints on any quantum-mechanical description of the de Sitter static patch: a conventional thermal two-point function but an anti-scrambling four-point function, with a violation of the thermal commutator bound in the two-sided continuation (4.38). The dS2/JT derivation is a genuine strength: the sign of the 1/b0 correction follows from the classical time-advance via the phase-shift relation, and the worldline calculation in Sec. 5 independently reproduces the four-point result. The paper is also admirably transparent about its own limitations, explicitly flagging the dS3 inelastic-scattering gap, the undetermined fold integer n, and the Stokes-phenomenon risk. However, the headline claims about 'de Sitter space' and about a quantum realization via Euclidean folds are not yet at the same level of support as the dS2 calculation. The dS3 claim depends on a truncation that is not justified in this manuscript, and the Euclidean-fold idea remains a conjecture rather than a demonstrated realization. Because those two elements are load-bearing for the paper's broadest conclusions, the manuscript requires revisi","major_comments":[{"comment":"The advertised generalization to dS3 is load-bearing for the paper's headline, but it rests on an unproven s-wave truncation. The text concedes that in dS3 'quantum mechanically there can be inelastic scattering out of the s-wave sector' and then restricts to heavy operators chi^n at large n with the assertion that 'the s-wave sector dominates'. No partial-wave expansion or estimate of inelastic matrix elements is provided, and the systematic treatment is explicitly delegated to references [27,29]. Moreover, Sec. 3.5 identifies a concrete non-spherically-symmetric contribution, the recoil effect, which is 'systematically included' in [27,29], not here. Since the abstract and title make a general statement about de Sitter space, and the strong conclusion that every quantum description of the dS static patch must reproduce these correlators depends on the sign of the 1/b0 correction, the d","section":"Sec. 4, closing paragraph; App. B, Eq. (B.9)"},{"comment":"The Euclidean-fold proposal is not yet a demonstration that anti-scrambling can be realized in a quantum system with a Hamiltonian bounded both above and below. The fold prescription contains an arbitrary integer n, and the text states: 'At present we do not have a way of selecting a particular value of n' and 'we do not at the moment know how to turn this observation into a general rule'. The prescription is also tied to a specific operator ordering, and the paper acknowledges that the analytic continuation could fail if a Stokes phenomenon disrupts the semiclassical limit. The abstract, however, states that anti-scrambling 'can be realized in a quantum system whose Hamiltonian is bounded from both above and below using correlators that are folded in Euclidean time.' As written, this overstates the status of the proposal. To support the claim, the paper should either provide a concrete","section":"Sec. 6, Eqs. (6.9)-(6.12)"}],"minor_comments":[{"comment":"The paper states that previous papers [24-26] found a Lyapunov exponent 4pi/beta, while the present work finds 2pi/beta saturating the chaos bound. This discrepancy is not reconciled. Since the difference might be due to a different time coordinate, to a different physical effect, or to an error in the earlier works, the authors should add an explicit comparison. This is important for readers trying to place the new 'anti-scrambling' claim relative to the existing literature.","section":"Introduction and Sec. 2"},{"comment":"The two-point function (5.31) is obtained in a renormalization scheme where q=1, chosen so that the renormalized Euclidean time has period 2pi. The constant -Delta^2/(pi b0) in (5.31) looks scheme-dependent. Please state explicitly whether the physical correlator is scheme-independent at this order, and if so, where the scheme dependence cancels.","section":"Sec. 5.2, Eq. (5.31)"},{"comment":"Minor typos and formatting: in Sec. 1, 'the operatore τ Hwithτ>0' lacks spaces; in Sec. 6, the figure label contains 'ds Sitter' instead of 'de Sitter'; and the text alternates between 'dS2' and 'dS 2' in a way that should be unified. These do not affect the scientific content.","section":"Various"}],"recommendation":"major_revision","confidential_remarks":"The dS2/JT core is solid and likely publishable. The main weaknesses are the dS3 s-wave-dominance assertion and the incompleteness of the Euclidean-fold realization. If the authors can provide a partial-wave estimate for the inelastic channels, or explicitly reframe the paper's general claims as applying to dS2/JT only, the manuscript could become acceptable. The fold proposal would be more convincing with at least one concrete toy model. I would not recommend rejection, because the central dS2 calculation is coherent, cross-checked, and clearly a useful contribution to the de Sitter holography discussion."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The punchline: this paper gives the first clean semiclassical calculation of an observer OTOC in de Sitter where the gravitational correction has the opposite sign to the black-hole case, and a Lyapunov exponent 2π/β saturating the chaos bound. The dS2/JT part is solid; the advertised dS3 generalization is conditional on an unproven truncation, and Sec. 6 should be read as a proposal, not a derivation.\n\nWhat is genuinely new: the anti-scrambling sign is derived from the classical time advance (3.34) via the eikonal phase (4.20), so it is not fitted. The two-shell solutions in sections 3 and Appendix A are explicit, with stated validity regimes, and the worldline theory in Sec. 5 reproduces the four-point function by an independent route (5.37 equals 4.36). The separation of the kinematic recoil effect of [25] from the gravitational backreaction is useful, as is the observation that a two-scrambling-time separation produces a crunch. The paper is honest about its own limitations, which appear in the text.\n\nThe soft spots, in proportion. The largest is the continuation from dS2 to dS3. The last paragraphs of Sec. 4 concede that quantum mechanically there can be inelastic scattering out of the s-wave sector; the restriction to heavy operators χ^n at large n is asserted, not derived, and the systematic treatment is delegated to [27,29]. If inelastic channels enter at the same order 1/b0, the sign of the phase δ — and therefore the sign of the anti-scrambling correction — could be altered. The dS2 result does not depend on this, but the abstract's general claim about 'de Sitter space' and the stronger statement about 'every quantum description' of the static patch do depend on it. The factor-of-two discrepancy with [24-26] (4π/β versus 2π/β) is acknowledged but not quantitatively reconciled, which weakens the comparison to previous work. The Euclidean fold prescription in Sec. 6 has an unselected integer n, a Stokes phenomenon caveat, and a renormalization choice q=1 in the worldline theory; all are stated, but they are not consequences.\n\nNone of this undermines the dS2 computation. The central chain — shell crossing, time advance, eikonal phase, OTOC, and the worldline cross-check — is coherent, and the paper is scoped carefully enough that its main claims are checkable. The dS3 sign and the folds are where a referee should push.\n\nRecommendation: yes, send to peer review. The dS2 anti-scrambling result is a significant datum for de Sitter holography and deserves referee time. The authors should be asked to either demonstrate the s-wave dominance in dS3 or explicitly restrict the claim to JT, and to either make the fold mechanism definite or label it as a conjecture. I would cite the dS2 result in my own work.","headline":"The dS2 anti-scrambling OTOC is the real result; the dS3 generalization is conditional on an unproven s-wave truncation, and the Euclidean folds are a sketch.","tokens_in":51950,"tokens_out":3475,"would_cite":true,"duration_ms":32732,"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 paper claims that in de Sitter space the first gravitational correction to the observer four-point function has the opposite sign to the black hole result, and that this anti-scrambling sign is a key test of any quantum description of t","keywords":["anti-scrambling","de Sitter space","observer correlators","out-of-time-order correlator","Jackiw-Teitelboim gravity","Euclidean folds","scrambling time","Lyapunov exponent"],"falsifier":"Compute the four-point function including the full set of non-spherically-symmetric (inelastic) scattering channels, for instance using the localized perturbation calculation of section 3.5, at the same order in G. If the first gravitational correction to the OTOC (4.36) becomes negative once inelastic scattering is included, the anti-scrambling claim is false. Alternatively, test the Euclidean fold prescription in a concrete bounded-energy quantum system such as the SYK model: if the analytically continued correlator does not match (4.32) in the semiclassical limit, the proposal fails.","tokens_in":50710,"feed_emoji":"🌀","tokens_out":6401,"duration_ms":64803,"temperature":0.7,"pith_summary":"This paper gathers gravitational 'data' about the de Sitter horizon by computing correlation functions on the worldline of an observer. The two-point function looks thermal, but the out-of-time-order four-point function displays anti-scrambling: the first gravitational correction has a positive sign (time advance) rather than the negative sign (time delay) found for black holes. The Lyapunov exponent is 2π/β, saturating the chaos bound, and for perturbations separated by more than two scrambling times the backreaction produces a big crunch singularity. The paper then proposes that such anti-scrambling correlators could arise in an ordinary quantum system whose Hamiltonian is bounded both above and below, through correlation functions folded in Euclidean time.","feed_headline":"De Sitter horizons emit particles earlier, not later","feed_subtitle":"The sign of the first gravitational correction is opposite to the black hole case, a concrete test for quantum de Sitter.","key_machinery":"The central object is the eikonal 2–2 gravitational scattering phase δ(p₊p₋) between the two shells, whose derivative gives the time advance; its negative sign δ ≈ −2p₊p₋/b₀ (Eq. 4.23) is what turns scrambling into anti-scrambling. This phase enters the four-point function (4.19) via e^{iδ}, and the resulting exact expression (4.32) is a perfect match to the black hole out-of-time-order correlator except for the replacement w → −w. The second piece of machinery is the Euclidean fold: the continuation (6.9) reverses the sign of sin((τ₄₃+τ₂₁)/2) without changing the product of two-point functions, converting a scrambling correlator into the anti-scrambling one.","core_discovery":"The central discovery is that the observer four-point function in de Sitter space, computed with classical gravity in the eikonal regime, takes the form (4.32)–(4.36): a product of two-point functions times a factor whose first correction is positive — the opposite sign to the black hole result. This sign traces directly to the negative sign of the 2–2 gravitational scattering phase δ ≈ −2p₊p₋/b₀ (4.23), which in turn is a consequence of the anti-scrambling time advance: throwing a shell through the cosmological horizon moves the horizon away from the observer, so probe particles arrive earlier rather than later. The paper also finds a Lyapunov exponent 2π/β saturating the chaos bound (not 4","pith_inferences":["If the anti-scrambling sign is robust, the requirement of a Hamiltonian bounded both above and below for Euclidean folds suggests that a microscopic description of the static patch might be inherently finite-dimensional, much like a spin model; this is a nontrivial constraint on holographic proposals.","A direct test of the proposal would be to compute the four-point function in a concrete bounded-energy system, such as a finite-size SYK model or a spin chain, using the fold prescription (6.9) and compare with (4.32); the paper itself flags the SYK model as a test of the analytic continuation.","The big-crunch regime at 2t_scr, if robust, offers a sharper prediction than the sign: a microscopic theory must exhibit a divergence or breakdown at that exact time separation, which could discriminate among candidate duals.","The negative-temperature interpretation of the de Sitter static patch (entropy decreases when energy increases) is a natural consequence of the anti-scrambling sign and could serve as an organizing principle for future constructions."],"forward_implications":["Any fundamental quantum description of the de Sitter static patch must reproduce the anti-scrambling four-point function (4.36), including the positive sign of the first gravitational correction — a sharp, quantitative test.","The Lyapunov exponent in de Sitter is 2π/β, saturating the chaos bound, not the 4π/β reported in earlier work.","Two-sided observer correlators in de Sitter violate the thermal bound (4.38), showing that ordinary thermal quantum mechanics cannot produce them; a non-standard ingredient such as folded Euclidean time is required.","For time separations beyond 2t_scr, gravitational backreaction creates a big crunch singularity in the post-collision region: a thermal-scale perturbation can destroy the universe if thrown in more than two scrambling times before another emission.","If the Euclidean fold proposal is correct, de Sitter correlators would be reproduced by a finite (bounded-above-and-below) quantum system, effectively operating at negative temperature."],"fun_headline_variants":["De Sitter scrambles in reverse: probes arrive early","Horizon repels, not attracts: anti-scrambling in de Sitter","Black holes delay; de Sitter horizons rush out","First sign flip: de Sitter's quantum test is anti-scrambling","Lyapunov bound intact, but scrambler runs backward"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The anti-scrambling sign rests on assuming the two-shell gravitational collision is dominated by the spherically symmetric (s-wave) sector and that a single eikonal phase δ ≈ −2p₊p₋/b₀ captures the scattering; if inelastic scattering out of the s-wave sector contributes at the same order, the sign can flip. The paper asserts, but does not derive, that restricting to heavy operators χⁿ makes the s-wave sector dominate.","fun_headline_variants_meta":{"raw":{"variants":["De Sitter scrambles in reverse: probes arrive early","Horizon repels, not attracts: anti-scrambling in de Sitter","Black holes delay; de Sitter horizons rush out","First sign flip: de Sitter's quantum test is anti-scrambling","Lyapunov bound intact, but scrambler runs backward"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000239,"raw_usage":{"total_tokens":1319,"prompt_tokens":682,"completion_tokens":637,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":426,"completion_tokens_details":{"reasoning_tokens":557}},"tokens_in":426,"tokens_out":637,"duration_ms":7070,"temperature":1.0,"reasoning_tokens":557,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T01:54:50.100135+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the four-point function including the full set of non-spherically-symmetric (inelastic) scattering channels, for instance using the localized perturbation calculation of section 3.5, at the same order in G. If the first gravitational correction to the OTOC (4.36) becomes negative once inelastic scattering is included, the anti-scrambling claim is false. Alternatively, test the Euclidean fold prescription in a concrete bounded-energy quantum system such as the SYK model: if the analytically continued correlator does not match (4.32) in the semiclassical limit, the proposal fails.","supporting_citations":[],"review_version":2}