{"id":"38b979cb-ed99-4552-ac19-a1826b1c86c3","arxiv_id":"2604.20170","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A two-gap divergence in holographic complexity obstructs both reaching extremality and reaching a naked singularity, unifying the third law and weak cosmic censorship.","lead":"The paper proposes that two well-known black-hole rules—the third law and weak cosmic censorship—are actually one rule about computational effort: making an exactly extremal black hole costs infinite complexity, and so does crossing it to reach a naked singularity. It supports this with a calculation for overcharged black holes in anti-de Sitter space.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Reverse triangle inequality is the unproved load-bearing premise; without it, divergent complexity of formation does not imply divergent relative complexity between the extremal and naked-singularity sectors.","rationale":"The reader's weakest assumption—the unproved use of the reverse triangle inequality—is exactly the load-bearing step in the paper. Eq. (22) determines the near-singularity action of the naked-singularity WdW patch, hence C(B_sprx) relative to the AdS reference. The dynamical obstruction, however, requires the complexity distance between the extremal state and the naked-singularity sector to diverge. The paper connects these by a single sentence invoking the reverse triangle inequality, with no derivation or citation. This is not a minor gap: if pairwise complexity distance can be finite while the two complexities relative to the reference differ divergently, then a finite-complexity circuit could map E to B_sprx, and the central conclusion fails. Standard circuit complexity does satisfy this inequality, but CA/CV complexity as used in this paper is not shown to be such a metric. I also note the reader's secondary point about Eq. (8): for D>4, the claimed finiteness of the Einstein-Hilbert bulk term appears incorrect, undermining the abstract's exact-cancellation statement, though not the no-logarithmic-divergence core. Since the main concern is precisely the one the reader flagged, the CONDITIONAL verdict remains appropriate and no adjustment is needed.","tokens_in":6910,"tokens_out":11585,"duration_ms":86357,"concrete_test":"Independently derive the reverse triangle inequality for CA/CV complexity from the definitions in Refs. [3,4,13,16]. Specifically, show that for any three states A,B,C, the complexity distance D satisfies D(A,C) ≤ D(A,B)+D(B,C). If the derivation fails, then replace the RTI step by a direct computation of D(E,B_sprx) for D=4 RN-AdS (e.g., via the action of an interpolating WdW patch or a Vaidya-type charged shell geometry) and check whether it diverges. If neither can be shown, the two-gap obstruction is unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central dynamical conclusion requires that a divergent difference in complexities of formation, |C(E)-C(B_sprx)|, implies a divergent complexity distance between the two states. In the 'Extremal Barrier' section, immediately after Eq. (22), the paper states: 'the reverse triangle inequality for complexity distance then implies divergent relative complexity.' This sentence is the only bridge between Eq. (22), which computes C(B_sprx) as I_WdW/(πℏ) relative to the AdS vacuum, and the claim that the transition E→B_sprx is dynamically obstructed. The reverse triangle inequality D(E,B_sprx) ≥ |C(E)-C(B_sprx)| holds only if (i) state complexity is defined as a distance from a fixed reference and (ii) the pairwise complexity distance D is a metric satisfying the triangle inequality. Neither property is established for CA/CV complexity, and no proof or citation is supplied. If D(E,B_sprx) were finite despite the divergence in C(E)-C(B_sprx), a simple circuit could connect the two states, and the two-gap obstruction would evaporate. A secondary, non-central issue: the claim in Eq. (8) that the Einstein-Hilbert bulk term is finite for all D>4 appears incorrect; with R-2Λ ~ r^{-(2D-4)} near r=0, the integrand scales as r^{-(D-2)}, giving an ϵ^{-(D-4)} divergence for D>4. This affects the exact-cancellation statement in the abstract, but not the no-logarithmic-divergence core.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes that the extremal surface in the AdS black-hole parameter space is a complexity-geometric barrier. In both CA and CV prescriptions, the complexity of formation of extremal RN-AdS black holes is logarithmically divergent, while every subextremal state is finite. The author computes the WdW action of overcharged RN-AdS and argues that the naked-singularity complexity carries at most a power-law divergence, never a logarithmic one. Combining the two gaps with the reverse triangle inequality for a complexity distance and a finite complexity-growth-rate assumption, the paper concludes that no finite-time physical process can cross the extremal surface, giving a complexity origin for the third law and weak cosmic censorship.","tokens_in":7288,"tokens_out":10194,"duration_ms":96119,"significance":"The question addressed is timely and potentially important: connecting holographic complexity to the dynamical inaccessibility of naked singularities would be a genuinely new structural result. The paper has clear strengths: the WdW calculations are direct and explicit, no parameters are fitted, the extremal logarithmic divergence is imported from external results [9-11], and the conclusion is explicitly framed as a conditional obstruction. The universality scaling analysis is also useful. However, two load-bearing points prevent acceptance in the present form: the claimed finiteness of the Einstein-Hilbert bulk term (Eq. (8)) is incorrect for D>4, and the reverse triangle inequality for 'complexity distance' is asserted without definition, proof, or citation. The no-logarithmic-divergence core may survive a corrected calculation, but the exact-cancellation statement in the abstract and the central dynamical conclusion need substantial revision.","major_comments":[{"comment":"The claim that the Einstein-Hilbert bulk contribution is finite because r^{D-2} is integrable at r=0 ignores the divergence of the on-shell Ricci scalar. For Einstein-Maxwell-AdS, the trace of the Einstein equations gives R - 2Λ = 4Λ/(D-2) - 2G(D-4)/(D-2) q^2/r^{2D-4} + ... . Hence √-g(R-2Λ) ~ r^{D-2} r^{-(2D-4)} = r^{-D+2}, and the integral in Eq. (8) diverges as ϵ^{-(D-3)}, not finite. This invalidates the abstract's statement that EH, Maxwell, and GHY terms 'cancel exactly' and also undermines the coefficient K_D(γ) in Eq. (22). The no-logarithmic-divergence conclusion may survive, but the explicit cancellation claim requires recomputation.","section":"Eq. (8), 'WdW Action for Overcharged RN-AdS'"},{"comment":"The only bridge from Eq. (22) to the claim that the transition E→B_sprx is dynamically obstructed is the sentence 'the reverse triangle inequality for complexity distance then implies divergent relative complexity.' The paper never defines the pairwise complexity distance D(E,B_sprx), nor proves or cites any property that CA/CV complexity satisfies the triangle inequality. A divergent difference of complexities of formation relative to the AdS vacuum, |C(E)-C(B_sprx)|, does not by itself imply a divergent pairwise relative complexity. If complexity distance is not a metric, a finite circuit could connect the two states and the two-gap obstruction would not follow. This assumption is load-bearing and must be either proved, cited, or explicitly added as an axiom with a discussion of its status.","section":"'Extremal Barrier, the Third Law, and WCCC', sentence after Eq. (22)"},{"comment":"The paper repeatedly states that the obstruction is 'prescription-independent' and holds in both CA and CV. But the prescription independence is established only under the unproved metric assumption of the previous comment, and the CA calculation is quantitatively affected by the Eq. (8) error. The universality criterion based on f(r) ~ a r^{-p} is plausible, but it is used to conclude that GHY terms are the only possible power-law source; once the EH bulk is included, the scaling criterion must be extended to curvature invariants. As written, the universality claim is stronger than what the explicit calculations support.","section":"'Universality' and general framework"}],"minor_comments":[{"comment":"The LMPS counterterm written for pure global AdS contains Λ^{D-2} power-law terms; calling it 'finite' is confusing unless the vacuum subtraction is applied immediately. Please clarify that the finiteness refers to the vacuum-subtracted quantity.","section":"Eq. (14)"},{"comment":"The term I_GHY^{(r=0)} is written with r=0 but is later evaluated at r=ϵ. Use a consistent notation, e.g., I_GHY(r=ϵ), to avoid confusion about where the regulator surface sits.","section":"Eq. (7) and surrounding notation"},{"comment":"The symbol ΔC is used both for the difference of complexities of formation between two states and for a pairwise 'relative complexity.' These are conceptually different objects; define the latter explicitly and use distinct notation until the metric property is established.","section":"General presentation"}],"recommendation":"major_revision","confidential_remarks":"No conflicts of interest. The paper is ambitious and the topic is suitable for the journal. The two major issues are both reparable in principle: the EH calculation can be corrected, and the reverse-triangle assumption can be made explicit and conditional. If the author is unwilling to weaken the claim to a conditional statement, the manuscript would not be acceptable. I recommend major revision rather than rejection because the central no-logarithmic-divergence calculation likely survives correction, and the paper's conditional framing provides a reasonable path forward."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear —,\n\nThis is worth a read if you work on holographic complexity, and it deserves a serious referee, but the central inference needs more than the paper gives it.\n\nThe genuinely new part is the explicit WdW evaluation for overcharged RN–AdS in general D. The claim that the naked-singularity complexity carries no logarithmic divergence in any electromagnetic prescription is concrete and, as far as I can see, the CV calculation supports it. The two-gap framing—subextremal vs extremal, extremal vs naked singularity—is a nice unification of the third law and WCCC, and the paper is appropriately honest that this is conditional on CA/CV and finite complexity growth.\n\nThe soft spots are in proportion. First, Eq. (8) says the EH bulk is finite because r^{D-2} is integrable, but the Ricci scalar also diverges. For D>4, R ~ r^{-(2D-4)} from the Maxwell trace, so the integrand is r^{-D+2} and the EH term has a power-law divergence of the same order as the GHY and Maxwell terms. The exact cancellation claimed in the minimal prescription is therefore not established by the text. That might be repairable—the coefficients could cancel—but as written it's an error.\n\nSecond, and more serious, the step from 'no log divergence in C(B_sprx)' to 'divergent relative complexity ΔC(E,B_sprx)' relies on the reverse triangle inequality for complexity distance. This is asserted in one sentence, with no proof or citation. The inequality is true if state complexity is a distance from a fixed reference and the pairwise distance is a metric, but that is not a standard established property for either CA or CV complexity. Without it, a divergent difference in complexities of formation does not imply the states are far apart in complexity distance. This is load-bearing: the entire obstruction to crossing the extremal surface depends on that bridge.\n\nWhat holds up: the no-logarithmic-divergence core for the naked singularity looks right, the CV calculation is correct, and the universality criterion is useful. The citation pattern is appropriate, and the paper distinguishes its own conditional status clearly.\n\nBottom line: this is a serious proposal with a real calculation, and I would send it to a referee with a request to address the two issues above. The reverse triangle inequality needs a derivation or at least a citation to a model where it holds; the EH counting needs fixing. I'd bring it to a reading group first, though.","headline":"A genuinely new two-gap complexity proposal for WCCC, with a real calculation in the overcharged sector, but the reverse triangle inequality is doing heavy lifting without a proof, and the EH term is miscounted in D>4.","tokens_in":7765,"tokens_out":5594,"would_cite":true,"duration_ms":411224,"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":"Holographic complexity of formation of overcharged AdS spacetimes carries no logarithmic divergence, so relative complexity across the extremal surface diverges in both directions, making the naked-singularity sector unreachable in finite t","keywords":["holographic complexity","complexity=action","complexity=volume","weak cosmic censorship","third law of black-hole mechanics","Reissner-Nordström-AdS","Wheeler-DeWitt action","naked singularities"],"falsifier":"A specific test: compute the complexity of formation of overcharged RN–AdS in D=5 (near-singularity exponent p=4) using a differently shaped regulator at the singularity; the paper's claim predicts the WdW action is at worst power-law divergent in ϵ and never contains ln ϵ. If any valid regulator yields a logarithmic term, the 'no logarithmic divergence' input fails. Separately, constructing an explicit finite-complexity circuit in the dual CFT that prepares the extremal state from a subextremal state would falsify the first gap.","tokens_in":6796,"feed_emoji":"🕳️","tokens_out":7105,"duration_ms":56095,"temperature":0.7,"pith_summary":"The paper proposes that the third law of black-hole mechanics and weak cosmic censorship share a single origin in the geometry of quantum complexity. In the complexity=action and complexity=volume prescriptions, every subextremal black hole has finite complexity, the extremal black hole has logarithmically divergent complexity, and the overcharged naked-singularity sector has no logarithmic divergence. The paper computes the Wheeler–DeWitt action of overcharged Reissner–Nordström–AdS spacetimes and finds that near-singularity divergences are at most power-law, with a prescription-dependent coefficient that can even cancel. Because the extremal complexity divergence is logarithmic, no cancellation is possible, so the relative complexity across each adjacent gap diverges. If physical state preparation has a finite complexity growth rate, neither gap can be crossed in finite boundary time, making the naked-singularity sector computationally inaccessible.","feed_headline":"Complexity gaps block the route to extremality and naked singularities","feed_subtitle":"A holographic calculation finds divergent computational cost on crossing the extremal surface in either direction.","key_machinery":"The central machinery is the Wheeler–DeWitt (WdW) action in the complexity=action proposal and the maximal-volume functional in the complexity=volume proposal. For overcharged RN–AdS, the Maxwell bulk term and the Gibbons–Hawking–York boundary term at a small-radius regulator each diverge as ϵ^{-(D−3)}, but their combined coefficient K_D(γ) is prescription-dependent; the null, joint, and counterterm contributions are finite. No logarithmic divergence appears, so it cannot cancel the logarithmic divergence of the extremal complexity. In complexity=volume, the maximal volume near the singularity scales as O(ϵ^{2D−4}), removing any divergence. The near-singularity scaling f ~ a r^{-p} is then u","core_discovery":"The central claim is that two successive divergences in relative complexity protect the extremal surface from being crossed. The first gap separates any subextremal black hole (finite complexity) from the extremal black hole (logarithmically divergent complexity of formation). The second gap separates the extremal state from the naked-singularity sector: a direct on-shell Wheeler–DeWitt evaluation in general dimension D≥4 shows that the overcharged Reissner–Nordström–AdS complexity of formation contains only prescription-dependent power-law divergences or finite terms, never a logarithmic divergence. Since the extremal state's divergence is logarithmic, the difference in absolute complexitie","pith_inferences":["A direct test would be to compute the complexity distance, not just the absolute-complexity difference, between extremal and overcharged RN–AdS states in a concrete circuit model; a finite distance would undercut the two-gap conclusion even if the individual complexities diverge as stated.","The universality criterion suggests a classification program: for any static naked-singularity geometry with near-singularity scaling f ~ a r^{-p}, the absence of a log divergence in the WdW action is what protects the extremal surface; rotating Kerr–AdS is the natural next target, though its ring singularity changes the regulator geometry.","One could sharpen the obstruction by estimating the boundary time needed to approach the extremal surface as a function of the near-extremal parameter; if the required complexity gap diverges while the growth rate is bounded, the crossing time should diverge, giving a quantitative form of the barrier.","The paper's local near-singularity mechanism suggests computational censorship may extend beyond AdS asymptotics: because the relevant divergence is determined by local data near the singularity, similar obstructions might appear in other asymptotically flat or de Sitter settings where a holographic complexity notion exists."],"forward_implications":["A subextremal charged AdS black hole cannot be driven to extremality in finite boundary time under any finite-complexity-growth process, giving a complexity-geometric form of the third law.","Overcharging or overspinning gedanken experiments that attempt to reach a naked singularity must cross the extremal surface; in both directions the relative complexity diverges, so the naked-singularity sector is dynamically inaccessible.","The obstruction is prescription-independent: it holds in both complexity=action and complexity=volume and for any electromagnetic boundary prescription, since the naked-singularity sector never has a logarithmic divergence to cancel the extremal one.","The result generalizes to any static, spherically symmetric geometry with f ~ a r^{-p} near the singularity, covering RN black holes in D≥4 and broader multi-charge or dilatonic families with the same scaling.","If the paper is right, weak cosmic censorship is not primarily about the geometry of the endpoint but about the computational cost of preparing the endpoint state."],"fun_headline_variants":["Extremal black holes are computationally unreachable","Complexity gaps guard cosmic censorship","Naked singularities blocked by complexity divergence","Two complexity gaps shield cosmic censorship","Censorship by complexity: no route past extremality"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that complexity obeys the reverse triangle inequality, so a divergent absolute difference |C(E)−C(B_sprx)| implies a divergent complexity distance between the two states; the paper states this without proof or citation. The subsidiary finite-complexity-growth assumption is also needed to convert that divergence into an obstruction to finite-time crossing, and the paper explicitly frames the result as conditional evidence, not a proof of the full co","fun_headline_variants_meta":{"raw":{"variants":["Extremal black holes are computationally unreachable","Complexity gaps guard cosmic censorship","Naked singularities blocked by complexity divergence","Two complexity gaps shield cosmic censorship","Censorship by complexity: no route past extremality"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000193,"raw_usage":{"total_tokens":1192,"prompt_tokens":751,"completion_tokens":441,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":495,"completion_tokens_details":{"reasoning_tokens":375}},"tokens_in":495,"tokens_out":441,"duration_ms":4141,"temperature":1.0,"reasoning_tokens":375,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T15:59:51.462059+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A specific test: compute the complexity of formation of overcharged RN–AdS in D=5 (near-singularity exponent p=4) using a differently shaped regulator at the singularity; the paper's claim predicts the WdW action is at worst power-law divergent in ϵ and never contains ln ϵ. If any valid regulator yields a logarithmic term, the 'no logarithmic divergence' input fails. Separately, constructing an explicit finite-complexity circuit in the dual CFT that prepares the extremal state from a subextremal state would falsify the first gap.","supporting_citations":[],"review_version":3}