The generalized second law (GSL) — monotonic increase of the generalized entropy Sgen=⟨A⟩/4GN+SoutS_{\rm gen}=\langle A\rangle/4G_N+S_{\rm out} of a black hole horizon — was given an operator-algebraic foundation by Chandrasekaran, Penington and Witten (CPW): in the strict large-NN limit of AdS/CFT, the entropy of a Type II∞_\infty crossed-product algebra of single-trace operators equals the generalized entropy of the black hole bifurcation surface, and, for horizons that have first equilibrated, the GSL follows from monotonicity of algebra entropy under trace-preserving inclusions. CPW could not extend this to horizon cuts that are out of equilibrium: they found no Type II algebra whose entropy equals the generalized entropy of such a cut, and conjectured that new ingredients — perhaps entropies of subspaces of observables rather than of algebras — are required. Later work has resolved substantial stationary sectors — Type II entropy constructions for general subregions and all Killing-horizon spacetimes, the GSL for arbitrary cuts of Killing horizons, an exact but modified GSL with dynamical cuts in perturbative gravity, and cut-by-cut Type II∞_\infty edge-mode algebras valid locally on non-stationary horizons — but a global structure whose entropy reproduces the generalized entropy of arbitrary non-equilibrated horizon cuts, with the second law following from its monotonicity, remains open.

Background

The generalized second law (GSL) of black hole thermodynamics states that the generalized entropy of a black hole horizon never decreases. The generalized entropy of a horizon cut (a spatial cross-section of the horizon) is Sgen=⟨A⟩4GN+Sout,S_{\rm gen}=\frac{\langle A\rangle}{4G_N}+S_{\rm out}, the expectation value of the cut's area functional divided by 4GN4G_N plus the entropy of the exterior quantum fields. The GSL was proven semiclassically by Wall, using monotonicity of the quantum relative entropy between successive horizon slices, for rapidly changing fields and arbitrary horizon slices. What has been missing since Bekenstein and Hawking is "a precise microscopic principle" behind the GSL: an identification of SgenS_{\rm gen} with the ordinary entropy of some well-defined subsystem, from which the second law would follow as a structural monotonicity property rather than a gravitational computation. Chandrasekaran, Penington and Witten (CPW) supplied such an identification in the strict large-NN limit of AdS/CFT (the holographic duality between quantum gravity in an asymptotically anti-de Sitter spacetime and a conformal field theory on its boundary), for a black hole in equilibrium — a two-sided eternal black hole in the thermofield-double (Hartle–Hawking) state, or a one-sided black hole formed from collapse and allowed to equilibrate for a parametrically long time. The relevant operator-algebraic notions: a von Neumann algebra is a collection of observables closed under addition, multiplication, and adjoint, and it is a Type II∞_\infty factor when it admits a trace — a nontrivial linear functional that is cyclic (invariant under cyclic permutations) and satisfies Tr⁡(e−βH)=∞\operatorname{Tr}(e^{-\beta H})=\infty, yet whose value on finite projections can be normalized so that states have finite, well-defined entropy. CPW crossed the algebra of single-trace CFT operators (the boundary operators dual to weakly coupled bulk fields, closed at leading order in 1/N1/N) with the time-translation group generated by the boundary Hamiltonian H0H_0: the resulting crossed-product algebra AR=Amatter⋊σR\mathcal{A}_R=\mathcal{A}_{\rm matter}\rtimes_{\sigma}\mathbb{R} is a Type II∞_\infty algebra on which the renormalized time function acts, and the entropy of a semiclassical state on AR\mathcal{A}_R equals the generalized entropy of the black hole bifurcation surface (the cross-section where the past and future branches of the horizon meet). Because a trace-preserving inclusion of Type II algebras cannot decrease the entropy of a state, the GSL for equilibrated horizons — and for operator insertions separated by parametrically large but sub-scrambling times (longer than the thermalization time but shorter than the scrambling time — the time of order the black hole's inverse temperature times its entropy, after which perturbations are spread over all degrees of freedom), where appropriate Type II subalgebras still exist — follows without any gravitational computation. That derivation, however, needs the horizon to have equilibrated: the crossed product is built from the time-translation Hamiltonian, which is available only in a stationary or equilibrated limit. CPW state: "Unfortunately we have not been able to extend this line of argument to horizon cuts that are out of equilibrium, because we were not able to find a Type II algebra whose entropy is equal to the generalized entropy of such a cut." They conjecture that "some new ingredient beyond Type II von Neumann algebras is necessary," offering one candidate: the generalized entropies of out-of-equilibrium cuts may describe "the entropies of subspaces of observables, rather than actual algebras" — subspaces not closed under multiplication, for which CPW define a candidate entropy as a supremum, over all states on the ambient algebra that agree with the given state on the subspace, of the state's entropy (their eq. 4.10) — a definition motivated by the coarse-grained apparent-horizon entropies of Engelhardt–Wall and of Bousso–Chandrasekaran–Shahbazi-Moghaddam — but they cannot show that this supremum reproduces the generalized entropy without restricting it to semiclassical states. A closely related puzzle: black holes formed from collapse never equilibrate (the Unruh vacuum, i.e. the state of a collapse-formed hole which is empty on the past horizon, is not KMS — Kubo–Martin–Schwinger, meaning thermal with respect to horizon time), so the CPW framework does not even nominally apply to astrophysical black holes. Subsequent work substantially advanced the stationary sectors. Jensen–Sorce–Speranza constructed Type II factors for arbitrary subregions in the GN→0G_N\to 0 limit, assuming states with geometric modular flow (the modular flow of a state — the canonical state-dependent one-parameter evolution attached to an algebra and state by Tomita–Takesaki theory — being generated by a spacetime vector field); Kudler-Flam–Leutheusser–Satishchandran proved a structure theorem for linear fields on any spacetime with a Killing horizon (the null surface on which a stationarity-generating Killing vector field becomes null) in stationary (not necessarily KMS) states; Faulkner–Speranza derived the GSL for arbitrary cuts of Killing horizons from crossed-product algebras, leaving open beyond-leading-order corrections and genuinely dynamical horizons; Kirklin proved an exact but modified GSL (with a free-energy term) in perturbative gravity using idealized dynamical cuts; Chandrasekaran–Flanagan constructed, for each cut of a possibly non-stationary event horizon, an edge-mode Type II∞_\infty factor (edge modes: auxiliary degrees of freedom attached to the cut that restore gauge invariance) whose entropy matches SgenS_{\rm gen} to first order near that cut — but the algebras are defined cut-by-cut, to first order in a local Rindler (perturbative) neighborhood of a Killing background, so they do not cover genuinely non-Killing or non-linearized horizons. No existing construction yields a global, non-perturbative-in-the-nonequilibrium structure whose entropy is the generalized entropy of arbitrary out-of-equilibrium cuts.

Problem Statement

For an arbitrary horizon cut of a black hole that has not equilibrated — including black holes formed from collapse and horizons crossed by infalling matter in arbitrary states, settings in which no KMS or stationary state with geometric modular flow on the horizon is available — identify the mathematical structure (a Type II von Neumann algebra, an entropy functional on subspaces of observables that are not closed under multiplication, or some new ingredient) whose renormalized entropy equals the generalized entropy Sgen=⟨A⟩/4GN+SoutS_{\rm gen}=\langle A\rangle/4G_N+S_{\rm out} of that cut, and show that the generalized second law for such cuts — non-decrease of SgenS_{\rm gen} along the horizon — follows from a structural monotonicity property of this structure, rather than from a semiclassical gravitational computation. The problem therefore asks for: (i) the algebraic or generalized-algebraic object attached to out-of-equilibrium cuts and the limit in which it is defined (large-NN, GN→0G_N\to 0, or perturbative); (ii) the proof that its entropy, suitably renormalized, equals SgenS_{\rm gen} of the cut, including the treatment of the area operator and bulk entropy; (iii) the proof of the monotonicity property from which the GSL for such cuts follows, controlling or explicitly bounding quantum corrections beyond leading semiclassical order; and (iv) verification of the limiting cases: agreement with the CPW equilibrated-horizon derivation, with Wall's semiclassical GSL for arbitrary slices, and with the Faulkner–Speranza result for cuts of Killing horizons. Further open issues from the source paper are not part of this problem's completion boundary.

References

  1. Venkatesa Chandrasekaran, Geoff Penington, Edward Witten, "Large N algebras and generalized entropy", JHEP 04 (2023) 009, DOI:10.1007/jhep04(2023)009, arXiv:2209.10454, https://arxiv.org/abs/2209.10454
  2. Venkatesa Chandrasekaran, Roberto Longo, Geoff Penington, Edward Witten, "An algebra of observables for de Sitter space", JHEP 02 (2023) 082, DOI:10.1007/jhep02(2023)082, arXiv:2206.10780, https://arxiv.org/abs/2206.10780
  3. Edward Witten, "Gravity and the crossed product", JHEP 10 (2022) 008, DOI:10.1007/jhep10(2022)008, arXiv:2112.12828, https://arxiv.org/abs/2112.12828
  4. Aron C. Wall, "A proof of the generalized second law for rapidly changing fields and arbitrary horizon slices", Phys. Rev. D 85 (2012) 104049, DOI:10.1103/PhysRevD.85.104049, arXiv:1105.3445, https://arxiv.org/abs/1105.3445
  5. Netta Engelhardt, Aron C. Wall, "Decoding the Apparent Horizon: A Coarse-Grained Holographic Entropy", Phys. Rev. Lett. 121 (2018) 211301, DOI:10.1103/PhysRevLett.121.211301, arXiv:1706.02038, https://arxiv.org/abs/1706.02038
  6. Raphael Bousso, Venkatesa Chandrasekaran, Arvin Shahbazi-Moghaddam, "From black hole entropy to energy-minimizing states in QFT" (arXiv preprint title: "Ignorance is Cheap: From Black Hole Entropy To Energy-Minimizing States In QFT"), Phys. Rev. D 101 (2020) 046001, DOI:10.1103/PhysRevD.101.046001, arXiv:1906.05299, https://arxiv.org/abs/1906.05299
  7. Kristan Jensen, Jonathan Sorce, Antony Speranza, "Generalized entropy for general subregions in quantum gravity", JHEP 12 (2023) 020, DOI:10.1007/jhep12(2023)020, arXiv:2306.01837, https://arxiv.org/abs/2306.01837
  8. Jonah Kudler-Flam, Samuel Leutheusser, Gautam Satishchandran, "Generalized black hole entropy is von Neumann entropy", Phys. Rev. D 111 (2025) 025013, DOI:10.1103/PhysRevD.111.025013, arXiv:2309.15897, https://arxiv.org/abs/2309.15897
  9. Thomas Faulkner, Antony J. Speranza, "Gravitational algebras and the generalized second law", JHEP 11 (2024) 099, DOI:10.1007/jhep11(2024)099, arXiv:2405.00847, https://arxiv.org/abs/2405.00847
  10. Josh Kirklin, "Generalised second law beyond the semiclassical regime", JHEP 07 (2025) 192, DOI:10.1007/JHEP07(2025)192, arXiv:2412.01903, https://arxiv.org/abs/2412.01903
  11. Mohd Ali, Vardarajan Suneeta, "A local Generalized second law in crossed product constructions", arXiv:2404.00718, https://arxiv.org/abs/2404.00718
  12. Geoff Penington, Elisa Tabor, "The algebraic structure of gravitational scrambling", arXiv:2508.21062, https://arxiv.org/abs/2508.21062
  13. Stefan Hollands, Robert M. Wald, Victor G. Zhang, "Entropy of dynamical black holes", Phys. Rev. D 110 (2024) 024070, DOI:10.1103/PhysRevD.110.024070, arXiv:2402.00818, https://arxiv.org/abs/2402.00818
  14. Venkatesa Chandrasekaran, Éanna É. Flanagan, "Subregion algebras in classical and quantum gravity", arXiv:2601.07915, https://arxiv.org/abs/2601.07915
  15. Avinandan Mondal, Kartik Prabhu, "Perturbative semiclassical entropy of dynamical black holes", arXiv:2603.02320, https://arxiv.org/abs/2603.02320
  16. Marc S. Klinger, Jonah Kudler-Flam, Gautam Satishchandran, "Generalized Entropy is von Neumann Entropy II: The complete symmetry group and edge modes", arXiv:2601.07910, https://arxiv.org/abs/2601.07910
  17. Andreas Blommaert, Jonah Kudler-Flam, Erez Y. Urbach, "Absolute entropy and the observer's no-boundary state", arXiv:2505.14771, https://arxiv.org/abs/2505.14771
  18. Ronak M. Soni, "A Type I Approximation of the Crossed Product", arXiv:2307.12481, https://arxiv.org/abs/2307.12481
  19. Eugenia Colafranceschi, Xi Dong, Donald Marolf, Zhencheng Wang, "Algebras and Hilbert spaces from gravitational path integrals: Understanding Ryu-Takayanagi/HRT as entropy without AdS/CFT", JHEP 10 (2024) 063, DOI:10.1007/JHEP10(2024)063, arXiv:2310.02189, https://arxiv.org/abs/2310.02189
  20. Vijay Balasubramanian, Tom Yildirim, "The Nonperturbative Hilbert Space of Quantum Gravity With One Boundary", arXiv:2506.04319, https://arxiv.org/abs/2506.04319
  21. Shadi Ali Ahmad, Ro Jefferson, "Crossed product algebras and generalized entropy for subregions", SciPost Phys. Core 7 (2024) 020, DOI:10.21468/SciPostPhysCore.7.2.020, arXiv:2306.07323, https://arxiv.org/abs/2306.07323