Holography vs. Scale Separation

Alek Bedroya, Paul J. Steinhardt · 2025

Read Holography vs. Scale Separation by Alek Bedroya, Paul J. Steinhardt (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-25313.txt.

Introduction

Alek Bedroya and Paul Steinhardt argue in 2025 that two prized ideas in string theory cannot both hold. One idea, holography, says a region of space can be described by a theory on its edge. The other seeks spaces with a huge gap between light and heavy particles. The pair show the edge theory can only split off from gravity if the inner field energy meets a strict test.

Well-known examples pass the test but give up the large gap. Popular setups that chase the gap, such as DGKT and KKLT, fail the test, the authors conclude. Their result narrows the search for realistic models of the universe built from strings.

Catalog record

CREATORAlek Bedroya, Paul J. Steinhardt
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE133 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-25313

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Holography vs. Scale Separation Alek Bedroya, Paul J. Steinhardt Abstract In this work, we point out a contradiction between holography and scale-separated AdS (i.e. parametrically large mass gap) in string theory, making the standard assumption that the holographic CFT describes the IR degrees of freedom on a brane that decouple from gravity. We show that the CFT can only decouple from gravity if the scalar potential in the dual AdS satisfies a certain criterion. Namely, there must exist a scalar field trajectory that follows the gradient of the scalar potential to the asymptotic region of the scalar field space in which limit V ∼ m α V\sim m^{\alpha} , where α ≤ 2 \alpha\leq 2 is a constant and m m is the mass scale of the lightest tower of states. This condition, which implies lack of scale-separation, is satisfied in the standard examples of AdS/CFT. However, proposed attempts at achieving scale separation, such as DGKT, employ scalar potentials that scale as m α m^{\alpha} with α > 2 \alpha>2 in the asymptotic region of moduli space. We therefore conclude that the CFT duals of DGKT vacua cannot exist in string theory. Barring fine-tuning, our conclusions apply to other Ricci-flat flux compactifications including the KKLT scenario which relies on scale separation to obtain a metastable de Sitter uplift. 1 Introduction UV completing a gravitational effective field theory is notoriously more difficult than for an ordinary quantum field theory. A key reason is that sufficiently massive states are black holes, which obey universal properties any UV completion must respect. Although a precise UV completion that can answer, for example, scattering experiments at arbitrarily high energies remains out of reach, string theory has provided compelling evidence that certain theories are UV completable. Two structural features underpin this success. First, string theory admits weak coupling limits, where the string coupling goes to zero and the theory can be controlled at arbitrarily high energies compared to the Planck scale. Second, string dualities relate these weakly coupled descriptions to each other, ensuring consistency across the full moduli space at finite couplings. Indeed, almost all precise results in string theory exploit these weak coupling regimes. Such limits exhibit universal behavior captured by the Swampland Distance Conjecture [ 1 ] . As the expectation value of a scalar field is taken to infinity, a tower of light states emerges with exponentially decreasing masses, and the quantum gravity cutoff falls accordingly [ 2 ] . In Minkowski space this provides a consistent weakly coupled regime, but in curved backgrounds the situation is more subtle. In AdS, moving to infinite distance in moduli space drives the cutoff below the curvature scale, invalidating the gravitational description. This is precisely the regime where the dual CFT becomes weakly coupled. Therefore, unlike in Minkowski space, there is no arbitrarily weakly coupled gravitational description of AdS. In the absence of infinite-distance limits where the gravitational description remains valid, there are two natural ways to generalize the Distance Conjecture to AdS spaces. One is to rely on the dual CFT [ 3 ] ; the other is to take limits in the space of AdS solutions, rather than in the moduli space of a single AdS. Since the moduli parametrize the space of solutions, the latter approach is natural. Following this logic, the authors of [ 4 ] conjectured that there exist constants A and \alpha , depending only on the spacetime dimension, such that a tower of states appears with mass spacing m such that \displaystyle Am^{\alpha}<|\Lambda_{\rm AdS}|\,, (1) in Planck units. A stronger form of the conjecture sets \alpha=2 , asserting the existence of a tower of light states satisfying \displaystyle Am^{2}\lesssim|\Lambda_{\text{AdS}}|\,. (2) This no-scale-separation conjecture states that the AdS scale cannot be parametrically separated from the KK scale 1 1 1 We comment on the difference between the version of the AdS Distance Conjecture used here and the strong AdS Distance Conjecture formulated in [ 4 ] in Section 2.3 . We approach this question holographically. Constructions with curved internal geometries naturally produce |\Lambda_{\rm AdS}|\sim m_{\rm KK}^{2} , since the curvature of the internal space contributes terms of order m_{\rm KK}^{2} to the potential that help stabilize the volume modulus. However, proposed scale-separated vacua such as DGKT [ 5 ] and KKLT [ 6 ] instead rely on Ricci-flat compactifications with fluxes. In these cases the scalar potential is not proportional to m^{2} , raising the possibility of scale separation. It is important to settle the validity of proposed scale–separated AdS vacua, or to understand precisely what goes wrong in those constructions. For example, KKLT AdS vacua serve as a stepping stone to KKLT de Sitter constructions, and the question of their validity carries important implications for the string landscape and its role in cosmology. By contrast, DGKT offers a more controlled scenario: unlike KKLT, which cannot be parametrically controlled (since the \alpha^{\prime} corrections cannot be made arbitrarily small), DGKT admits families of AdS solutions labeled by N that, in principle, allow for arbitrarily small Ultimately, the question of whether an AdS background is UV completable in quantum gravity is decided not by the gravitational description, but by the existence of its dual CFT. This is because the gravitational description can never become arbitrarily weakly coupled: as soon as one takes any coupling to zero, the appropriate description switches to that of the CFT. Therefore, while scale separation in terms of string coupling and internal volume may suggest parametric control, true UV completeness of an AdS background requires identifying the dual CFT. Our key observation is that the brane supporting the dual CFT can be described as a scalar field solution of the AdS effective action, where the scalar field runs to infinity. This perspective allows us to test whether problematic asymptotic behavior of the potential—specifically, behavior that develops scale separation in the interior—obstructs the existence of a decoupling limit. We show that for a brane theory to decouple from bulk gravity, the scalar potential at infinity must scale as |V|\gtrsim m^{2} . Otherwise, the worldvolume CFT cannot be realized as a brane theory in string theory. In particular, if DGKT AdS exists, then string theory would fail to contain its superconformal field theory (SCFT) dual. This idea, that varying scalar backgrounds encode essential consistency conditions beyond isolated critical points, has proven fruitful in constraining positive potentials as well [ 7 , 8 ] . Finally, we extend our methods to de Sitter. Interpreted in the context of dS/CFT [ 9 ] , our analysis shows that no Euclidean CFT dual to de Sitter can be realized on a spacelike brane in string theory. This conclusion is qualitatively different from the AdS case: while only scale-separated AdS vacua fail the decoupling test, no dual can exist for de Sitter. The organization of the paper is as follows. In Section 2 , we review what is known about infinite-distance limits in the string landscape and introduce the motivation for the absence of scale separation in AdS. There, we also provide a novel formulation of the no scale-separation condition that can be stated sharply within the gravitational description, without relying on the CFT. This formulation is based on the asymptotic behavior of the

Read the complete text