Could the correspondence fail only for specific spacetime dimensions or boundary CFT structures rather than universally?

Your question asks whether the AdS/CFT correspondence (or more broadly, holographic duality) might break down selectively—in certain dimensions or under particular boundary conditions—rather than being universal. This is a genuinely sophisticated question, and the archive shows that researchers are indeed exploring exactly this kind of dimensional and structural dependence.

Dimension-dependence and modified structures

The correspondence does appear sensitive to spacetime dimension in specific ways. [AG-2025.01-691] studies holographic correlators in two-dimensional boundary and crosscap CFTs, computing stress-tensor correlators in detail. The fact that researchers perform explicit calculations at lower dimensions suggests the correspondence may have unique features there—two dimensions is special because CFTs are far more constrained. Similarly, [AG-2026.04-745] works explicitly in three bulk dimensions when computing entanglement entropy in their holographic construction, treating lower-dimensional cases as controlled calculational arenas.

Boundary structure dependence

The archive shows the correspondence is sensitive to boundary conditions. [AG-2025.01-691] examines both tensionless and tensionful end-of-the-world branes in BCFT, and finds that brane tension deforms correlators—meaning the holographic map itself depends on boundary structure. This is not a breakdown, exactly, but it shows the correspondence is not universal *across* different boundary CFT architectures.

More radically, [AG-2026.02-334] questions whether the correspondence holds with full conformal invariance at all. It shows that in non-unitary field theories (conjectured dual to cosmological correlators), scale invariance does *not* necessarily imply conformal invariance. Modified gravity theories like Einstein-Aether theory lack bulk boost symmetry, generating only scale invariance on the boundary. This is a structural failure of the usual correspondence: the bulk and boundary symmetries decouple, suggesting the dS/CFT (or cosmological) version of the correspondence may not exist universally.

Constraints on bulk-to-boundary maps

[AG-2026.01-916] shows that bulk-to-boundary correlators in Poincaré-invariant theories are highly constrained by fall-off conditions at null infinity. Crucially, they identify a critical fall-off index Δ = 1 below which behavior changes qualitatively. This hints that the correspondence may have sharp phase transitions or structural changes depending on which operator dimensions (and hence which CFT primaries) you consider.

The bottom line

Rather than a universal law, the evidence suggests the holographic correspondence is a toolkit that adapts to structure: it works robustly in AdS with standard conformal boundaries, becomes subtle in two dimensions, breaks down when you lose conformal symmetry (as in cosmological duality), and has hidden critical thresholds in operator dimensions. Whether this constitutes "failure" depends on how universally you expected it to work in the first place.