Wormhole Cosmic Censorship Conjecture

• Wormhole cosmic censorship conjecture is a framework where the wormhole throat isolates a naked ring singularity from both null and timelike geodesics.
• It reinterprets Penrose's censorship by replacing event horizon shielding with a throat-induced barrier that prevents causal curves from reaching the singularity.
• Analytical proofs in Kerr-like phantom and Einstein–Maxwell–Dilaton wormholes utilize Hamilton–Jacobi separation and effective potential analysis to ensure causal inaccessibility.

Searching arXiv for the cited wormhole cosmic censorship papers and related review. Wormhole cosmic censorship conjecture denotes a class of proposals in which a spacetime singularity remains causally inaccessible not because it is enclosed by an event horizon, but because the intrinsic geometry of a wormhole throat prevents causal geodesics from reaching it. In the literature assembled around Kerr-like phantom wormholes and later exact Einstein–Maxwell–Dilaton wormholes, the conjecture is formulated for ring singularities that are naked in the horizon sense yet are nevertheless “untouchable,” meaning that no null or timelike geodesic from the asymptotically flat exterior reaches the singular set (Matos et al., 2012). The conjecture is therefore a wormhole analogue of weak cosmic censorship rather than a statement about strong cosmic censorship, global hyperbolicity in full generality, or the generic evolution problem (Águila et al., 2018).

1. Definition and relation to Penrose cosmic censorship

Penrose’s cosmic censorship conjecture is presented in two standard forms. Weak cosmic censorship states that singularities are hidden from distant observers by event horizons, whereas strong cosmic censorship concerns inextendibility of the maximal Cauchy development and the preservation of deterministic evolution. Wormhole cosmic censorship modifies only the shielding mechanism: the singularity is not hidden by a horizon, but is instead causally disconnected by the wormhole throat itself (Bixano et al., 18 Feb 2026).

In the formulation introduced for Kerr-like wormholes supported by phantom matter, the central claim is that a naked ring singularity can be “fully protected by the intrinsic properties of a wormhole’s throat,” so that no future-directed null geodesic from either asymptotically flat exterior region reaches the singularity (Matos et al., 2012). In the later analytical treatment, this statement is sharpened to causal inaccessibility for both null and timelike geodesics in the slowly rotating limit, yielding a wormhole version of weak cosmic censorship: there is no event horizon, but there is still no causal geodesic from either asymptotic region that can touch the ring singularity (Águila et al., 2018).

A recurring misconception is to equate the conjecture with ordinary weak cosmic censorship. The distinction is explicit in the cited works: ordinary weak cosmic censorship uses horizon shielding, whereas wormhole cosmic censorship uses throat-induced geodesic exclusion. Another misconception is to regard the result as a version of strong cosmic censorship. The 2018 analytical proof explicitly states that it does not address strong cosmic censorship; its conclusion is geodesic inaccessibility of the naked singularity, not a full theorem on global hyperbolicity or generic inextendibility (Águila et al., 2018).

2. Kerr-like phantom wormholes and the original conjecture

The conjecture was first advanced in the study of a Kerr-like wormhole supported by a phantom scalar field, an exact solution of the Einstein–phantom field equations (Matos et al., 2012). In Boyer–Lindquist-like coordinates $(t,l,\theta,\phi)$, the metric is

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$

with

$$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$

and

$$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$

The parameters $l_0$ and $k_1>0$ set the throat scale and the scalar-field strength, respectively. The spacetime has two asymptotically flat regions, $l\to\pm\infty$, connected by a throat at $l=0$ (Matos et al., 2012).

The ring singularity is located at

$$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$

equivalently $r=l_0,\theta=\pi/2$ after introducing $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$0. The same analysis shows that curvature invariants diverge at this locus. The metric function $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$1 has one-sided discontinuous behavior at the ring: approaching along $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$2 from $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$3 gives $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$4, while $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$5 gives $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$6. This one-sided structure is part of the local geometry responsible for the geodesic deflection mechanism (Matos et al., 2012).

The matter source is a massless phantom scalar. In the conventions used there, the Einstein equations are written as

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$7

with $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$8, and the solution is tied to the metric function through

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$9

Because the field is phantom, the null energy condition is violated, which the paper presents as part of the matter model sustaining the throat (Matos et al., 2012).

3. Geodesic barrier and the 2012 null-geodesic analysis

The initial support for the conjecture was geodesic. For null geodesics, the Hamiltonian is

$$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$0

with $$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$1, and the stationary-axisymmetric symmetries imply conserved quantities $$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$2 and $$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$3 (Matos et al., 2012).

For the sector $$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$4 and normalized $$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$5, the null constraint reduces to

$$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$6

This relation functions as an effective barrier for radial-polar motion. Far from the throat, $$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$7, but near the ring singularity the discontinuity of $$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$8 makes the right-hand side discontinuous at $$\Delta = l^2 + l_0^2 \cos^2\theta,\qquad \Delta_1 = l^2 + l_0^2,\qquad K = \Delta / \Delta_1,$$9. The consequence described in the paper is that freely falling null trajectories develop turning points and are deflected away from the ring rather than reaching it (Matos et al., 2012).

The analytic near-ring expansion was performed in the southern hemisphere, $$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$0 and $$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$1. Using the geodesic equations, the paper derives exact forms for $$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$2 and $$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$3 when $$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$4, and from the null constraint obtains

$$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$5

Because $$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$6 for $$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$7, the right-hand side is negative in the domain of validity, and there is no real solution for $$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$8. This is the central 2012 analytic statement: no null geodesic coming from the exterior can attain the ring singularity (Matos et al., 2012).

The same work emphasizes that there are no event horizons. Since $$f = \exp[ -(k_1/2\Delta)\cos\theta ] = e^{-\lambda}.$$9 with $l_0$0 except at the singular locus, and $l_0$1 everywhere, the metric does not contain a Kerr-like horizon structure. The singularity is therefore locally naked in the horizon sense, but globally unreachable by exterior null geodesics. This is the original sense in which the throat acts as censor (Matos et al., 2012).

4. Analytical proof in the slowly rotating Kerr-like phantom wormhole

The 2018 work provides an analytical proof of cosmic censorship for a Kerr-like phantom wormhole in the slowly rotating limit (Águila et al., 2018). In Boyer–Lindquist-like coordinates $l_0$2, the metric is

$l_0$3

with

$l_0$4

$l_0$5

The throat is at $l_0$6, and the ring singularity at $l_0$7. With $l_0$8, $l_0$9, and $k_1>0$0, the throat becomes the two-surface $k_1>0$1, while the ring singularity is $k_1>0$2 (Águila et al., 2018).

The slowly rotating limit imposes

$k_1>0$3

together with $k_1>0$4. To first nontrivial order,

$k_1>0$5

The domain of validity of this first-order approximation is controlled by

$k_1>0$6

for $k_1>0$7. The paper states that this excludes only a tiny neighborhood of the ring singularity while still reaching sufficiently near it to reveal the relevant repulsive effect (Águila et al., 2018).

The geodesic Hamiltonian is

$k_1>0$8

with $k_1>0$9 for null and $l\to\pm\infty$0 for timelike geodesics. The Killing vectors $l\to\pm\infty$1 and $l\to\pm\infty$2 imply

$l\to\pm\infty$3

Using the Hamilton–Jacobi ansatz

$l\to\pm\infty$4

the equations separate and yield a fourth conserved quantity, a Carter-like constant $l\to\pm\infty$5 (Águila et al., 2018).

The separated first integrals are

$l\to\pm\infty$6

where

$l\to\pm\infty$7

$l\to\pm\infty$8

Allowed motion requires $l\to\pm\infty$9 and $l=0$0. At the ring singularity,

$l=0$1

with $l=0$2 (Águila et al., 2018).

This identity is the core of the proof. To pass through the throat at $l=0$3, one needs $l=0$4, equivalently $l=0$5. But then automatically $l=0$6. Hence any admissible geodesic that opens the throat is excluded from $l=0$7, the equatorial plane at the throat where the ring singularity resides. The point $l=0$8 is therefore forbidden for all null or timelike geodesics compatible with throat traversability (Águila et al., 2018).

The stronger discriminant-based sufficient conditions are written in terms of

$l=0$9

and lead to the explicit set

$$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$0

Under these inequalities, $$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$1 for all $$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$2, while $$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$3 has allowed bands $$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$4 that necessarily avoid $$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$5. For timelike geodesics, $$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$6 requires $$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$7, so $$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$8. The result is simultaneous traversability between the two asymptotic universes and causal inaccessibility of the ring singularity (Águila et al., 2018).

5. Hidden symmetry, causal structure, and comparison with Kerr-type singularities

A technically important feature of the 2018 proof is hidden symmetry. In the slowly rotating limit, the inverse metric separates as

$$\Delta = 0 \quad\Longleftrightarrow\quad l=0,\ \theta=\pi/2,$$9

with

$r=l_0,\theta=\pi/2$0

and explicit $r=l_0,\theta=\pi/2$1, $r=l_0,\theta=\pi/2$2 given in the paper. This yields a rank-2 Killing tensor

$r=l_0,\theta=\pi/2$3

whose contraction reproduces the Carter-like constant $r=l_0,\theta=\pi/2$4. The censorship mechanism is therefore encoded not merely in a heuristic effective potential, but in exact Hamilton–Jacobi separability and the associated polynomial structure of the geodesic flow (Águila et al., 2018).

The same work also constructs the causal picture on fixed $r=l_0,\theta=\pi/2$5 and fixed $r=l_0,\theta=\pi/2$6 slices. The induced two-dimensional metric can be rendered conformally flat by introducing

$r=l_0,\theta=\pi/2$7

so that

$r=l_0,\theta=\pi/2$8

Compactification with

$r=l_0,\theta=\pi/2$9

gives a Penrose diagram with two asymptotically flat ends and no horizons; the throat is at $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$00, while the ring $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$01 is removed from the geodesic domain by the condition $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$02 whenever $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$03 (Águila et al., 2018).

The contrast with the negative-mass Kerr black hole is explicit. Both spacetimes lack event horizons and contain a naked ring singularity. In the negative-mass Kerr solution, however, geodesics can penetrate the ring region into $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$04, and closed timelike curves can occur near the ring. In the Kerr-like phantom wormhole, by contrast, the ring lies at the throat and is shielded by the effective polar barrier

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$05

whenever the throat is open. The ring is thus untouchable by admissible null or timelike geodesics (Águila et al., 2018).

A further conceptual distinction concerns topological censorship. Because the phantom source violates the null energy condition, classical topological censorship theorems based on the averaged null energy condition are inapplicable to the 2012 and 2018 phantom-supported wormholes. The censorship argument itself does not rely on those theorems; it is geodesic and separability based (Águila et al., 2018).

6. Extensions to exact Einstein–Maxwell–Dilaton wormholes

Later work generalized the theme from phantom-supported Kerr-like wormholes to exact stationary, axisymmetric Einstein–Maxwell–Dilaton wormholes with a ring singularity that remains causally disconnected (Bixano et al., 3 Aug 2025). In Weyl–Lewis–Papapetrou form, the metric is written as

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$06

with spheroidal coordinates

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$07

For the exact solution highlighted in the review, one has

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$08

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$09

with $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$10 given explicitly as a rational function of $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$11 (Bixano et al., 18 Feb 2026).

In the 2025 space-time analysis, Boyer–Lindquist-type coordinates are introduced through

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$12

and the ring singularity is again at

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$13

The throat is the two-sphere $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$14, equivalent to $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$15, and the paper concludes that the throat “lines” or “dresses” the ring singularity. The throat is analytic for $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$16, but closes at the equator $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$17, so that no causal curve can pass through the equatorial section to meet the ring (Bixano et al., 3 Aug 2025).

The corresponding Carter–Penrose construction fixes $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$18 and then $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$19, giving

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$20

with $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$21, $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$22, followed by the compactification $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$23, $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$24. Since $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$25, the conformal metric is regular away from the ring. The resulting diagram contains two asymptotic regions glued across the throat and a forbidden region containing the ring singularity. For $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$26, the throat crossing occurs at a finite offset; for $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$27, one has

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$28

meaning the throat closes precisely in the equatorial plane. This is the geometrical realization of the “untouchable naked singularity” described in that paper (Bixano et al., 3 Aug 2025).

The 2026 review expands the framework to exact Einstein–Maxwell–Dilaton solutions and formulates the conjecture in terms of the singular set $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$29: for every future-directed causal curve whose past endpoint lies in the domain of outer communication of either asymptotic region, $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$30. More strongly, all future-directed causal geodesics either pass through the throat to the other asymptotic region at finite affine parameter or asymptotically approach the throat without intersecting $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$31. In that review, the throat is said to “suck in” geodesics before the ring can be encountered (Bixano et al., 18 Feb 2026).

These later solutions are also relevant because they introduce branches in which the null energy condition can be satisfied. Substituting the exact solution into Einstein’s equations yields

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$32

so that

$$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$33

Accordingly, the review states that the null energy condition holds for the dilaton branch $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$34 and fails for the phantom branch $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$35. This suggests that wormhole cosmic censorship, as a causal-disconnection mechanism, is not restricted to NEC-violating matter models, at least within the exact solution families presented there (Bixano et al., 18 Feb 2026).

7. Limitations, scope, and open directions

The strongest analytical proof currently summarized in this literature is limited to the slowly rotating Kerr-like phantom wormhole. Its proof relies on the slowly rotating limit $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$36, the first-order approximation for $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$37, and Hamilton–Jacobi separability in that regime. The authors state explicitly that a fully nonperturbative proof for arbitrary spin $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$38 remains open, and that extending the argument to faster rotation would require controlling divergences in the zero-angular-momentum angular velocity and reanalyzing separability or identifying a different constant of motion (Águila et al., 2018).

The scope of the geodesic results must also be distinguished carefully. The 2012 analysis concentrates primarily on null geodesics, although the generalized constraint accommodates timelike motion. The 2018 paper upgrades the statement to both null and timelike geodesics, but still within the slowly rotating approximation. The later exact Einstein–Maxwell–Dilaton constructions supply Penrose-diagrammatic and effective-potential evidence that the ring singularity is causally disconnected, together with references to geodesic completeness results, but the review presents these as part of an ongoing program rather than as a universal theorem for all rotating wormhole geometries (Matos et al., 2012, Bixano et al., 18 Feb 2026).

The conjecture also does not collapse into a single claim about horizons, topology, or causality violations. In these works, no event horizon is present in the relevant wormhole solutions. In the 2025 exact EMD example, a Killing horizon associated with $$ds^2 = - f dt^2 + (K/f) dl^2 + (\Delta_1/f) [ K d\theta^2 + \sin^2\theta d\phi^2 ],$$39 may occur, with zero surface gravity, but it is explicitly stated not to be an event horizon and not to determine causal accessibility to the ring (Bixano et al., 3 Aug 2025). Likewise, while some of the later papers discuss global hyperbolicity of the exterior two-region manifold and the existence of a global Cauchy surface, the conjecture itself is about the causal inaccessibility of the singularity, not the full strong cosmic censorship program (Bixano et al., 3 Aug 2025).

Taken together, the cited literature defines wormhole cosmic censorship as a censorship mechanism in which a ring singularity is naked in the absence of an event horizon but excluded from the causal domain of all admissible null or timelike geodesics from the asymptotic exterior. In the phantom Kerr-like wormholes of the original papers, the exclusion follows from a throat-induced geodesic barrier and, in the slowly rotating case, from a fully separated Hamilton–Jacobi analysis. In the later exact Einstein–Maxwell–Dilaton wormholes, the same idea is recast geometrically: the throat lines the singularity and closes at the equator, so the singular ring is causally disconnected even though no event horizon exists (Águila et al., 2018).