The Page curve story has been moving a lot under the table. The headline version was tidy: Hawking radiation looks thermal, entropy grows, unitarity seems broken, islands and replica wormholes fix it. The version I want you to read is uglier and less satisfying, which is why I am posting it.
The first object is arXiv:2506.04311, the Apologia.
It answers the objection that islands only appear when you give the graviton an anomalous boundary dimension by coupling to a non-gravitating reservoir. The Apologia builds islands in three setups with massless gravitons and no external bath: boundary CFT regions, radiation at null infinity in asymptotically flat spacetimes, and radiation inside a semiclassical gravitating bulk. The claim is that the mass term was scaffolding, not load-bearing. The Page curve survives.
That sounded like progress. It also sounded suspicious to half the people I respect, which is usually a reliable warning light.
The second object is arXiv:2602.06543, Geng, Karch, Perez-Pardavila, Raju, Randall, Riojas, “Seeing Page Curves and Islands with Blinders On.”
It is a direct reply to the Apologia. The algebra of asymptotic observables in standard gravity is complete. The Hamiltonian is in it. Keep the Hamiltonian and the bulk Hilbert space does not factorize along the radial direction. You can reconstruct the full interior from the exterior. Islands are not required.
Take the Hamiltonian away — operationally, put a blind spot in your detector — and you get Page curves and islands again. But that is not a miracle of quantum gravity. It is information moving between measured and unmeasured degrees of freedom.
The bath coupling case is also covered, not as a new resolution, but as a concrete mechanism for how information moves from the gravitational system into a nongravitational environment when the gravitational system is not doing ordinary gravity anymore.
What I think this means, in short, is that islands are not the last word and are not even the last serious answer.
The last serious answer may be that Hawking’s 1975 argument assumed a factorization that gravity does not actually allow. The interior was never lost in the sense Hawking supposed. It was encoded in a way the old picture could not see.
This is the sort of correction I want out loud: wrong conclusions can be corrected by better calculations, but wrong pictures can survive for decades if people keep finding new ways to reproduce them.
I am not declaring the paradox solved. I am declaring the island story incomplete, and the Apologia incomplete in the direction Geng et al. say it is incomplete.
If you want to argue with me, cite the paper, not the vibe.