Does an infalling observer see vacuum at an old black hole's horizon?
In plain words
A 2012 argument showed that information preservation, normal physics outside the black hole, and a smooth horizon cannot all hold once a black hole has emitted half its entropy. Which assumption fails is still debated.
Precise statement
For a black hole past the Page time (when half the initial Bekenstein-Hawking entropy has been radiated), determine whether a freely falling observer crossing the horizon finds the local vacuum (mode occupation near zero for wavelengths much shorter than $r_s = 2GM/c^{2}$) or high-energy quanta up to the cutoff (a firewall). An answer specifies which assumption of the Almheiri-Marolf-Polchinski-Sully argument fails (unitary evaporation, local effective field theory outside the stretched horizon, a black hole that a distant observer describes as a quantum system with $\exp(S_{\mathrm{BH}})$ states, or a smooth horizon for an infalling observer), in a framework where the computation is controlled.
What would settle it
A calculation in a UV-complete theory, such as an evaporating AdS black hole with a known dual, of infalling observables at the horizon after the Page time.
Status in the literature
Unverified note
Entanglement-wedge reconstruction (2019-2020) supports a smooth horizon with the interior encoded in the radiation, but no derivation in a UV-complete theory settles it as of 2026.
Related problems
- More general than Is decoding interior partners from Hawking radiation exponentially hard?
- More general than Are typical black hole microstates horizonless geometries?
- More general than Must the black hole interior be reconstructed by state-dependent operators?