The entropy of a semiclassical closed baby universe in AdS/CFT is revisited: can it have a large number of states rather than a one-dimensional Hilbert space? Results on Haar random coding show a breakdown of complementary recovery: logical operators cannot be recovered from subsystems. An interpretation is proposed: the baby universe represents logical degrees of freedom inaccessible from a single boundary, with pseudorandom correlator dynamics. The cloning and singularity paradoxes for microstates are examined; both are resolved because no single observer can access the states — reflecting random coding complementarity. Observers arise naturally: the heavy operator that creates the geometry sets an observer-dependent microstate.
According to one physical hypothesis, our voluminous world might be a projection from a flat surface. From such a Universe, tiny closed bubbles — baby universes — sometimes pinch off. For a long time, they were thought to be empty, devoid of internal diversity.
It turned out that the information inside them is scattered, like shards of a broken mirror. Each external observer sees only the reflection in their own shard — which means the degree of disorder is high, and the full picture is inaccessible to anyone. This explains why no one can peer into a black hole or grasp what's going on in curved spacetime during collapse.
This approach resolves old paradoxes like the disappearance of information. Hawking was the first to ponder these baby universes, while Maldacena and Susskind built a holographic model.
🎯 The idea of tiny universes first emerged in Stephen Hawking's work, but was long considered mathematical exotica.
🎬 In science fiction, similar ideas play out in the movie 'Interstellar,' where other dimensions opened up behind a bookshelf.