What physical phenomena require gravity itself to be quantized that stochastic coupling cannot address?

This is an excellent question that cuts to a genuine tension in contemporary quantum gravity research. The preprints here actually illuminate the boundary quite sharply.

What stochastic gravity cannot do

Stochastic coupling—where gravity remains classical but couples to quantum matter via fluctuating or averaged fields—hits a fundamental wall when you need gravity itself to *entangle* quantum systems.

The clearest evidence comes from a direct comparison study: when researchers modeled Newtonian gravity three ways, they found that semiclassical gravity (where the gravitational field is determined by the quantum average of matter) and stochastic gravity (where the field has classical noise added) both fail to entangle spatially separated quantum bodies, even though quantum mechanics says they should become entangled through gravitational interaction [AG-2025.10-576]. Only when the gravitational tidal field itself is quantized—treated as a quantum object with its own degrees of freedom—does entanglement actually emerge [AG-2025.10-576].

This matters because entanglement is the signature of quantum mechanics. If two particles can become entangled through their mutual interaction, that interaction must itself be quantum-mechanical.

The deeper conceptual issue

There's a foundational problem lurking underneath. Standard semiclassical gravity violates a basic principle: linearity of quantum dynamics. When you have classical geometry coupled to quantum matter via the expectation value (the "mean field"), you're essentially asking the quantum system to evolve both linearly *and* be influenced by a classical field that depends on its own state. This creates conceptual monsters—like the famous Schrödinger cat problem in gravitational contexts [AG-2025.01-695].

Stochastic versions add noise to patch this, but the underlying issue remains: you still have a classical external degree of freedom (the geometry) telling the quantum system how to evolve, which breaks the symmetry between system and environment that quantum mechanics requires [AG-2025.01-695].

Phenomena requiring quantized gravity

1. Gravitationally induced entanglement between macroscopic objects — the litmus test [AG-2025.10-576]. If gravity can entangle two quantum systems, gravity must be quantum.

2. The spectrum of fundamental particles and mixing angles — why do electrons, muons, and tau leptons have the masses they do, and why do they mix in the specific patterns we observe? One proposal argues these emerge only when spacetime geometry itself is dynamical and quantized, not as a background [AG-2024.07-055].

3. Quantum superpositions of distinct spacetime geometries — in regions where the geometry itself must exist in superposition (near black hole singularities, or the very early universe), stochastic fluctuations of a classical field cannot capture the quantum coherence needed [AG-2024.07-055].

4. High-order quantum interference in gravitational systems — certain multi-path interference experiments can only be explained if gravity mediates genuine quantum coherence, not classical noise [AG-2024.07-055].

What stochastic models *can* do

This doesn't mean stochastic gravity is useless. One elegant framework shows that a purely stochastic Poisson process, with no external quantum machinery, naturally generates Dirac-like wave equations for both electromagnetism and linearized gravity when properly rescaled by the Planck length [AG-2025.08-267]. The quantum character emerges *from the stochastic process itself*, not imposed externally. This is a remarkable result but applies to linearized gravity, not the full theory.

In short: stochastic coupling fails precisely where gravity must answer the question "can you create quantum coherence?"—and the archive shows the answer is no, not without quantizing gravity itself.