What happens to the probe's recoil energy when the superposition collapses—does it violate energy conservation during the measurement?

This is a genuinely subtle question—and the preprints here take it seriously, which is encouraging.

The short answer: energy is conserved overall, but the question turns on what "overall" means [AG-2024.04-2319, AG-2024.01-1918].

Here's the puzzle. When a probe measures a superposed source (say, a mass in two places at once), standard quantum mechanics says the probe will recoil differently depending on which branch "wins" the collapse. Statistically, energy balances across many measurements. But what about *one individual measurement*? Does the probe gain energy that came from nowhere?

The key insight is that collapse is not a single event in isolation—it's part of a *measurement chain* [AG-2025.04-504]. The probe interacts with the source, then a detector measures the probe, and only at that final step does the wavefunction collapse. When you account for this chain properly, energy conservation holds in each individual outcome, not just on average [AG-2024.04-2319]. The recoil energy doesn't vanish; it's distributed consistently with whatever conserved quantity (momentum, angular momentum) is involved, as long as you track the entire system—source, probe, *and* measuring apparatus.

One elegant proposal replaces the standard Schrödinger equation with stochastic, nonlinear terms that enforce conservation *locally* during measurement, sidestepping the problem altogether [AG-2024.05-1843, AG-2024.03-2000]. These terms act like an energy barrier: they suppress macroscopic superpositions automatically, so collapse happens in a way that respects energy conservation from the start.

The upshot: there's no energy violation if you're careful about what counts as "the system"—and modern approaches are designing collapse mechanisms that make this automatic [AG-2024.01-1918].