The Lyth bound is an inequality relating the tensor-to-scalar ratio r — a measure of primordial gravitational waves — to the total field excursion of the inflaton. Large tensor signals require super-Planckian field excursions. This is a problem: field excursions larger than the Planck scale take the inflaton into a regime where quantum gravity corrections should become important, making the effective field theory unreliable. The bound says: you can have detectable gravitational waves OR a controlled theory, but not both.
Note (2026-11-12; read against the paper's full text): the paper states the Lyth bound as holding within the standard single-field slow-roll paradigm, and says it demonstrates that a violation of the bound occurs naturally (its title speaks of bypassing the bound, its abstract of an evasion) when the observable gravitons are entangled with a hidden sector. The essay's statements that the bound was never violated and that it was derived under an assumption that the graviton state is pure are not the paper's wording; the paper contrasts its mixed state with the pure excited states of single-sector models. The paper also assumes that the hidden gravitational sector sits in a pure de Sitter background with its own cosmological constant and acts as a quantum reservoir for the observable tensor modes, and it describes the late-time noise enhancement in interferometers as a forecast of something that may appear.
Akama and Lin (arXiv:2602.20734) evade the bound by entangling gravitons across sectors. Two dynamically decoupled gravitational sectors — one observable, one hidden — become entangled during inflation. When the hidden sector is traced out, the reduced density matrix of the observable sector shows a parametrically enhanced tensor power spectrum. The enhancement comes not from the inflaton rolling further, but from quantum correlations with degrees of freedom we don't observe.
The mechanism is purely quantum mechanical. No modified gravity, no exotic fields, no classical dynamics. The observable tensor-to-scalar ratio exceeds 0.01 with sub-Planckian inflaton excursions. The bound is not violated — it never applied. The Lyth bound relates r to field excursion under the assumption that the graviton state is pure. A mixed state, produced by entanglement with a hidden sector, has no such constraint.
The signature is distinctive: oscillatory features in the power spectrum and a scale-dependent enhancement of the squeezed-limit bispectrum — a “quantum birthmark” that records the entanglement history. Additionally, the late-time stochastic noise in gravitational wave detectors may carry a residual enhancement from this primordial entanglement.
The general point: bounds derived under purity assumptions can be evaded by mixed states. Entanglement with unobserved degrees of freedom changes the accessible parameter space for the observed sector. What looks forbidden for pure states becomes available when the system is part of a larger entangled whole.