One shared state
Take the Bell pair Φ+, distributed before Alice decides how to measure her half.
Joint correlations depend on which bases are later compared.
Entangled observers can recover strong joint correlations after comparing records. Bob cannot learn Alice's freshly chosen basis from his own outcomes alone.
Ideal Bell-state model · No FTL experimentTake the Bell pair Φ+, distributed before Alice decides how to measure her half.
Joint correlations depend on which bases are later compared.
Without receiving Alice's measurement outcome, Bob has the same reduced density matrix.
The theoretical receiver-only probability is exactly 1/2 for either result.
Each trial assigns Alice a fresh random Z or X setting; Bob always measures Z. The settings are revealed only to group the experimental records afterward.
Diagnostic control
Bob's two outcomes remain equally likely, whatever basis Alice selected.
Finite empirical mutual information can be slightly positive under a truly independent null. This is not a signalling detection.
| Measurements | Alice = Bob |
|---|---|
| Both Z | — |
| Alice X, Bob Z | — |
The difference appears when two observers compare their records through an ordinary classical channel.
Without a classical side channel, Bob's local statistics cannot reveal Alice's choice. The second mode deliberately overwrites some of Bob's simulated results using Alice's setting. It should create measurable dependence—but that dependence is an injected conventional leak, not entanglement communication.
The browser uses deterministic pseudorandom Monte Carlo samples of an ideal mathematical state. No photon is transmitted. No relativistically separated measurement is carried out.
In the original quantum teleportation proposal, transmitting an unknown state requires both previously shared entanglement and an additional classical communication step [1].
Browser script and equations work offline. No external scripts or font requests. Input updates animate gently and respect reduced-motion preferences.