[ExI] Google’s Willow Quantum Chip: Proof of the Multiverse?

[Adrian Tymes]

The AI explanation failed to address the question.  Under MWI, the worlds
are separate after splitting, with no way to interact.  And yet, this MWI
explanation for this requires them to interact.  By what means does the
information get from one world to another after splitting?

On Mon, Oct 13, 2025 at 1:34 PM John Clark <johnkclark at gmail.com> wrote:

> On Mon, Oct 13, 2025 at 10:12 AM Adrian Tymes via extropy-chat <
> extropy-chat at lists.extropy.org> wrote:
>
> *> Even MWI would not seem to explain it, as I read it, though I may
>> be misunderstanding.*
>
>
> *The thought experiment is that you are the foreman in a factory that
> makes bombs that are so sensitive if a single photon hits the fuse it would
> go off, but your manufacturing process is not perfect and sometimes you
> make a defective fuse, however your customers demand perfection, they
> insist that each fuse be tested so it could be guaranteed to work. This
> would seem to present an impossible problem because if you test it with a
> photon and it goes off you know it did work but now you don't have the bomb
> any longer so you can't sell it. But actually there is a way to do this and
> it involves a Mach–Zehnder interferometer (which has half silver mirrors)
> and whether or not which-way information has been preserved because if you
> have which-way information about a photon in a Mach–Zehnder interferometer
> it will behave differently than it would if you had no information about
> which path the photon took. The device looks like this:*
>
> [image: CC42E819-DD07-404F-91BE-7FA8ADF3CC84.png]
>
> *I then got lazy and asked GPT the following question:*
>
> *"How could the Many Worlds interpretation explain how the Elitzur–Vaidman
> bomb tester could work?" And this is what GPT said:*
> *===*
>
> "The Elitzur–Vaidman bomb tester is a quantum thought experiment using an
> interferometer to identify live bombs without detonating them. In a classic
> Mach–Zehnder setup, a single photon is sent into a beam splitter so that it
> travels along *both* paths in superposition. A sensitive bomb sits on one
> path: if a photon hits it, the bomb explodes (destroying the photon); if
> not, the photon continues. Remarkably, there are cases where a detector
> click reveals the bomb is live even though it did *not* explode – seeming
> to detect the bomb “without touching” it. As Elitzur and Vaidman explain,
> this is a new nonlocal effect of quantum mechanics: one can “ascertain the
> existence of an object in a given region of space without interacting with
> it” In practice the outcomes are: the bomb explodes (50%); the photon
> emerges at one detector (25%); or the photon emerges at the other detector
> (25%). A click at the normally silent detector (say “D”) certifies the bomb
> is live even though it has not been triggered. This counterfactual result
> relies on quantum interference and poses a conceptual puzzle: *how can
> the photon learn about the bomb’s presence without interacting with it?*
> Many-Worlds Interpretation (MWI) of the Bomb Tester
>
> In the Many-Worlds Interpretation (MWI) of quantum mechanics, the photon’s
> superposition leads to *real branching*: every possible outcome occurs in
> some “world” of the universal wavefunction, with no collapse. Applied to
> the bomb tester, the incoming photon at the first beam splitter creates two
> coherent branches (or worlds). In one branch the photon travels along the
> lower path toward the bomb; in the other branch it takes the upper path. If
> the bomb is live, the *lower-path branch* causes the bomb’s detector to
> trigger and explode. In MWI this branch becomes a “world” where the photon
> and bomb are both destroyed (a macroscopic outcome). Meanwhile, the *upper-path
> branch* is a world where the photon avoided the bomb entirely and
> continues on. Crucially, these branches *initially interfere* until a
> measurement or irreversible event occurs. Because the bomb exploded in the
> lower branch, it decouples (or decoheres) that branch from the upper
> branch. The photon in the upper branch then reaches the second beam
> splitter without its partner, so interference is *lost*. So whenever the
> photon hits the first mirror two worlds are effectively created and they
> continue to interfere until a photon measurement is made in either of
> them. In practice, this means our branch (photon on the upper path) is
> still influenced by its counterpart that went toward the bomb. If that
> counterpart is absorbed by the bomb the link between our world and the
> parallel one is disrupted and interference ceases, and it is possible for
> our own photon to hit the normally dark detector.
>
>    -
>
>    *Branch 1 (Bomb branch):* Photon hits the bomb. The bomb’s detector
>    absorbs the photon and explodes. This is a measurement event that yields
>    one real outcome in that world (bomb destroyed, photon gone)
>    -
>
>    *Branch 2 (Safe branch):* Photon takes the alternate (upper) path.
>    Because the twin photon in Branch 1 was absorbed, there is no two-path
>    interference. The surviving photon now behaves like a classical particle at
>    the second beam splitter, reaching detectors C or D with equal probability.
>    If it hits Detector D (the normally “dark” port), this *in our branch*
>    signals “live bomb”.
>
> From the perspective of *our branch*, the photon never touched the bomb
> yet we infer its presence. In MWI this is not magic but the result of the
> parallel world’s event. As Vaidman notes, “obtaining information about the
> region without anything being there” is only possible because “the photon
> which we sen was there, but in another world”. In fact, three worlds emerge
> in the experiment: one with explosion, and two where the photon is detected
> at the two detectors. In the world where detector D clicks (informing us
> the bomb is live), the bomb did not explode *in that world*. But that
> world’s information is possible because in the explosion world the photon
> *did* touch the bomb. In Vaidman’s words, “we get information about the
> object without touching it in one world but we ‘pay’ the price of
> interacting with the object in the other world”. In short, under MWI the
> counterfactual measurement simply means the photon’s interaction happened
> in a parallel branch, not in our own.
> Branching Structure with a Live Bomb
>
> The presence of a *working* bomb changes the branching structure of the
> quantum state. If the bomb were a *dud* (transparent), the two photon
> paths recombine and interfere perfectly at the second beam splitter, always
> yielding one deterministic detector (say C) and a single effective branch
> (the photon’s amplitude for Detector D cancels). But a *live* bomb
> introduces irreversible events that split the worlds:
>
>    -
>
>    *Splitting at the First Beam Splitter:* When the photon hits the first
>    beam splitter, it enters a superposition of “lower path” and “upper path.”
>    In MWI this creates two coherent branches (“worlds”).
>    -
>
>    *Bomb Interaction (Conditional):* In the lower-path branch, the photon
>    encounters the bomb. If the bomb is live, the photon is absorbed and the
>    bomb explodes. This is a macroscopic interaction that effectively
>    “measures” that branch. In MWI terms, that branch now has a definite
>    outcome (explosion) and becomes decohered.
>    -
>
>    *Surviving Branch:* In the upper-path branch, the photon never
>    interacts with the bomb. However, because the other branch just exploded,
>    the upper-branch photon no longer has an interference partner. At the
>    second beam splitter it thus has a 50/50 chance of emerging at either
>    detector C or D, just like a classical particle.
>    -
>
>    *Outcome Worlds:* If Detector D clicks, that branch (our branch) knows
>    “bomb is present.” If Detector C clicks (or if the bomb explodes), those
>    are the other branches. Importantly, whenever Detector D fires without
>    explosion, *another world* witnessed the explosion.
>
> Thus, a live bomb guarantees a *multi-world outcome structure*. The
> bomb’s “measurement” of the photon forces branching into at least three
> final outcomes: (i) explosion (no photon detected), (ii) photon detected at
> C, and (iii) photon detected at D. In the exploded world the photon was
> absorbed; in the detected worlds the photon survived. If the photon is
> detected at D (a sign of the bomb), our branch succeeded in “testing” the
> bomb without destruction. But the logical cost is borne by the explosion
> branch: the interference link was broken by the photon’s absorption there.
> Our photon’s otherworldly shadow photon carried information; if the “OK”
> interference signal does not come, it means the shadow photon has fallen,
> that is was absorbed in the parallel world.
> “Detecting” the Bomb Without Detonating in MWI
>
> In practice, when a detector click (e.g. at D) tells us the bomb is live
> yet it hasn’t exploded in our world, what does this mean philosophically in
> MWI? It means that the measurement result we see is correlated with an
> event in another branch. In our branch the bomb remains intact, but the
> detection of the photon at D happened *because* in the exploded branch
> the photon was absorbed. We are thus inferring the bomb’s existence
> *indirectly*, via the missing interference. There is no mystery
> communication: the “signal” comes from the fact that the branches’
> evolution was linked until the bomb’s interaction. Once the explosion
> occurred (in the other branch), the branches decohered and our photon’s
> behavior changed.
>
> In short, *“interaction-free” becomes almost a misnomer: globally, there
> was an interaction (in another world), but locally in our branch the bomb
> was not touched. MWI resolves the paradox by distributing reality: the
> photon did interact in the universe, but not in the branch we experience.*
> There is no nonlocal action at a distance in the many-worlds view – just
> the usual quantum evolution with branching. All the possible outcomes
> “happened” somewhere, so detecting the bomb unscathed simply means we find
> ourselves in the branch where the interaction happened elsewhere."
>
> ==
>
> *John K Clark*
>
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20251013/9bdad5f7/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CC42E819-DD07-404F-91BE-7FA8ADF3CC84.png
Type: image/png
Size: 53265 bytes
Desc: not available
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20251013/9bdad5f7/attachment-0001.png>