I had always read that you cannot use entangled quantum states to transfer information, certainly not faster than light. This article claims otherwise:
The summary paragraph reads:
Please tell me it’s wrong.
Curious why you want it to be. It most likely has some issue like even though the data arrives faster the the SOL, the ability to decode the q-bits need to be transmitted conventionally. But it seem you want it to be wrong.
This article tells the same story and doesn’t use the phrase “faster than the speed of light” once. So …
Back in the 20th century when they invented quantum teleportation, one of those people explained to me how it made use of certain entangled quantum states. You still need to transmit “classical” information to complete the measurement.
If this new scheme is doing something substantially different, what is it?
ETA the quantum measurement part is just a type of measurement, nothing is being sent superluminally either (the particles are entangled of course)
Science topic + The Daily Mail + “Einstein is wrong!” = the triple-threat recipe for you should completely ignore everything the article says.
The underlying science itself is quantum teleportation, for which there are numerous still-correct explanations online about how information transfer does and doesn’t work.
Take two tokens, one red and one blue, and put them into two envelopes. Shuffle the envelopes so you don’t know which is which. Keep one and mail the other to your cousin in Bemidji. Agree that you will open the envelopes at some predetermined precise time using prearranged precise timekeeping devices. When your cousin opens it and sees a blue token, they instantly know that you – in that very instant – are seeing a red token, assuming the plan went off without hiccups. This non-quantum non-spooky experiment would be called “faster than light communication” by the logic used in these sorts of articles.
Or just one of those factors. The middle one.
Yes I know I should have. But I clicked on it and I would still like some who actually understand the science to come along and say “Balderdash”.
As to why I want it to be wrong, it is that I have read in many places that information transfer is limited to SOL. Now I obviously don’t expect the Daily Mail to understand the science, but they must have gotten that claim from somewhere.
To be more explicit: the problem is that you still have to send the envelopes. You still have to send the entangled particles somewhere else, and that necessarily means that said particle will travel subluminally. So then the information itself also cannot travel faster than the speed of light.
Besides sending the envelope with the token somewhere, you also have to have the two observers far apart from each other who must make some pre-arranged plan, via conventional communication, about how and when to open the envelopes, and what meaning to give to the results they observe. That is necessarily part of the experiment. The superluminal component of the experiment can happen but cannot be given any meaningful interpretation (IOW no information is passed) without the prior conventional communication. That is what kanicbird meant by “the ability to decode the q-bits need to be transmitted conventionally.”
What I want to know is, how do you even get the two observers to agree on when to “simultaneously” open their envelopes and observe their q-bits. We already understand that two observers, distant from each other, can never truly agree on what time-events are simultaneous. How do they deal with that?
That’s not quite the issue. The thing is, rather, this:
In order to complete a quantum teleportation protocol, (classical) information has to be sent to the receiving party, to let them know which of a certain set of transformations to perform on their end. In the simplest case, you need two bits of information, to choose between four different final transformations. The interesting thing is, however, that it would need far more than two bits of information to fully specify a generic quantum state—that’s where the (somewhat misleading) name of ‘teleportation’ comes from.
So perhaps to make things a little more clear, the protocol is as follows:
- Two parties, A and B, share an entangled two-qubit state, consisting of qubits qA and qB (this can be exchanged a long time in advance, provided both have the resources to keep their part of the state coherent for a long enough time)
- A possesses a qubit in an arbitrary state qX
- A performs a measurement (a so-called Bell measurement) on their two qubits, qA and qX (this will break the entanglement between qA and qB, entangling qA and qX instead)
- This measurement yields two bits of information
- A sends a message containing those two bits—either 00, 01, 10, or 11—to B; this can obviously only be done at light speed, maximally
- Based on A’s message, B performs one of four transformations on qB (well, one of those is ‘do nothing’, so either that or one of three transformations)
- Afterwards, qB is now in the state qX was before A’s Bell measurement
So, as already pointed out correctly, it is not possible to send information faster than light with this method.
What this experiment shows is that non-locality is not preserved, which means that some sort of information did transfer faster than the speed of light. But, it is information about the entangled states of the particles, that they are able to communicate their states to eachother.
This has already been more or less proven by Bell’s Inequality experiments. It shows that it is not “hidden information” that is revealed, as the envelope analogy says, but that the states are actually determined upon measurement.
While it does show that there is some sort of underlying nature to entangled particles that allows them to maintain their connection and “instantly” resolve into their respective states, it still does not allow any sort of actual communication.
There is no way to tell if the transmitted particle has been measured, or what that measurement was, until you actually receive the other half of the data through conventional means. There is no way to influence the state of the particle that you are to observe without breaking that entanglement.
If there was a way to entangle particles, and then make one of those particles be in a particular state without breaking that entanglement, that would give you the ability to transfer data faster than light. It is almost entirely certain that this is not possible. So the “information” that is “transmitted” is inherently random.
The unhackable networks part is actually valid, though. For that, you take your data as 1’s and 0’s and then you do measurements on your end. If the particle in question is spin up, you flip the bit, if it’s spin down, you don’t. Then you transmit your data. Only the person who has received the other entangled bits will be able to decode it. These entangled particles cannot be intercepted or spied upon without breaking the entanglement.
This was less an experiment learning new physics or more about the underlying nature of entanglement, and more an engineering issue of being able to transmit these entangled particles without them losing their entanglement. It’s impressive what they can do, but it is still not violating the causality of the speed of light.
It is interesting from a cryptographic standpoint, in that the largest challenge of cryptography is how to securely transmit your key, something that has never been 100% resolved, there is always some way of copying or intercepting a conventional key, if someone is determined enough. This, so long as the “no cloning theorem” stays true, is actually pretty much 100% secure. (Assuming of course, that the transmitting and receiving systems themselves are secure, which is a whole different issue, but at least while in transit it is.)
Hmmm… but you’ve deliberately stripped out the spookiness here. The predetermined red & blue tokens are hidden variables, which we know is not the correct account of entanglement.
What happens is more along these lines:
You start with two head/tail coins, but they are magic coins. You shake them up together in a magic box, then you separate them, and send one coin to your cousin in Bemidji, keeping the other coin here. Along with the magic coin, you send instructions to your cousin for the precise time at which you will both flip your coins. When you flip your coins, neither of you can control the outcome, the head/tail outcome is unkown beforehand, it’s truly random. But the magic is that both coins always give the same result. If your cousin happens to flip a head, he knows instantaneously that your coin has come up heads too.
There is no way that you can transfer information that you want to transfer instantaneously, you cannot (for example) tell your cousin in Bemidji that the weather here is sunny or cloudy, because you can’t control the head/tail outcome of your own coin flip. But the instant that you know the outcome of your own coin flip, you do instantaneously know the outcome of your cousin’s coin flip is the same. So it does seem to require that the magic coins are spookily communicating with each other.
Good analogy, but the one thing that I want to point out is that there is no need for simultaneity.
You could do the same, but either of you could open your envelopes and flip your coins at any time. Then you will know what the other one will get when they do the same.
You could even flip your coin, then call up your cousin, and tell him that when he flips his coin, you know how it will come up.
It is more interesting when you do flip the coins close enough together that you can show that they could not have communicated at or slower than light speed, but it is not necessary.
I’m not sure if I’m parsing this correctly, but I think this gets things backwards: it’s only if there is ‘hidden information’ that is being revealed upon experiment that one needs non-local influences in quantum mechanics; if observables don’t have predetermined values (over and above those predicted by the quantum formalism), you also don’t need non-locality.
You can account for entanglement in terms of hidden variables, but then, you need for those hidden variables to non-locally influence one another (as in Bohmian mechanics). Contrariwise, and somewhat against what you seem to be saying later on, you do not need non-local influences if you don’t insist that all quantities in a quantum measurement have pre-defined values—i. e. if you don’t postulate hidden variables.
No experiment today has been able to rule out the possibility that quantum mechanics is a local theory; it’s only if one adds values that the quantum formalism doesn’t predict—i. e. hidden variables—that one finds that there must be non-local influences.
I’m going to admit that this is a bit above my paygrade, but I’m pretty sure that that’s not the case here.
If there are hidden variables, if the particle that I send to you actually is spin up, and we just don’t know it until we measure it, then that doesn’t require non-locality.
If there are no hidden variables, if the particle that I send to you is not spin up or spin down, and is actually in a superposition of states corresponding to the particle I hold, which is not just revealed but determined only upon measuring them, then that implies non-locality.
That was pretty much the entire point of Bell’s Inequality, to prove that there were no hidden variables. The alternative is quite literally called “local hidden variables theory”. Though the experiments to show Bell’s Inequality are not perfect, they say that either realism or locality must be incorrect.
What exactly are you meaning by “hidden variables”, if you are not saying that the state of the entangled particles are predetermined before they are measured?
To my cousin in Bemidji, I would mail a curling stone*. But I already understand Bell’s theorem that says this analogy is faulty since the fixed color of the stones is hidden information. The point is that the entangled particles don’t have a quantum state until it is actually measured. That Aspect has also been verified experimentally. I still wonder where the Mail reporter got the idea from. But I guess we can stop here.
*I never heard of Bemidji until it turned out they were curling enthusiasts.
My take on incorrect science clearly caused confusion. My entire point was that even if you do something that has nothing interesting going on at all, science reporting at the low level of this article could still conclude faster-than-light communication. I had hoped this point would have been made clear with:
That’s as deep as this “Einstein disproved!” logic tends to even go.
Envelopes, the USPS, and small Minnesota towns do not make for an analogy with quantum teleportation, nor did I intend them to.
Actually, this is not true.
This has been solved 100% for all practical purposes by public-key cryptography, which is routinely used by everyone on the internet.
In theory you could crack a key pair, but even the fastest supercomputers that exist would take millions of years to do so. If computers ever did become fast enough, you could simply increase the key size.
The assumption that there are hidden variables is the assumption of ‘realism’ in Bell experiments. So the experimental violation of a Bell inequality means that at least one of the following assumptions must be wrong:
- Realism/hidden variables (every observable has a predetermined value that is uncovered by experiment)
- Locality (neither measurement settings nor measurement outcomes influence one another instantaneously)
- Freedom of choice (each experimenter is able to freely choose—say, based upon a random coin throw—which measurement to perform)
- Single-valuedness (every experiment has one and only one outcome—this is the assumption violated by many-worlds type interpretations, and is usually unstated)
Thus, if you say that there’s no hidden variables, you can reconcile Bell inequality violations with a theory that obeys locality, freedom of choice, and single-valuedness; contrarily, if you say that there are hidden variables—that every experiment uncovers a pre-existing reality that is ‘hidden’ to the quantum formalism, such as a value for momentum for a system in a position-eigenstate—, then you have to give up one of those. Bohmian mechanics gives up locality, many-worlds theories give up single-valuedness, and exotic things like 't Hooft’s cellular automaton theory give up freedom of choice to introduce ‘superdeterminism’. (Bell himself, by the way, was an advocate of the existence of hidden variables, subscribing to Bohmian mechanics.)
Typically, this is glossed by saying that Bell experiments refute ‘local realism’—that is, you either have to give up locality, or the idea that there are predetermined values for every observable quantities. But really, Bell inequalities are just necessary conditions for all variables in an experiment to have a joint probability distribution. Their violation then means that there is no joint distribution; this could be the case because they influence one another (non-locality), because you’re not getting a fair sample (superdeterminism), or because there aren’t actually any predetermined values in the first place (things like Copenhagen quantum mechanics, or QBism, and the like).