Can quantum entanglement send messages faster than light?

What did Einstein find so spooky?

In 1935, Einstein and two colleagues, Boris Podolsky and Nathan Rosen, argued that quantum theory must be leaving something out. The problem was entanglement. Two particles that have interacted can share one quantum state, the math physicists use to predict results. Measure one, and you learn at once about the other. The Royal Swedish Academy of Sciences’ popular science background on the 2022 Nobel Prize notes that Einstein talked about “spooky action at a distance.”

Einstein and his colleagues argued that a more complete description must exist. One way that could work is if the particles carried hidden instructions from the start, which physicists call hidden variables. Then nothing spooky would be needed. In 1964, John Bell showed how that idea could be tested. If particles carry built-in instructions, the matching between results can never go above a fixed limit, now called a Bell inequality. Quantum theory predicts more matching than that. Starting with a 1972 experiment by John Clauser and Stuart Freedman, tests kept siding with quantum theory. In 2022, Alain Aspect, John Clauser and Anton Zeilinger won the Nobel Prize in Physics “for experiments with entangled photons, establishing the violation of Bell inequalities and pioneering quantum information science.” Our story How Scientists Tested Einstein’s Hunch tells that history, and Did Einstein lose the dice argument? looks at what it means for his larger hope.

Why can’t the link carry a message?

Go back to Alice and Bob. To send a message, Alice would need to change something Bob can see on his own, like making his particle give “up” more often. She can’t. The physicist Asher Peres summed up what quantum theory says: two people sharing an entangled system cannot send any information at all using only their own local measurements, without sending real objects between them. Bob’s odds come out the same whatever Alice chooses to measure.

Here is an analogy, and it has limits. Imagine two sealed decks of cards that were shuffled together in a special way. Each of you draws cards and gets what looks like a random run. Only when you compare the runs do you see they match in a surprising pattern. You can’t pick which card you draw, so you can’t spell out a word. The analogy breaks down in one big way. Real cards carry their faces from the start, and Bell tests show entangled particles do not behave like that. What carries over is the key point: neither side controls its own results.

The Stanford Encyclopedia of Philosophy lists proofs of this by Eberhard (1978), by Ghirardi, Rimini and Weber (1980), and by Page (1982). In standard quantum theory, quantum correlations cannot be used to send signals faster than light. Caltech’s Science Exchange puts it simply: it is a common misconception that the particles talk to each other faster than light, and quantum physics cannot be used to send faster-than-light communications. Thomas Vidick, a professor of computing and mathematical sciences at Caltech, puts the idea in five words: “There can be correlation without communication.”

How far and how fast has it been tested?

• Across a building. In a 2015 test at the National Institute of Standards and Technology (NIST), the two detectors were 184 meters apart. Light needs 617 billionths of a second to cross that gap. Each pair was measured more than 40 billionths of a second before any light-speed signal could have linked the two sides.
• Between two villages. In 2008, Salart and colleagues ran a Bell test for more than 24 hours between two villages 18 kilometers apart, with the source in the middle. They reported that if some hidden influence really did link the particles, then under their assumptions it would have to move at least 10,000 times faster than light.
• Through space. In 2017, Yin and colleagues used a satellite to send entangled photons to two ground stations 1,203 kilometers apart, and the matching still broke the Bell limit, they reported in Science.

In all of these, the matching was there. These were tests of the matching, not attempts to send messages, and the proofs above say no message could ride on it.

What is the strongest objection?

The best objection goes like this. If Bell tests rule out built-in instructions, then surely something passes between the particles instantly. And if something passes instantly, why can’t we ride it?

Bell himself took the first half seriously. In 1964 he wrote that if extra hidden details were added to quantum theory to decide each result, without changing its predictions, the setting of one measuring device would have to influence the reading of another, however far away, and the influence would have to act instantly. Some versions of quantum theory do include something like that. The Stanford Encyclopedia points to the de Broglie-Bohm theory, in which the setting on one side can affect the result on the other, yet which still does not allow signaling. Whether Bell’s theorem shows a real clash between quantum theory and relativity depends, the Stanford Encyclopedia says, on what one takes relativity to require, and Bell’s own view changed over time.

So the deep question, what entanglement is, stays open. The practical question does not. Whatever links the particles, nobody can steer it to carry a chosen message. Peres added one more point. He showed that a particular kind of change to quantum theory, a “nonlinear” one, would let a message through. As the theory stands, it cannot.

What do people often get wrong?

“It’s like a pair of gloves.” Mail one glove of a pair to a friend and keep the other. Open your box, see a left glove, and you know your friend has the right one. The entanglement story explains why this picture falls short: gloves carry their answers from the start, and Bell tests show entangled particles show more matching than that allows. The gloves do get one thing right. Opening your box sends nothing to your friend.

“Quantum teleportation is faster-than-light travel.” It is not. In the 1993 proposal by Charles Bennett and colleagues, the sender makes a measurement and then must send the result to the receiver as ordinary information. Only then can the receiver rebuild the state. Zeilinger and his colleagues first did this in 1997, the Royal Swedish Academy of Sciences’ popular science background notes. Nothing arrives before the ordinary message does.

“Someone already built a quantum phone.” In 1981, the journal Foundations of Physics sent Asher Peres a paper by Nick Herbert called “FLASH,” which proposed a faster-than-light communicator. Peres wrote later that he knew it was wrong, but he recommended publishing it because he expected that finding the error would lead to real progress. Another referee, GianCarlo Ghirardi, recommended rejecting it. Herbert’s scheme leaned on copying the state of a single photon. In 1982, Wootters and Zurek showed in Nature that the rules of quantum theory forbid such copying. Peres writes that John Wheeler contributed their title, “A single quantum cannot be cloned.” In Peres’s telling, the failed phone helped lead to a real result, the no-cloning theorem.

“If it can’t carry messages, it’s useless.” Not so. Entangled particles can give two people matching random numbers to use as a key for coded messages. In Artur Ekert’s method, described in the Stanford Encyclopedia, the two people still compare notes over an ordinary public channel. The entanglement adds security, not speed.

How can you check this yourself?

• Ask what each side sees alone. When a video says entanglement sends signals, ask: could Bob learn anything before hearing from Alice? If not, no message has been sent.
• Look for the ordinary channel. Real protocols, such as teleportation and Ekert’s code keys, include a normal message. Find it in the description.
• Read the free popular science background. The Royal Swedish Academy of Sciences’ 2022 background uses black, white and gray balls to show what Bell tests rule out.
• Connect it to the speed limit. Our answer Why is the speed of light the universe’s speed limit? explains why signals and objects cannot pass light in the first place.

The answer stays the same after all this detail. Entanglement is real, tested and strange. It cannot send a message faster than light.