Quantum Entanglement
Quantum entanglement is a physical link between two or more qubits whose measurement outcomes stay correlated no matter how far apart the qubits are, so measuring one instantly tells you something about the other even across a great distance. It’s one of the core ingredients that gives a quantum computer its power, because it lets many qubits act as one connected system instead of a pile of independent bits. It also underpins one family of QKD protocols. The one thing entanglement does not do, despite how it’s often described, is carry a message faster than light.
The short version:
- Entanglement links qubits so their measurement results are correlated, and that correlation holds however far apart the qubits are.
- It’s a resource quantum computers spend. Along with superposition, entanglement is what lets qubits perform coordinated multi-qubit operations, and it’s widely held to be necessary for a quantum computer to outrun a classical one.
- It’s also a defensive tool. Entanglement-based QKD (the E91 protocol) turns the physics of entanglement into an eavesdropper alarm.
- It does not enable faster-than-light communication. Each side of an entangled pair sees only random outcomes, and pulling any usable message out of the correlation still needs an ordinary channel capped by the speed of light.
- Entanglement is real and heavily verified in the lab. The distance from there to a code-breaking quantum computer is still large, and it’s mostly an engineering problem of keeping many qubits entangled without them falling apart.
An everyday way to picture it
Imagine you have a pair of magic coins. You give one to a friend who flies to Tokyo while you stay in New York. Whenever you flip yours and get heads, you learn that your friend’s coin, flipped at the same moment, came up tails, every single time, with no wire between them and no delay. That perfect matching of two random-looking outcomes is the flavor of entanglement. The catch that keeps it honest is control. You can’t force your coin to land heads; you only get to flip and see what comes up. So even though the two coins are perfectly linked, you have no way to steer which face appears, which means you can’t spell out a message to Tokyo. The link is real, the coordination is instant, and yet nothing you can steer travels between the coins.
What is quantum entanglement?
Quantum entanglement is a state of two or more quantum particles (for our purposes, qubits) in which you can’t fully describe one qubit on its own, because its properties are bound up with the others. Measure one member of an entangled pair and you immediately constrain what the other will show when it’s measured, and that holds no matter how far apart they are. The physicist John Bell put this on a rigorous footing in 1964, showing that these correlations are stronger than any theory based on the two particles carrying hidden, pre-agreed instructions could produce. That’s the mathematical fingerprint that separates genuine entanglement from ordinary correlation.
The idea traces back to a 1935 paper by Einstein, Podolsky, and Rosen, who found the phenomenon so strange they used it to argue quantum mechanics had to be incomplete. Einstein famously called it “spooky action at a distance.” Decades of experiments have since come down firmly on the other side: the correlations are real, and the 2022 Nobel Prize in Physics went to Alain Aspect, John Clauser, and Anton Zeilinger for the experiments that nailed this down and violated Bell’s inequalities.
A pair of maximally entangled qubits is often called a Bell pair or an EPR pair, and it’s the simplest building block of everything below.
Sources: Einstein, Podolsky, Rosen, “Can Quantum-Mechanical Description of Physical Reality Be Considered Complete?”, Physical Review 47, 1935, journals.aps.org; J. S. Bell, “On the Einstein Podolsky Rosen Paradox,” Physics 1, 1964, cds.cern.ch/record/111654; The Nobel Prize in Physics 2022, nobelprize.org.
How does entanglement work?
You don’t need any math to hold the mechanism. Think of it in three moves:
- You link the qubits. A quantum computer applies a sequence of operations (see quantum gates) that couples two qubits together. After that, the pair has a single joint description, and you can’t fully describe either qubit on its own. The only complete description is of the pair as one connected object.
- The link encodes a rule, not a value. Before anyone measures, neither qubit has a settled answer. What the entanglement fixes is the relationship between their eventual answers, for example “these two will always come out matching” or “these two will always come out opposite.”
- Measuring reveals the correlation. The moment you measure one qubit, it settles into a definite outcome, and the rule guarantees what the partner will give when it’s measured. Each individual result still looks random. The pattern only shows up when you compare the two sets of results side by side.
The strange part, the part Einstein balked at, is step 3 holding true even when the two qubits are far apart, with nothing passing between them at the moment of measurement. Bell’s work is what proves this coordination can’t be faked by the qubits secretly agreeing on their answers in advance. The correlation is a property of the joint quantum state itself.
One more fact worth carrying: entanglement is fragile. Stray interaction with the outside world (decoherence) breaks the link, which is exactly why keeping many qubits entangled long enough to compute is so hard.
Source: J. S. Bell, “On the Einstein Podolsky Rosen Paradox,” Physics 1, 1964, cds.cern.ch/record/111654.
Why does entanglement matter for quantum computing?
Entanglement is one of the two ingredients (with superposition) that separates a quantum computer from a very fast classical one. Superposition lets each qubit hold a blend of 0 and 1. Entanglement is what lets a group of qubits act as one connected system, so an operation on the whole register can create and exploit relationships across all of them at once. That coordination is where quantum algorithms get their leverage.
Concretely, it matters in two ways:
- Multi-qubit operations need it. The gates that make a quantum algorithm do anything interesting are the ones that entangle qubits together. Without entangling operations, you just have a set of qubits each doing its own thing, which a classical computer can simulate cheaply.
- It’s tied to quantum advantage. For pure-state quantum computing, a well-known result by Jozsa and Linden shows that any exponential speedup over classical computing requires the amount of entanglement to grow with the size of the problem. In plain terms, entanglement is broadly regarded as a necessary resource for a quantum computer to meaningfully outrun a classical one. Entanglement alone doesn’t guarantee a speedup, yet the big speedups reliably require it.
This is the thread that connects entanglement to the cryptographic story. The reason people worry about Shor’s algorithm breaking RSA and elliptic-curve cryptography is that Shor runs on a large quantum computer, and a large quantum computer only delivers its advantage by orchestrating superposition and entanglement across many qubits. Entanglement is part of the engine under the threat.
Source: R. Jozsa and N. Linden, “On the role of entanglement in quantum computational speed-up,” Proceedings of the Royal Society A 459, 2003, arXiv:quant-ph/0201143.
Why does entanglement matter for cryptography and QKD?
Entanglement shows up on both sides of the quantum-cryptography ledger, the offense and the defense.
On offense, it’s a component of the machine that threatens today’s public-key cryptography, as described above: a cryptographically relevant quantum computer running Shor’s algorithm leans on entanglement to work.
On defense, entanglement is the operating principle behind one whole family of QKD protocols. The 1991 E91 protocol, proposed by Artur Ekert, distributes entangled photon pairs to two parties who each measure their half and derive a shared secret key from the correlated results. The clever part is the built-in alarm: because genuine entanglement produces correlations that violate Bell’s inequality, the two parties can statistically test whether their photons are still fully entangled. An eavesdropper who intercepts and measures the photons unavoidably weakens that entanglement, which shows up as a drop in the measured correlation. The security rests on physics rather than on a math problem an attacker might one day solve.
This is a real capability, and it also has real limits. Distributing entanglement over long distances is hard because photons get lost in fiber, and a practical quantum repeater to extend entanglement across long links without a trusted relay is still research-grade. As a concrete marker of what’s achievable, China’s Micius satellite distributed entangled photon pairs to two ground stations roughly 1,200 kilometers apart in 2017, using the satellite as a relay. That experiment is a useful anchor for the “no matter the distance” claim, and a reminder that scaling it into everyday infrastructure is an open engineering problem. QKD is a specialized tool, not a drop-in replacement for the post-quantum algorithms most organizations will actually migrate to.
Sources: A. K. Ekert, “Quantum cryptography based on Bell’s theorem,” Physical Review Letters 67, 661, 1991, journals.aps.org; J. Yin et al., “Satellite-based entanglement distribution over 1200 kilometres,” Science 356, 1140, 2017, science.org.
Does entanglement allow faster-than-light communication?
No. This is the single most common misconception about entanglement, and it’s worth being precise about why. When you measure your half of an entangled pair, your outcome is random, and you have no way to control it. The partner qubit’s outcome becomes correlated with yours instantly, but the person holding it sees only their own random-looking result. Nothing about their local measurement tells them whether you’ve measured yet, or what you got. To actually learn that the two results are correlated, both parties have to get on an ordinary channel (a phone call, an internet link, a radio) and compare notes, and that comparison crawls along at or below the speed of light like any other message.
This is formalized as the no-communication theorem in quantum information: entanglement produces correlations, and correlations alone can’t transmit information. You can’t encode a message into which outcome you get, because you don’t pick the outcome. So entanglement never beats the cosmic speed limit, and it never will. It’s a source of shared randomness with a guaranteed relationship, and that’s genuinely useful for cryptography, but it is not a faster-than-light telephone.
The reason this matters for a security professional is calibration. Vendors and headlines sometimes lean on the “instantaneous” language to imply something magical. The honest version is narrower and still valuable: entanglement gives two parties correlated secret randomness they can verify, which is the real basis for entanglement-based QKD.
Source: M. A. Nielsen and I. L. Chuang, “Quantum Computation and Quantum Information,” Cambridge University Press, 10th anniversary edition, 2010 (no-communication theorem; foundational treatment of entanglement).
How real is entanglement, and how close is the threat?
Entanglement itself is settled physics. It’s been produced and measured in thousands of experiments, tested to extraordinary precision, distributed over more than a thousand kilometers by satellite, and confirmed by loophole-free Bell tests that earned the 2022 Nobel Prize in Physics. There’s no serious scientific doubt that it’s real.
What’s still far off is a quantum computer that can hold enough qubits entangled, accurately and stably enough, to run Shor’s algorithm against real-world key sizes. That’s the reality gap. Today’s machines are in the noisy, intermediate-scale era, where entanglement is real but delicate and constantly degraded by decoherence. Building a cryptographically relevant machine means keeping large-scale entanglement alive long enough to compute, which is the central engineering challenge and the reason estimates land years out rather than next quarter. Entanglement being real does not mean the code-breaking computer is here. The physics is proven; the machine is an unfinished build.
Source: The Nobel Prize in Physics 2022, nobelprize.org.
Where does entanglement sit in the bigger picture?
| Setting | What entanglement does | Why it matters here |
|---|---|---|
| Quantum computing | Links qubits so a circuit can perform coordinated multi-qubit operations | It’s a necessary resource behind quantum speedups, including Shor’s threat to public-key crypto |
| Entanglement-based QKD (E91) | Distributes correlated secret bits whose integrity can be tested with a Bell inequality | An eavesdropper degrades the entanglement, which is detectable, giving physics-based key security |
| Long-distance quantum links | Correlations survive across large distances (satellite-demonstrated to ~1,200 km) | Shows the “any distance” property is real, while exposing the repeater and loss limits that keep it specialized |
| Communication | Produces correlations only, never controllable outcomes | The no-communication theorem means it can’t send a message faster than light, a common misconception to correct |
Common misconceptions
- “Entanglement lets you send information faster than light.” It doesn’t. The outcomes are random and uncontrollable, so no message rides the correlation, and the no-communication theorem makes this a hard rule of physics.
- “Entanglement means one particle physically pushes the other.” Nothing travels between them at the moment of measurement. What’s shared is a joint quantum state that guarantees how their results relate, which Bell’s work proves can’t be explained by pre-agreed hidden instructions.
- “Entanglement is just ordinary correlation, like two gloves in separate boxes.” Ordinary correlation can be explained by properties fixed in advance. Entanglement produces correlations too strong for any such explanation, which is precisely what Bell inequality violations demonstrate.
- “Entanglement alone is what breaks encryption.” The threat to public-key cryptography is Shor’s algorithm running on a large quantum computer. Entanglement is one ingredient that makes such a computer powerful, working together with superposition and the right algorithm.
- “Because entanglement is proven, a code-breaking quantum computer must be close.” Proven physics and a working cryptanalytic machine are different milestones. Keeping enough qubits entangled and stable to run Shor at scale is an unsolved engineering problem.
- “QKD uses entanglement, so QKD is what everyone should migrate to.” Most organizations will migrate to post-quantum algorithms like ML-KEM, not QKD. Entanglement-based QKD is a specialized, distance-limited tool rather than a general replacement for public-key cryptography.
Questions people ask
Do I need physics to understand entanglement? No. The working intuition is that two qubits become linked so their measurement results are correlated at any distance, while each individual result still looks random. That’s enough to follow why it powers quantum computers and why it can’t send a faster-than-light message.
Why does entanglement threaten encryption? Indirectly, as part of the machine. A quantum computer only outruns a classical one by orchestrating entanglement and superposition across many qubits, and that’s what lets Shor’s algorithm break RSA and ECC. Entanglement is a component of the threat, not the attack by itself.
Does a quantum computer that uses entanglement exist yet? Yes, today’s quantum computers already create and use entanglement, but only at small, noisy scale. A machine that can keep enough qubits entangled and stable to break real cryptographic keys does not exist yet, and building one is the core engineering challenge.
Can entanglement really work “no matter the distance”? The correlation itself has no built-in distance limit, and entangled photon pairs have been distributed by satellite over roughly 1,200 kilometers. The practical limit is holding onto the entanglement against loss and noise over long links, which is why quantum repeaters are an active research problem.
If it’s instant, why can’t it beat the speed of light? Because you can’t control your own measurement outcome, so you can’t encode a message into it. Learning that two results are correlated requires comparing them over an ordinary channel, and that comparison is bound by the speed of light like any other communication.
How is entanglement different from superposition? Superposition is a single qubit holding a blend of 0 and 1. Entanglement is a relationship across two or more qubits, where their outcomes are linked. Quantum algorithms use both together, and the entangling operations are what create genuinely quantum coordination.
Is entanglement-based QKD the same as post-quantum cryptography? No. QKD uses quantum physics to distribute keys over specialized hardware, while post-quantum cryptography is new math-based algorithms that run on ordinary computers. National security agencies generally steer organizations toward post-quantum algorithms for most migrations, treating QKD as a niche tool.
Is quantum computing just hype, then? No, but the timeline gets hyped. Entanglement and the underlying physics are real and demonstrated. What’s uncertain and often oversold is how soon a large, stable, cryptographically relevant machine arrives, which is why serious planning tracks the threat timeline instead of any single headline.
Everything here is the map, given freely. When your team needs this turned into a cryptographic inventory and a dated plan for your own estate, that’s what an alignment briefing is for.
Last verified 2026-07-09 · Maintained by Addie LaMarr, LaMarr Labs.