Quantum computing and AIMiddleLesson 3087 min read

Hype check: where quantum stands

Today's quantum computers are small and noisy. Learn what is proven, why error correction is the big goal, and why post-quantum cryptography matters for security right now.

✦ Ask the tutor 🧪 Lab · 25 min 📝 Lesson test

Speed

Lesson in motion

In 60 seconds

Hype check: where quantum stands

Today's quantum computers are small and noisy. Learn what is proven, why error correction is the big goal, and why post-quantum cryptography matters for security right now.

1/5

In simple words

Imagine trying to build a tall tower of cards in a room where a fan is blowing. Each card you add might get knocked. That is what building a big quantum computer is like: every extra qubit and every extra step is another chance for noise to knock things over. Engineers are learning to build the tower in a way that can fix fallen cards as it goes.

Quantum computing is real, and it is also heavily hyped. A good way to think clearly is to sort claims into three boxes: what is proven in theory, what is shown on real machines today, and what is still an open question.

What is proven in theory

Algorithm	Problem	Speed-up	Needs
Shor's algorithm	Factoring large numbers, and related problems used in public-key cryptography	Far faster than any known classical method (polynomial instead of the best known classical, which grows much faster)	A large, error-corrected quantum computer
Grover's algorithm	Searching an unstructured list for a marked item	Quadratic: about √N steps instead of about N	Many reliable qubits; gain is smaller than exponential
Quantum simulation	Simulating molecules and materials that follow quantum rules	Expected to be large for some problems; a natural fit	Size and accuracy depend on the problem

Notice the pattern. Proven big gains are for specific problems, and most need machines far larger and more reliable than exist today. A quadratic speed-up like Grover's can also be eaten up by the overhead of error correction and slower quantum operations, so it does not automatically help in practice.

What machines can do today

Today's quantum computers have a modest number of qubits, and each operation has a small chance of error. Errors pile up as circuits get longer, so only fairly short circuits give reliable results. This period is often called the era of noisy intermediate-scale quantum computers. Researchers have shown impressive experiments, but a useful, practical advantage on a real-world problem is still a goal, not an everyday fact.

Error correction: the big goal

Classical computers protect data by keeping extra copies and checking them. Quantum error correction is harder, because you cannot copy an unknown qubit, and measuring it directly destroys the state. The solution is clever: spread one logical qubit over many physical qubits and measure only checks that reveal errors without revealing the data. If physical error rates are below a certain threshold, adding more physical qubits makes the logical qubit more reliable.

The simplest idea behind it is the classical repetition code: store a bit three times and take a majority vote. Run this to see how much it helps when each copy flips with probability p:

Python: majority vote of three noisy copiespython

import numpy as np

rng = np.random.default_rng(0)
trials = 100_000
for p in [0.01, 0.05, 0.1, 0.3, 0.6]:
    flips = rng.random((trials, 3)) < p          # which of the 3 copies flipped
    wrong_after_vote = flips.sum(axis=1) >= 2     # majority is wrong
    theory = 3 * p**2 - 2 * p**3
    print(f"p = {p:.2f}  one copy wrong: {p:.3f}  after vote: {wrong_after_vote.mean():.4f}  (theory {theory:.4f})")

When p is small, the vote makes errors much rarer. When p is too large, above one half here, the vote makes things worse. Quantum error correction has the same flavour: it only helps once hardware is good enough. Real quantum codes also have to handle phase errors, and need many physical qubits per logical qubit.

Why security teams care now: post-quantum cryptography

Much of the internet's security, including the key exchange that protects websites and apps and many digital signatures, uses public-key methods such as RSA and elliptic-curve cryptography. A large, error-corrected quantum computer running Shor's algorithm could break these. No such machine exists today, but the risk is taken seriously for two reasons:

Harvest now, decrypt later. An attacker can record encrypted traffic today and keep it, hoping to decrypt it years later. Data that must stay secret for a long time, like health or government records, is at risk now.
Changing cryptography takes years. Banks, telecoms and governments have to update software, hardware, cards and devices. Starting late is dangerous.

Post-quantum cryptography (PQC) means new public-key methods, running on normal computers, that are believed to resist attacks from both classical and quantum computers. Standards bodies have published the first PQC standards, and many systems are starting to adopt them. Symmetric encryption, like AES, is affected less: Grover's algorithm gives only a quadratic speed-up, and longer keys are the usual answer.

Not the same thing

Quantum key distribution: uses real quantum hardware and special links to share keys

Post-quantum cryptography

New maths-based algorithms that run on today's computers and networks

Note

India has a National Quantum Mission, approved in 2023, to support quantum research, technology and skills, including computing, communication and sensing. For current programmes and calls, check the official government websites.

Do this

A calm summary: quantum computing is real science with real progress. Big proven speed-ups exist for specific problems. Practical machines for those problems are not here yet. Quantum AI speed-ups are not proven. And the security risk is real enough that moving to post-quantum cryptography is sensible now.

Try this. In the code, add p = 0.5 to the list. What happens after the vote, and why?

Watch and read more

NIST: Post-Quantum Cryptography projectNIST · reading

Lab

A one-page hype check: five quantum claims sorted into proven, shown today, or open question, with a reason for each.

~25 min

The problem

This is a thinking and writing lab. For each claim below, write 2–4 lines: put it in one box (proven in theory, shown on real machines today, open question, or false), and say why.

"A large, error-corrected quantum computer could factor the numbers used in RSA much faster than known classical methods."
"Quantum computers will make training AI models exponentially faster."
"Entanglement lets us send messages faster than light."
"Grover's algorithm gives a quadratic speed-up for unstructured search."
"Quantum computers try every answer at once and choose the best."

Finally, write three sentences for a friend who works in a bank's IT team: why post-quantum cryptography matters to them now, and one first step they could suggest (for example, making a list of where public-key cryptography is used).

You are done when

Claims 1 and 4 are placed as proven in theory, with the note that claim 1 needs a large error-corrected machine
Claim 2 is an open question, not proven
Claims 3 and 5 are false, with a correct reason (no-communication; measurement gives one outcome, interference does the work)
The bank note mentions harvest-now-decrypt-later or long migration times, and a practical first step

Hard questions

Try to answer before you reveal. If you can answer these, you understood the lesson.

Q1If Grover's algorithm gives a quadratic speed-up for search, why does doubling a symmetric key length restore security?Reveal

A brute-force search over a k-bit key takes about 2ᵏ tries classically. Grover's algorithm needs about √(2ᵏ) = 2^(k/2) steps. So a 256-bit key would give roughly the same Grover effort as a 128-bit key does classically, which is still out of reach. This is why symmetric encryption is less affected than RSA and elliptic-curve methods, which Shor's algorithm breaks outright.

Q2Why is "more qubits" alone a poor measure of how useful a quantum computer is?Reveal

Usefulness depends on how accurate each operation is, how long qubits keep their state, how qubits connect to each other, and whether errors can be corrected. Many noisy qubits may run only short circuits before errors swamp the result. A smaller machine with much lower error rates, or with working logical qubits, can do more useful work.

Please sign in to continue.

Questions people ask

Will quantum computers make my laptop obsolete?

No. Quantum computers are expected to be special tools for particular problems, used alongside ordinary computers, much like a specialised lab machine. Email, games, video and most AI will keep running on classical hardware.

Is my UPI payment or WhatsApp message at risk today?

Not from any quantum computer that exists today. The concern is about the future and about data recorded now and decrypted later. That is why banks, app makers and governments are working on moving to post-quantum cryptography. Keep your apps updated so you get those changes.

If Grover's algorithm speeds up search, does it help web search?

Not in the way the name suggests. Grover's algorithm is for searching an unstructured space where you can only check items one by one, and it gives about √N steps instead of N. Web search uses indexes and structure, which classical computers already exploit very well.

How do I tell a serious quantum claim from hype?

Look for: a named, specific problem; comparison with the best classical method; honest error bars; whether it ran on real hardware; and whether other researchers have checked it. Be wary of claims like "quantum AI will solve everything" with no details.

Lesson test

5 questions. Get 3 right (60%) to pass and complete this lesson.