Understanding Quantum Computing: A Plain-English Explainer

In 2019, Google announced "quantum supremacy" — the claim that their 53-qubit Sycamore processor had solved a calculation in 200 seconds that would take the world's fastest classical supercomputer 10,000 years. IBM immediately disputed the claim, arguing their classical Summit supercomputer could do it in 2.5 days.

Both were right about different things. Google's quantum computer did solve that specific, carefully chosen problem dramatically faster. But that problem — sampling from a specific random circuit distribution — has no practical application. It was a benchmark, not a breakthrough for real-world computing.

This is the central confusion about quantum computing: the gap between what the technology can theoretically do and what it can practically do today is enormous. Understanding that gap is essential to understanding what is genuinely coming — and what is hype.

What Makes a Quantum Computer Different

Classical Bits vs. Qubits

A classical computer processes information in binary bits — each bit is either 0 or 1 at any moment. This is perfect for most computing tasks. A quantum computer uses qubits, which can represent 0, 1, or — due to the quantum mechanical property of superposition — a combination of both simultaneously, described by probabilities.

Here is the critical nuance most explanations get wrong: a qubit is not "both 0 and 1 at the same time" in a way you can directly use. When you measure a qubit, it collapses to 0 or 1 with specific probabilities. The value of superposition is computational — you can process information about multiple possible states simultaneously before measurement forces a definite answer.

An analogy: imagine you are searching a maze. A classical computer tries each path one by one. A quantum computer can, in a sense, explore all paths simultaneously and then exploit interference (constructive for correct paths, destructive for wrong paths) to arrive at the answer efficiently. This works for specific problem structures — not every problem.

Superposition, Entanglement, and Interference

Three quantum phenomena power quantum computation:

• Superposition: A qubit can be in a combination of states until measured. With n qubits, a quantum computer can represent 2^n states simultaneously. 300 qubits can represent more states than there are atoms in the observable universe.
• Entanglement: Entangled qubits have correlated states — measuring one instantly tells you something about the other, regardless of distance. This allows quantum computers to process correlated information efficiently.
• Interference: Quantum algorithms manipulate superposition states so that correct answers amplify (constructive interference) and incorrect ones cancel out (destructive interference) — this is how useful quantum algorithms extract answers from quantum states.

The Qubit Quality Problem: Why "1,000 Qubits" Is Misleading

News headlines celebrate companies announcing computers with hundreds or thousands of qubits. But qubit count is almost meaningless without knowing qubit quality — specifically, error rates.

Current quantum computers are "noisy intermediate-scale quantum" (NISQ) devices. Every qubit operation has a small probability of error. With current physical qubit error rates of roughly 0.1–1%, a 100-step quantum algorithm on 50 qubits will produce a wrong answer more often than a right one. Larger computations are currently limited by cumulative errors.

IBM's quantum roadmap targets 100,000+ physical qubits by 2033, enough to support error-corrected "logical qubits" for useful computation. The roadmap is available at ibm.com/quantum/roadmap.

Error correction requires using multiple physical qubits to encode one reliable "logical qubit." Current estimates suggest you need roughly 1,000 physical qubits per logical qubit using current error correction codes. A useful quantum computer might need 1,000 logical qubits — meaning 1 million physical qubits at high quality. We are not there yet.

What Problems Are Quantum Computers Actually Good At?

Quantum computing is not universally faster than classical computing. It offers speedups only for specific problem classes.

Problems Where Quantum Wins

• Factoring large numbers (Shor's algorithm): A quantum computer can break RSA encryption by factoring the large numbers that underpin it. This is the most significant cryptographic threat — but requires error-corrected quantum computers we do not yet have.
• Simulating quantum chemistry (molecules and materials): Quantum computers can simulate molecular interactions that classical computers approximate expensively. This has direct applications in drug discovery, materials science, and battery development.
• Optimization problems with specific structure: Some logistics, financial portfolio optimization, and machine learning training problems have quantum algorithmic speedups. The practical advantage over best classical algorithms is still debated.
• Quantum machine learning: Theoretically, quantum algorithms for linear algebra operations (used in ML) offer speedups. In practice, the overhead of loading classical data into quantum systems often negates the advantage.

Problems Where Classical Wins (or Quantum Offers No Advantage)

• Most everyday computing (email, web browsing, video streaming, databases): These do not benefit from quantum speedups.
• Training most neural networks: Current evidence suggests classical supercomputers will remain dominant for AI training.
• Any problem solvable efficiently by classical algorithms: Quantum is not a universal speedup — it only helps when the problem structure aligns with quantum algorithmic advantages.

The Encryption Threat: The Most Important Real-World Implication

The most significant near-term implication of quantum computing is cryptographic. RSA, ECC, and Diffie-Hellman — the encryption algorithms protecting your banking, email, VPN, and HTTPS connections — can be broken by Shor's algorithm on a sufficiently powerful quantum computer.

This is not an immediate threat: the quantum computers needed to break RSA-2048 require approximately 4,000 error-corrected logical qubits running for hours. No existing quantum computer comes close. But the threat is significant enough that NIST has already finalized post-quantum cryptography (PQC) standards.

In August 2024, NIST released the first three post-quantum cryptography standards: ML-KEM, ML-DSA, and SLH-DSA. These replace RSA and ECC with lattice-based algorithms resistant to quantum attacks. See the announcement at nist.gov/news-events/news/2024/08/nist-releases-first-3-finalized-post-quantum-encryption-standards.

Who Is Building Quantum Computers and Why

IBM

IBM has the largest publicly accessible quantum computing program (IBM Quantum Network) and has published the most detailed roadmaps. Their 2024 Heron processor demonstrated improved error rates. IBM's approach focuses on superconducting qubits cooled to near absolute zero (−273°C / 15 millikelvin).

Google

Google's Quantum AI team made the 2019 supremacy claim and has continued publishing important research. Their focus is also superconducting qubits. In late 2024, they announced Willow — a 105-qubit chip that demonstrated improved error correction with more qubits, a key milestone.

Microsoft

Microsoft is pursuing topological qubits — a physically different approach that, if successful, would provide much lower error rates. Microsoft has invested heavily but has yet to demonstrate a fully working topological qubit system. If it works, it could leapfrog superconducting approaches.

IonQ, Quantinuum, and Others

Trapped-ion quantum computers (IonQ, Quantinuum) use a different physical technology — ions suspended in electromagnetic fields. They have higher qubit quality but are harder to scale. Photonic quantum computers (PsiQuantum) use photons and can theoretically operate at room temperature.

China

China has made significant quantum computing investments. USTC's Jiuzhang photonic quantum computer achieved its own form of quantum advantage in 2020 for specific sampling problems. China is also investing heavily in quantum communication (quantum key distribution networks).

When Will Quantum Computing Actually Affect Everyday Life?

Honest timeline:

• 2026–2030 (now): NISQ devices for niche scientific computation. Quantum chemistry simulations for drug discovery and materials science are the most plausible near-term applications.
• 2030–2035: Error-corrected quantum systems may begin demonstrating advantage over classical computers for specific commercial optimization and simulation problems.
• 2035–2040: If roadmaps hold, fault-tolerant quantum computers might begin impacting cryptography and large-scale optimization. This would require massive investment in post-quantum cryptography migration.
• 2040+: Broad commercial quantum advantage for chemistry, materials, pharmaceuticals, and possibly machine learning. This is speculative.

Quantum Computing for Non-Specialists: What You Actually Need to Know

For most people, quantum computing is a background technology whose implications you will feel without needing to understand the physics. The most important practical implications:

• Cybersecurity: The move to post-quantum encryption is underway. If you run a website or application, your SSL/TLS providers will handle this migration. Pay attention to updates from your hosting providers and security vendors over the next 5 years.
• Drug discovery: Quantum-aided molecular simulation could accelerate pharmaceutical development by 10x or more, potentially bringing drugs to market faster and at lower cost.
• Materials science: Battery technology, solar panel efficiency, and semiconductor design may all benefit from quantum simulation — accelerating the energy transition.
• Finance: Portfolio optimization and risk modeling are candidates for quantum advantage, though the practical benefit over existing tools remains unproven.

Frequently Asked Questions

Can I use a quantum computer today?

Yes — IBM provides cloud access to their quantum computers through IBM Quantum (quantum.ibm.com), with free access to some systems. You can write quantum programs in Qiskit, IBM's open-source quantum software. The practical limitation is that current systems are NISQ devices useful primarily for research and learning, not commercially useful computation.

Will quantum computers replace classical computers?

No. Quantum computers are specialized accelerators for specific problem types — not general-purpose computers. They require extreme cooling, careful isolation from vibration and electromagnetic interference, and specialized software. Classical computers will remain the workhorse for virtually all everyday computing for the foreseeable future.

Is the quantum internet real?

Quantum communication — using quantum properties to transmit unhackable keys — is real and has been demonstrated at small scale. China launched the first quantum communication satellite (Mozi/Micius) in 2016 and demonstrated quantum key distribution over 4,600km in 2020. A true "quantum internet" that replaces today's internet is decades away and architecturally different from current proposals.

The Bottom Line

Quantum computing is real, progressing, and significant — but the timeline for commercial impact is measured in decades, not years, for most applications. The cryptographic threat is the most near-term concrete issue worth taking seriously now, and NIST's post-quantum standards address it.

For curious non-specialists: IBM Quantum's free cloud access is the best way to experience quantum computing directly. For security professionals: start planning your post-quantum cryptography migration. For everyone else: follow the field with calibrated optimism — the applications in healthcare and materials science could be genuinely transformative on a 10–20 year horizon.