Quantum Computing Advantage: Google's 2019 Breakthrough
Photo: Google DeepMind / Pexels
In October 2019, Google announced that its Sycamore quantum processor had achieved "quantum advantage"—it solved a specific problem faster than any classical supercomputer could. The problem was artificial (sampling from a random quantum circuit distribution), not directly useful for real-world applications. Yet the achievement was momentous: it demonstrated that quantum computers could, in principle, outperform classical computers at certain tasks. Quantum computing, long relegated to theoretical physics and small-scale experiments, had achieved a concrete milestone.
The breakthrough sparked excitement and skepticism. Quantum computers still can't meaningfully factorize large numbers (threatening encryption), simulate molecular systems, or solve optimization problems better than classical approaches—the real applications that would justify quantum computing's promise. Yet Google's result proved the underlying physics works. If engineers can scale quantum computers to more qubits while maintaining quality, real quantum advantages for practical problems will follow.
Key Takeaways
- The Challenge: Building quantum computers with enough qubits (quantum bits) to outperform classical computers while maintaining quantum coherence
- The Breakthrough: Sycamore's 53-qubit processor achieved quantum advantage on a synthetic benchmark in 200 seconds (vs. 10,000 years for classical supercomputers)
- Real-World Impact: Proved quantum supremacy is possible; validated quantum computing as viable technology; sparked investment and competition
- Future Directions: Error correction, larger qubit counts, and new algorithms aim for practical quantum advantage in drug discovery, optimization, and cryptography
The Challenge
Quantum computers harness quantum mechanics's strange properties. Classical computers use bits (0 or 1). Quantum computers use qubits (quantum bits) that can be 0, 1, or both simultaneously—a superposition. This enables quantum computers to explore many possibilities in parallel.
Building a quantum computer requires:
Qubits: Physical systems maintaining quantum superposition long enough for computation. Candidates include superconducting circuits, trapped ions, photons, and topological qubits. Each approach has tradeoffs between qubit count, coherence time, and control precision.
Coherence: Quantum states are fragile. Environmental noise causes decoherence—qubits lose quantum properties and collapse to classical states. Maintaining coherence for long enough to complete computations is challenging. Current superconducting qubits maintain coherence for roughly 100 nanoseconds (billionths of a second).
Control: Manipulating qubits requires precise pulses (microwaves for superconducting qubits). Implementing quantum gates (operations on qubits) with high fidelity is difficult. Current gate errors are 0.1-1%—much higher than needed for practical computation.
Error Correction: Quantum error correction uses multiple physical qubits to encode one logical qubit, protecting against noise. However, error correction itself introduces errors. To benefit from quantum error correction, physical qubit error rates must be below a "threshold" (roughly 0.1%), below which additional redundancy reduces overall error. Current error rates don't reach this threshold for most systems.
Scalability: Scaling from 10 qubits to 100, 1000, or million qubits introduces immense technical challenges. Each additional qubit requires careful control signals, cryogenic infrastructure, and integration with other systems.
For decades, skeptics doubted quantum computers would ever work practically. The quantum decoherence problem seemed fundamental—perhaps insurmountable. Even optimists expected quantum advantage to remain theoretical for years.
The Breakthrough
Google built Sycamore, a superconducting quantum processor with 53 functional qubits. The team designed a benchmark: sample outputs from random quantum circuits of varying depth (complexity). This problem is easy for quantum computers (apply the quantum circuit and measure) but apparently hard classically (simulating the quantum circuit's probability distribution).
In September 2019, Sycamore ran the benchmark, generating samples in 200 seconds. Google claimed that the strongest classical supercomputer (IBM's Summit) would require 10,000 years to do the equivalent. IBM disputed the claim, arguing more efficient classical algorithms could do it in 1-3 days, and questioned whether the benchmark reflected real-world relevance.
The dispute was partly academic. Google's point wasn't that they'd solved a practical problem—they hadn't. The point was principle: a quantum computer exceeded classical computers' capability at a well-defined task. Whether that task was useful was secondary.
By the standards of quantum computing research, this was revolutionary. It validated quantum mechanics (quantum computers behave as theory predicts), proved qubits could be controlled and measured, demonstrated computation was possible at scale (53 qubits, hundreds of quantum gates), and proved quantum advantage was achievable. All of these had been questioned.
How It Works
Understanding quantum computing requires understanding key quantum phenomena:
Superposition: A quantum system can exist in multiple states simultaneously. A qubit can be 0 and 1 at the same time (weighted by amplitudes). Measuring the qubit forces it into one state, with probability determined by the amplitudes.
Entanglement: Qubits can become entangled, meaning their states are correlated in ways without classical analog. Measuring one entangled qubit instantaneously affects the states of others (non-locally, though information can't be transmitted faster than light).
Quantum Interference: Quantum amplitudes can add constructively (amplifying certain outcomes) or destructively (canceling others). Clever algorithms exploit this: arrange computations so wrong answers destructively interfere (cancel) while correct answers constructively interfere (amplify).
Quantum Gates: Operations on qubits analogous to classical logic gates. Single-qubit gates rotate quantum states. Two-qubit gates (CNOT, CPHASE) create entanglement. These gates are implemented by pulses applied to qubits.
Measurement: Measuring a qubit collapses its superposition to a classical state (0 or 1), determined probabilistically by the quantum state before measurement. Extracting results from quantum computers requires repeated measurements (sampling).
The Quantum Circuit Model: Quantum computation proceeds as quantum circuits—sequences of quantum gates applied to qubits, ending with measurement. Sycamore implemented random circuits as a benchmark.
Why It's Hard to Simulate Classically: A quantum system with n qubits has 2^n possible states. Tracking all amplitudes requires storing 2^n complex numbers. For 53 qubits, that's about 10^16 complex numbers—more memory than classical supercomputers have. This is why simulating quantum circuits becomes exponentially harder as qubit count grows.
Real-World Impact
By 2024, Google's quantum advantage had catalyzed significant developments:
Research Investment: Google's success attracted billions in quantum computing investment. IBM, Microsoft, Amazon, Intel, and numerous startups entered the quantum computing race. Academic research intensified.
Quantum Computers Accessible: Google, IBM, and others offered cloud access to quantum processors. Researchers worldwide could experiment with quantum computing without owning hardware.
Error Correction Progress: Google's results motivated error correction research. In 2024, Google announced surface code implementations approaching the error-correction threshold—a major step toward practical quantum advantage.
Algorithm Development: Researchers developed quantum algorithms for optimization (QAOA—Quantum Approximate Optimization Algorithm), machine learning, and chemistry. While not yet faster than classical approaches, these algorithms point toward future applications.
Hardware Improvements: Qubit counts increased: Google moved from 53 qubits (Sycamore) to 70+ qubits (Willow). IBM, IonQ, and others increased qubit counts while improving error rates.
Hybrid Approaches: Systems combining classical and quantum processors are being developed. Classical computers prepare data and post-process results; quantum processors handle quantum-suitable sub-problems. This hybrid approach is practical for near-term devices.
Benchmark Development: Quantum computing benchmarks (like Willow's error correction demonstration) are becoming standardized, enabling fair comparison across platforms.
What's Next
The quantum computing frontier in 2024-2026 involves:
Error Correction: The grand challenge. If quantum error rates can be pushed below ~0.1%, error correction becomes beneficial, enabling "logical qubits" far more reliable than physical qubits. This enables scaling to practical qubit counts (millions).
NISQ to UESQ: Current devices are NISQ (Noisy Intermediate-Scale Quantum—dozens to hundreds of qubits, high error rates). The goal is UESQ (Useful-Scale Quantum), where errors are controlled well enough for practical computation.
Qubit Improvements: Higher coherence times, lower error rates, and better connectivity between qubits all improve quantum computer capability.
New Qubit Types: While superconducting qubits dominate, trapped ions, photonic qubits, and topological qubits are advancing. Different qubit types have complementary advantages.
Practical Algorithms: Development of quantum algorithms for real problems—drug discovery (simulating molecules), optimization (supply chains, machine learning), chemistry (catalysis)—remains critical.
Quantum Networking: Connecting multiple quantum processors over distances via quantum repeaters is an emerging direction, potentially enabling distributed quantum computation.
Fault-Tolerant Quantum Computing: the definitive goal: a quantum computer with error rates so low that algorithms run without significant degradation. Estimates suggest this requires 1 million+ qubits with error rates below 0.01%. Timelines are uncertain—anywhere from 5 to 15+ years away.
FAQ
Q: Does quantum advantage mean quantum computers are better than classical computers? A: Not yet, and not generally. Google's quantum advantage applies to a specific artificial problem. For most practical tasks, classical computers remain superior. Quantum computers excel at specific problem classes (sampling, certain optimization tasks, quantum simulation). As the field matures, quantum advantage will broaden.
Q: Could quantum computers break encryption? A: Potentially. Shor's algorithm, developed in 1994, shows quantum computers could factorize large numbers exponentially faster than known classical algorithms. This would break RSA encryption. However, large-scale quantum computers capable of this (millions of qubits, extremely low error rates) don't exist and may not exist for 10+ years. The threat motivates post-quantum cryptography development.
Q: Why is error correction so important? A: Quantum states are fragile. Noise causes errors. If you want a quantum computer to run long algorithms reliably, you must correct errors. Error correction trades physical qubits for logical qubit reliability—using multiple physical qubits to protect one logical qubit. This increases the qubit count needed but enables practical computation.
Q: How many qubits do we need for practical quantum computers? A: Depends on the application. Drug discovery simulations might require 1,000-10,000 qubits. Optimization problems vary. A rough estimate: practical quantum advantage across many domains likely requires 10,000-1,000,000 qubits with error rates below 0.1%.
Q: Is quantum supremacy the same as quantum advantage? A: "Quantum supremacy" was Google's original term. IBM criticized it, noting "supremacy" has uncomfortable connotations. The field now uses "quantum advantage" (a quantum computer outperforms classical for some task) and "quantum utility" (quantum computers solve practical problems faster than classical).
Q: Could quantum computers replace classical computers? A: Unlikely. Quantum computers excel at specific problems; classical computers remain superior for most tasks. The future likely involves hybrid systems—classical computers for general purpose, quantum co-processors for specific algorithms.
Q: Why is Google's benchmark artificial? Why not solve real problems? A: Google chose a benchmark specifically because: (1) it's easy for quantum computers (apply a circuit, measure), (2) it's hard for classical computers (exponentially hard to simulate), and (3) it's unambiguous—the answer is well-defined and verifiable. Real-world problems are messier; their solutions would be harder to verify.
Related Articles
- Transformers: The Architecture Behind Modern AI
- CERN Higgs Boson Discovery: Finding the God Particle
- AlphaFold: AI Solving Protein Folding in Hours
Sources:
- Arute, F., et al. (2019). "Quantum Supremacy Using a Programmable Superconducting Processor." Nature, 574(7779), 505-510. https://doi.org/10.1038/s41586-019-1666-5
- Google Quantum AI. (2024). "Willow: Solving Quantum Error Correction." Nature (in press).
Read next:
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles
Quantum Entanglement: Bell's Theorem Experiments
How quantum entanglement experiments revealed nature defies classical intuition
Protein Crystal Growth in Microgravity: What ISS Crystals Show
Why researchers grow protein crystals in orbit, what the experiments have measured since the 1980s, and where the evidence for better crystals is still mixed.
How Do Quantum Computers Actually Work? Beyond the Hype
Quantum computers don't replace your laptop — they solve a small set of problems exponentially faster. Here's what they really are, why they're hard to build, and what they'll change.
Quantum Error Correction Explained: Physical Qubits, Logical Qubits and Thresholds
Quantum error correction protects one logical qubit across many noisy physical qubits. Learn how syndromes, surface codes, and thresholds work in practice.
The Most Important Scientific Discoveries of 2025: Breakthroughs and Their Impact
2025 marked significant scientific progress across multiple disciplines: a structural biology breakthrough in protein folding, advances in quantum computing error correction, discovery of a novel class of high-temperature superconductors, clinical trials demonstrating efficacy of novel Alzheimer's disease therapies, successful gene therapy applications for hereditary blindness, and confirmation of predicted gravitational wave signatures. This article examines the most consequential discoveries of 2025 — those that represent genuine advances in human knowledge, demonstrate technological capability, or have immediate clinical applications — explaining the science, why each discovery matters, and the implications for future research.
Quantum Machine Learning Explained: Hype, Promise, and Reality
Quantum machine learning promises exponential speedups for AI—but separating genuine potential from overhype requires careful attention to the physics.