Quantum Computing is a form of computation that uses quantum-mechanical systems to represent and manipulate information. Its fundamental unit of information is the qubit, which differs from a conventional binary bit because a qubit can occupy a quantum superposition of basis states before measurement. Quantum computation uses controlled quantum operations, measurement and interference to transform probability amplitudes so that certain computational outcomes can be amplified while others are suppressed.
Quantum computers are not simply faster versions of classical computers and do not provide a universal speed advantage for ordinary software. Their potential comes from specialised computational models and algorithms that can provide significant theoretical advantages for particular problem classes. Practical quantum computing remains technically challenging because useful quantum states are fragile and difficult to control.
Quantum computation emerged from the recognition that classical computers may be inefficient for simulating quantum systems. Richard Feynman and other researchers discussed the limitations of classical simulation for quantum phenomena, while David Deutsch formalised the concept of a universal quantum computer. Peter Shor developed a quantum algorithm for integer factorisation and related problems, demonstrating a major theoretical implication for cryptography. Lov Grover developed a quantum search algorithm providing a quadratic speedup for unstructured search under its model. These algorithms helped establish quantum computing as a distinct field combining computer science, physics, mathematics and engineering. Experimental systems have subsequently been developed using technologies including superconducting circuits, trapped ions, neutral atoms, photons and other physical platforms.
A qubit can be represented mathematically as a combination of basis states commonly written as zero and one, with complex probability amplitudes. Measurement produces a classical outcome according to probabilities determined by the quantum state. Superposition should not be interpreted as a qubit simply containing two classical values that can both be read simultaneously; measurement extracts a classical result and generally changes the state.
Entanglement is another important resource in which the joint quantum state of multiple qubits cannot be represented as independent states of the individual qubits. Quantum interference allows amplitudes associated with computational paths to reinforce or cancel one another. These principles distinguish quantum algorithms from classical randomised algorithms and provide the mechanisms through which quantum computational advantages can arise.
Quantum gates manipulate qubit states through reversible transformations, and quantum circuits arrange gates into sequences that implement algorithms. Common conceptual gates include operations corresponding to bit flips, phase changes and transformations that create superposition, while controlled operations can generate entanglement. A quantum circuit begins with an initial state, applies a sequence of gates and ends with measurements. Quantum algorithms must be designed so that measurement probabilities encode useful information in the final state.
The no-cloning theorem states that an unknown arbitrary quantum state cannot be perfectly copied, which has important consequences for information handling and error correction. Quantum operations are also subject to physical constraints and noise, meaning that theoretical circuits often need substantial engineering support before they can be executed reliably on hardware.
Quantum error correction addresses the extreme sensitivity of quantum states to environmental disturbance and imperfect operations. Physical qubits can experience decoherence, gate errors, measurement errors and other forms of noise. A logical qubit can be encoded across multiple physical qubits so that certain errors can be detected and corrected without directly measuring and destroying the encoded quantum information. Error-correcting codes introduce substantial overhead, meaning that a useful fault-tolerant quantum computer may require many physical qubits for each logical qubit.
Fault-tolerant quantum computation aims to perform sufficiently reliable logical operations despite imperfect physical components. This distinction between physical and logical qubits is essential when evaluating quantum-computing capabilities because a large count of physical qubits does not automatically imply the ability to execute large, reliable algorithms.
Quantum algorithms have been developed for specific computational tasks. Shor's algorithm can factor integers and compute discrete logarithms efficiently in the relevant theoretical model, creating important implications for public-key cryptographic systems based on those problems. Grover's algorithm provides a quadratic speedup for unstructured search, although the improvement differs fundamentally from the exponential-type speedups associated with some quantum algorithms. Quantum simulation is a major prospective application because quantum systems can be difficult to simulate efficiently on classical computers.
Other research areas include optimisation, chemistry, materials science and machine learning, but claimed practical advantages in these areas must be evaluated according to specific algorithms, hardware assumptions, error rates and classical baselines. Quantum computing should therefore not be treated as a general replacement for classical computing.
Quantum hardware platforms make different engineering trade-offs. Superconducting qubits use fabricated electrical circuits operating at extremely low temperatures and can support fast operations but require demanding cryogenic infrastructure. Trapped-ion systems manipulate ions using electromagnetic fields and lasers and can achieve strong control and connectivity characteristics, although operations can be comparatively slow. Neutral-atom approaches arrange and manipulate atoms using optical techniques. Photonic approaches encode information in properties of light and can offer different networking and operating characteristics. Other approaches continue to be researched. No single physical implementation has established universal superiority across all practical criteria. Relevant measures include gate fidelity, coherence time, operation speed, connectivity, measurement accuracy, scalability, error-correction overhead, control complexity and total system requirements.
Quantum Computing has important limitations that are often obscured by popular descriptions. Quantum speedups apply only to particular algorithms and problem structures, and converting a real-world problem into a suitable quantum algorithm can itself be difficult. Quantum hardware is noisy, expensive and technically complex, while error correction can require substantial physical resources. Classical computers will remain necessary for control systems, data preparation, post-processing and many ordinary computational tasks.
Quantum algorithms also frequently require repeated circuit execution and statistical sampling rather than producing a useful result from a single run. Current research therefore focuses on improving hardware quality, increasing useful circuit depth, developing error-correcting architectures and identifying applications where quantum resources can provide measurable benefits compared with the best available classical methods.
