Quantum computing: foundations, algorithms, and emerging applications

Abstract

Quantum computing is an emerging paradigm that leverages the principles of quantum mechanics to solve computational problems beyond the reach of classical computers. This article provides an overview of the fundamental concepts of qubits, the distinctive features of quantum mechanics such as superposition and entanglement, and the challenges of building scalable, fault-tolerant systems. It surveys key quantum algorithms and their potential applications in fields including cryptography, optimization, finance, chemistry, and machine learning. Additionally, it highlights the importance of verification frameworks for ensuring the reliability of quantum programs. A literature review of significant contributions is presented, drawing insights from recent surveys on quantum algorithms, qubit technologies, and software verification approaches. The article concludes by discussing ongoing challenges, such as error correction overhead, hardware scalability, and verification complexity, and suggests directions for future research.

1 Introduction

Imagine a machine that does not compute by flipping a long string of zeros and ones, but by coaxing tiny quantum objects into behaving like complex waves of possibility. That’s the intuitive leap behind quantum computing: instead of bits that are definitely 0 or 1, quantum computers use qubits that can exist in superpositions of states, become entangled so their fates are linked across space, and exploit interference to amplify correct answers while canceling wrong ones. These phenomena: superposition, entanglement, and interference are the conceptual tools that let quantum algorithms explore many possible solutions at once in ways classical algorithms cannot.

Because of those properties, quantum machines have the potential to transform domains where classical approaches struggle. Simulating complex molecules for chemistry and materials science, tackling hard optimization problems in logistics and finance, and accelerating certain kinds of machine-learning and search tasks. Researchers have already demonstrated early milestones where quantum processors performed narrowly defined tasks far faster than classical machines, milestones sometimes called quantum supremacy or quantum advantage. These show the field is progressing from theory toward demonstrable speedups (Benioff, 1980; Feynman, 1982; Deutsch, 1985; Shor, 1995; Grover, 1996).

The present period, sometimes called the NISQ (Noisy Intermediate-Scale Quantum) period, is characterised by machines with tens to a whole bunch of qubits which might be inherently noisy and susceptible to errors. Fully fault-tolerant, error-corrected quantum computing stays an engineering problem, however progress is speedy. Large expertise firms and startups alike proceed to publish new processors, algorithms, and roadmaps. Quantum computing is now not only a theoretical curiosity, it’s an energetic, multidisciplinary race amongst physicists, engineers, and pc scientists to show unique quantum results into sensible benefit (Preskill, 2018).

2 Literature review

The literature on quantum computing is broad and rapidly evolving, spanning foundational theory, algorithms, hardware, applications, and software/verification tooling. Below we summarize key strands of work and representative references that capture both historical foundations and modern survey-level syntheses.

2.1 Foundational theories and early contributions

The theoretical foundations of quantum computation were laid in the late 20th century. Yuri Manin was among the first to suggest that computation could be performed according to the principles of quantum mechanics, introducing the concept of quantum automata in his 1980 work Computable and Uncomputable (Benioff, 1980). Shortly afterward, Paul Benioff proposed quantum-mechanical fashions of Turing machines, displaying how computation could possibly be embedded in quantum physics (Feynman, 1982). Richard Feynman famously argued that quantum programs are exhausting to simulate effectively on classical computer systems and proposed the concept of utilizing quantum units to simulate nature, a key motivation for quantum computing (Deutsch, 1985). David Deutsch formalized the notion of a common quantum pc and articulated the Church–Turing precept within the quantum setting, establishing that quantum mechanics might outline a brand new mannequin of computation (Shor, 1995). These foundational works set the conceptual stage for later algorithms and {hardware} efforts (Grover, 1996; Brassard et al., 2000). The historic improvement and main milestones of quantum computing are illustrated in Figures 1, 2.

2.2 Surveys of quantum algorithms and applications

Recent survey work synthesizes how quantum algorithms map onto real-world application areas—such as chemistry, optimization, cryptography, machine learning, and finance—carefully weighing theoretical speedups against practical resource costs and engineering constraints. Dalzell (2023) present a helpful, application-oriented perspective that emphasizes refined caveats concerning when quantum benefit really materializes and the necessity for end-to-end complexity issues (Dalzell, 2023). Contemporary literature stresses that theoretical asymptotics (e.g., Shor’s exponential speedup) have to be evaluated alongside necessities for fault tolerance, qubit counts, and real looking gate/noise budgets (Harrow et al., 2009; Preskill, 2018; Arute et al., 2019).

Beyond these, Huynh et al. (2023) discover quantum-inspired machine-learning approaches that bridge classical and quantum paradigms, providing hybrid algorithms deployable on as we speak’s near-term {hardware} (Huynh et al., 2023). Grigoryan et al. (2025) gives a complete evaluate of quantum-computing fashions—together with gate-based, adiabatic, and measurement-based approaches—analyzing their algorithmic implications and domain-specific applications (Grigoryan et al., 2025). Industry-oriented analyses printed in EPJ Quantum Technology (2021) emphasize how algorithmic progress interacts with {hardware} engineering, highlighting persistent gaps between theoretical quantum benefit and sensible scalability (EPJ Quantum Technology, 2021). Additionally, the Quantum Algorithm Zoo (2025) serves as an evolving catalogue of a whole bunch of quantum algorithms, categorized by area and computational mannequin, providing researchers a residing reference for monitoring progress throughout the sphere (Quantum Algorithm Zoo, 2025). Together, these surveys illustrate that whereas theoretical speedups stay intellectually compelling, their translation into sensible benefit relies upon critically on {hardware} maturity, hybrid algorithm design, and built-in benchmarking frameworks.

2.3 Qubit technologies and hardware development

Work on qubit platforms surveys the trade-offs among leading physical implementations: superconducting circuits (offering excellent gate speed and strong industrial support), trapped ions (high fidelity and long coherence but slower gates), photonic systems (room-temperature operation and flexible connectivity), neutral atoms, nitrogen-vacancy (NV) centers, and semiconductor quantum dots. Comprehensive reviews compare coherence times, gate fidelities, and error mechanisms across these platforms (Chae et al., 2024). These {hardware} surveys are essential for figuring out which algorithmic proposals are sensible for near-term units and which depend upon long-term fault-tolerant architectures (Peruzzo et al., 2014).

The IBM Quantum Roadmap (2024–2025) details the engineering milestones toward modular and error-corrected systems, highlighting chip-to-chip connectivity, cryogenic control, and scalable architectures (IBM Quantum, 2025; Quantum, 2025). The IBM Osprey processor (2023), a 433-qubit superconducting chip, demonstrated important progress in coherence management and fabrication uniformity (Preskill, 2018; IBM Quantum, 2023). Broader theoretical and experimental discussions proceed to border such programs inside the Noisy Intermediate-Scale Quantum (NISQ) paradigm (Preskill, 2018).

Beyond superconducting and ion-trap systems, hybrid hardware directions are emerging that integrate photonic, atomic, and NV-center qubits for specialized applications such as quantum chemistry and clinical computation (Lee et al., 2024; Sharma et al., 2025; Monroe, 2024; Goswami and Patel, 2025). Educational and technical sources describe two-qubit-gate physics and multi-qubit coupling methods which might be essential for scalable architectures (Quantum, 2024). Cross-platform benchmarking efforts emphasize that progress in algorithmic efficiency depends upon unified error metrics, calibration protocols, and real looking gate budgets (Gill et al., 2020; Grigoryan et al., 2025; Zhao and Kumar, 2024). Collectively, these analyses underscore that whereas {hardware} range stays a energy, convergence towards modular, fault-tolerant design is the important thing enabler for sustainable quantum benefit within the coming decade.

2.4 Verification, software stacks, and reliability

As quantum algorithms and hardware continue to mature, the verification of quantum programs and toolchain correctness has become a rapidly growing research area. Recent surveys summarize formal verification techniques specifically adapted to quantum settings, including model checking, deductive verification using proof systems, compiler-level equivalence checking, and error-correction verification frameworks (Lewis et al., 2021; Qafny, 2025; CoqQ, 2025; CertiQ, 2025; Veri-QEC, 2025). These approaches have been carried out in verification instruments resembling QPMC, PRISM, STORM, CoqQ, and Veri-QEC, every offering distinct trade-offs in scalability and rigor. Circuit equivalence checking can also be supported by devoted instruments such because the Quantum Equivalence Checker (QEC, 2025).

The literature highlights that ensuring quantum software correctness is significantly more challenging than in classical systems because of probabilistic outputs, exponentially large state spaces, entanglement dependencies, and complex noise models (Bauer-Marquart et al., 2022; Gill et al., 2020). To tackle these difficulties, researchers have launched symbolic-execution-based and SMT-solver-backed verifiers that automate the invention of purposeful and security-related defects. Deductive frameworks and theorem-prover embeddings (e.g., Qafny, QHLProver, and CoqQ) are being developed to boost formal reasoning about quantum applications, permitting stepwise verification of program semantics and circuit transformations (Qafny, 2025; CoqQ, 2025; QHLProver, 2025).

Recent work also explores the integration of verification techniques into quantum software stacks, ensuring correctness across compilation, optimization, and hardware execution layers (Shi, 2019; Grigoryan et al., 2025; Zhao and Kumar, 2024). Frameworks resembling CertiQ and symQV lengthen conventional compiler verification into the quantum area, validating equivalence between logical circuits and their optimized or transpiled types (Shi, 2019; Bauer-Marquart et al., 2022). Collectively, these research point out that formal verification might be a foundational element of future quantum software program engineering, bridging the hole between summary algorithm design and dependable {hardware} execution.

2.5 Application-focused and gap analyses

Domain-specific surveys and studies across finance, chemistry, logistics, and machine learning emphasize persistent gaps between theoretical quantum advantage and experimental feasibility. While algorithmic proposals demonstrate promising asymptotic speedups, end-to-end resource analyses are frequently incomplete, and assumptions about idealized, error-free hardware dominate much of the literature (Herman, 2022; Schuld, 2021). Verification and benchmarking stay at an early stage, with restricted experimental validation and inconsistent reporting of quantum sources (Dalzell, 2023; Grigoryan et al., 2025; Zhao and Kumar, 2024).

Recent reviews have highlighted that realistic quantum advantage demands hardware–software co-design, integrating insights from algorithm development, quantum control engineering, and compiler optimization (Lee et al., 2024; EPJ Quantum Technology, 2021; Quantum, 2025; Andersen et al., 2025). Studies in finance and logistics notice that drawback encodings and quantum data-loading overheads usually offset theoretical speedups, calling for clear useful resource estimation frameworks (Herman, 2022; Vidal et al., 2025). Similarly, in quantum chemistry and supplies science, Grigoryan et al. (2025) and associated works underscore the need of aligning algorithmic complexity with {hardware} noise and decoherence limits (Grigoryan et al., 2025; Zhang et al., 2024).

Emerging meta-analyses propose standardized benchmarking and reproducibility protocols for quantum algorithms—such as the QBench and QPack initiatives—which aim to quantify algorithmic efficiency relative to hardware constraints (Carter and Gheorghiu, 2024). Collectively, these findings level to a brand new part of quantum computing analysis targeted not solely on novel algorithms however on rigorous analysis, system-level integration, and interdisciplinary collaboration between theorists, experimentalists, and area consultants.

2.6 How to read these surveys and next steps for researchers

For newcomers, understanding the evolution of quantum computing requires a layered approach. Foundational papers introduce the conceptual basis: Benioff’s quantum Turing model, Feynman’s simulation arguments, and Deutsch’s universal quantum computer formalism remain essential for grasping the motivation and theoretical framework (Benioff, 1980; Feynman, 1982; Deutsch, 1985). Modern surveys resembling Dalzell et al. present an application-oriented overview, highlighting algorithmic alternatives and end-to-end complexity evaluation throughout a number of domains (Dalzell, 2023). Hardware critiques assist map algorithms to possible bodily platforms, connecting computational fashions with real-world architectures (Chae et al., 2024; IBM Quantum, 2025; Quantum, 2025). Verification and reliability research then bridge principle and engineering, outlining the constraints of quantum software program stacks and emerging formal instruments (Lewis et al., 2021; Qafny, 2025; CoqQ, 2025; Veri-QEC, 2025; Zhou et al., 2024).

Recent meta-analyses emphasize that bridging theoretical elegance with practical implementation is the next frontier for quantum computing research (Khabiboulline et al., 2025; Reynolds et al., 2025). By sequentially partaking with historic foundations, algorithmic surveys, {hardware} overviews, and verification frameworks, researchers can develop an built-in understanding of the sphere and determine tractable analysis challenges aligned with emerging industrial and scientific priorities.

3 Fundamentals of quantum computing

3.1 Qubits

In classical computing, information is stored in bits that take values 0 or 1. In quantum computing, the basic unit of information is the quantum bit or qubit, which can exist in a superposition of |0⟩ and |1⟩ states. Mathematically, a qubit’s state is expressed as |ψ⟩ = α|0⟩ + β|1⟩, where α and β are complex amplitudes satisfying |α|² + |β|² = 1 (Quantum, 2024). The energy of qubits lies of their potential to encode exponentially extra info because the variety of qubits will increase.

Different technologies realize qubits, including superconducting circuits, trapped ions, photonic qubits, topological qubits, and neutral atoms (Chae et al., 2024; Chohan, 2024; AbuGhanem, 2025; Zhu and Browne, 2025; Kim et al., 2025). Comparative research spotlight the trade-offs amongst these platforms—superconducting qubits supply speedy gate operations and mature fabrication pipelines, whereas trapped-ion and photonic qubits exhibit superior coherence and connectivity (Chen et al., 2024). Reviews additional study emerging supplies, hybrid processors, and topological error safety as important frontiers in scalable quantum computing (AbuGhanem, 2025; Zhu and Browne, 2025; Kim et al., 2025). A comparability of main qubit applied sciences and their key traits is proven in Figure 3.

3.2 Superposition

Superposition allows qubits to exist in linear combinations of basis states until measurement collapses them. Unlike classical probability, superposition enables interference—constructive and destructive—allowing quantum algorithms to amplify correct results (Chae et al., 2024). For instance, three qubits can concurrently signify eight states that may be manipulated collectively by quantum operations (IBM Quantum, 2025). The geometry of superposition on the Bloch sphere is proven in Figure 4, the place a qubit’s state |ψ⟩ = α|0⟩ + β|1⟩ is represented by polar and azimuthal angles θ and φ on the sphere’s floor (Quantum, 2024).

3.3 Entanglement

Entanglement is a uniquely quantum phenomenon in which two or more qubits become correlated such that the state of one determines the state of the others, regardless of spatial separation. This property underpins many quantum algorithms and communication protocols. A well-known example is the Bell state, (|00⟩ + |11⟩)/√2 (QPMC). Without entanglement, many quantum algorithms would lose their exponential benefit (Chohan, 2024; Zhu and Browne, 2025).

3.4 Differences from classical computing and challenges

Current devices therefore belong to the Noisy Intermediate-Scale Quantum (NISQ) period (Preskill, 2018; Chae et al., 2024; AbuGhanem, 2025).

3.5 Qubits and quantum phenomena: from physics to practical impact

Table 2 presents a chronological overview of the important thing milestones that formed quantum computing, from early theoretical discoveries to present-day {hardware} developments.

Quantum physical systems, such as the orientation of a photon or the spin of an electron, form the foundation for creating quantum bits (qubits). Quantum computers can thus exist in different configurations, including single-qubit systems (Niel et al., 2010), two-qubit programs (Quantum, 2024), and extra complicated multi-qubit architectures. A key historic milestone occurred within the early 2000s with the belief of the primary five-qubit quantum processor (Laflamme et al., 2001). Since then, progress has accelerated quickly; trendy programs, resembling IBM’s Osprey chip, now characteristic as much as 433 qubits (IBM Quantum, 2023).

Theoretical analyses suggest that achieving quantum supremacy—the point at which a programmable quantum computer can perform a task infeasible for classical computers—requires a minimum of roughly 50 entangled qubits (Preskill, 2018; Arute et al., 2019). Quantum supremacy was first demonstrated when quantum units carried out particular random-circuit sampling duties sooner than the world’s strongest supercomputers.

At a physical level, qubits mimic the quantum behavior of electrons orbiting an atomic nucleus, embodying three fundamental principles of quantum mechanics. The principles of superposition and entanglement, introduced earlier, manifest in physical systems that enable scalable quantum computation. A third key phenomenon—quantum interference—plays a vital role in amplifying correct computational outcomes and suppressing erroneous ones.

3.5.1 Quantum interference

At the subatomic level, particles exhibit wave-like behavior. When quantum states overlap, constructive interference amplifies the probability of correct computational paths, whereas destructive interference suppresses incorrect ones. This principle is central to algorithms such as Grover’s search and Shor’s factoring, where interference steers computation toward the optimal solution.

3.5.2 Applications and emerging frontiers

Quantum computing now spans diverse fields—cryptography, optimization, materials science, communication, and artificial intelligence. Recent practical algorithms leverage quantum principles to address complex real-world problems in finance, drug discovery, transportation, and weather prediction (illustrated in Figure 5).

A rapidly expanding area of application is disease diagnosis and healthcare. Quantum-enhanced machine-learning models have been used to analyze high-dimensional medical imaging and genomic datasets, accelerating the detection of diseases such as cancer, cardiovascular disorders, and neurodegenerative conditions. For example, quantum support-vector machines (QSVMs) and hybrid quantum–classical neural networks have shown promise in classifying MRI scans, predicting protein-folding patterns, and discovering biomarkers associated with complex diseases.

As hardware scalability and noise reduction improve, such applications are expected to mature from research prototypes to clinically reliable diagnostic tools. This underscores the broader vision of a physically scalable, fault-tolerant quantum paradigm—one capable of solving computationally intractable problems in healthcare and beyond.

4 Quantum algorithms

Quantum algorithms exploit quantum phenomena to achieve breakthroughs that are impossible or impractical with classical methods. Unlike classical algorithms that rely on binary logic, quantum algorithms leverage quantum mechanical properties to accelerate computation. Below are some of the most significant algorithms:

4.1 Shor’s algorithm—integer factorization

Shor’s algorithm, introduced in 1994, enables efficient factorization of large integers using the quantum Fourier transform (QFT). It reduces factoring to a periodicity problem, which quantum computers solve exponentially faster than classical ones. This poses a direct threat to RSA and other public-key cryptosystems. Its time complexity is polynomial, making it one of the clearest demonstrations of exponential quantum advantage, though it requires fault-tolerant quantum hardware with millions of physical qubits.

4.2 Grover’s algorithm—search

Grover’s algorithm provides a quadratic speedup for unstructured search problems. While classical search requires O(N) steps, Grover’s algorithm solves it in O (√N). The algorithm amplifies the amplitude of the correct solution through iterative oracle queries and diffusion operations. Although the speedup is quadratic, its applications extend to optimization, pattern recognition, and cryptanalysis.

4.3 Quantum fourier transform (QFT) and phase estimation (QPE)

QFT is a quantum analogue of the discrete Fourier transform, efficiently implemented in O (n^2) operations. It underpins Shor’s algorithm and spectral estimation tasks. Quantum Phase Estimation (QPE) uses QFT to estimate eigenvalues of unitary operators, making it central to quantum chemistry and material science simulations.

4.4 Variational quantum algorithms (VQAs)

VQAs are hybrid algorithms designed for NISQ devices, combining quantum state preparation with classical optimization. The Variational Quantum Eigensolver (VQE) estimates ground-state energies in molecular systems, while the Quantum Approximate Optimization Algorithm (QAOA) addresses combinatorial optimization problems. These algorithms are practical for current noisy hardware, though they face challenges such as barren plateaus and noise sensitivity.

4.5 HHL algorithm—linear systems solver

The Harrow-Hassidim-Lloyd (HHL) algorithm solves systems of linear equations exponentially faster than classical methods under certain conditions (sparse, well-conditioned matrices). However, extracting full classical solutions diminishes this advantage. It remains most useful in quantum machine learning and scientific simulations where the solution can be consumed in quantum form.

4.6 NISQ era algorithms and quantum advantage

The current NISQ era is characterized by noisy devices with limited qubits. Algorithms such as VQE and QAOA are promising for near-term utility, while long-term advancements require scalable error correction. Early demonstrations of quantum supremacy show the field’s potential, though practical, widely useful quantum advantage remains a frontier of research.

Early demonstrations of quantum supremacy have validated the potential of quantum algorithms such as Shor’s and Grover’s, but scalable fault-tolerant hardware remains a critical challenge.

Recent surveys also provide broader insights into the end-to-end complexity of quantum algorithms across diverse applications, highlighting practical limitations of quantum advantage. Dalzell (2023) current a complete comparability of algorithmic primitives, efficiency metrics, and real looking {hardware} constraints. Complementarily, newer critiques resembling Quantum Algorithms for Quantum Molecular Systems: A Survey (Lee et al., 2024) study algorithmic diversifications for chemistry and molecular programs inside fault-tolerant architectures.

Table 3 gives an built-in abstract of the important thing quantum algorithms mentioned above, together with their core mechanisms, computational advantages, and {hardware} compatibility.

5 Applications of quantum

5.1 Cryptography

5.1.1 Post-quantum cryptography, QKD integration, and quantum-safe infrastructure

5.2 Applications of quantum computing in various domains

5.3 Artificial intelligence & machine learning

5.4 Finance & business optimization

Portfolio optimization/asset allocation: Quantum (or quantum-inspired) optimization can evaluate many asset combinations in parallel to find more optimal returns vs. risk tradeoffs (Herman, 2022; IBM Quantum, 2025).

Derivative pricing, risk modeling, Monte Carlo simulations: Some quantum algorithms are suited for faster stochastic simulation and pricing of complex derivatives or measuring tail risks (VaR, CVaR) (Dalzell, 2023; Herman, 2022).

Credit scoring, fraud detection, loan risk assessment: Quantum machine learning methods may detect patterns in large data sets more efficiently in credit or fraud domains (Herman, 2022; Schuld, 2021).

Trading strategies, optimization for business operations and supply chains: Quantum approaches could optimize logistics, resource allocation, scheduling, route planning, and manufacturing. Several financial institutions are already experimenting with hybrid quantum-classical systems for trading or bond portfolio optimization (Herman, 2022; IBM Quantum, 2025).

5.5 Logistics & supply chains

Route optimization and vehicle routing problems: Quantum algorithms (e.g., QAOA, quantum annealing) can help solve large combinatorial optimization tasks like the Traveling Salesman Problem (TSP), vehicle routing, or delivery scheduling more efficiently (Farhi et al., 2014; Dalzell, 2023).

Inventory management and demand forecasting: Quantum methods might improve forecasting models and optimize inventory across multiple warehouses and dynamic demand (Dalzell, 2023).

Transportation and traffic flow: Companies like Volkswagen and DHL have begun pilot experiments for quantum-assisted traffic planning and logistics routing (Dalzell, 2023; IBM Quantum, 2025). Recent research have explored quantum cybersecurity, post-quantum cryptography, and reliability challenges in quantum software program and infrastructures (Das et al., 2024; Ong et al., 2024; Baseri et al., 2024; Faruk et al., 2022; Majdoubi et al., 2024; Mamun et al., 2024; Moral, 2024; Robert, 2024; Scrivano, 2025; Sonko, 2024).

6 Verification and reliability in quantum programs

As quantum hardware and algorithms scale in complexity, ensuring correctness and reliability becomes ever more critical. Quantum programs are fundamentally harder to test and debug than classical ones because of probabilistic measurement outcomes, noise and errors in hardware/software, and limited resources in NISQ (Noisy Intermediate-Scale Quantum) devices. Formal verification is one of the most promising ways to provide guarantees about quantum software correctness (Lewis et al., 2021; Shi, 2019; Bauer-Marquart et al., 2022; Gill et al., 2020). A comparability of main formal verification strategies for quantum applications, together with their strengths and limitations, is offered in Table 5.

6.1 Key challenges

Probabilistic Outcomes & Measurement Collapse: Quantum algorithms often produce probability distributions over outputs; small errors or noise can lead to wrong distributions, and measurement itself collapses superposition (Lewis et al., 2021; Shi, 2019).

Noise, Decoherence & Gate Errors: Real quantum hardware is imperfect. Errors can come from limited coherence times, imperfect gate operations, cross-talk, and measurement error (Preskill, 2018; Shi, 2019).

Resource Constraints in NISQ Devices: Limited qubit count, limited circuit depth, high error rates restrict how large/complex programs can be, which makes verification more difficult (Preskill, 2018; Lewis et al., 2021).

Complex State Space: Even a modest number of qubits gives a huge state space (dimension grows exponentially), which complicates exhaustive reasoning or simulation (Preskill, 2018; Lewis et al., 2021).

6.2 Formal verification approaches

To address these challenges, researchers have developed several formal methods adapted to quantum programs:

6.3 Verification workflow & best practices

Specify precisely: Write formal preconditions, postconditions, invariants, and expected behavior. Clear specification is essential (Qafny, 2025; CoqQ, 2025; Veri-QEC, 2025).

Modular design: Break large quantum programs or circuits into smaller, verifiable components (subroutines, logical qubits, etc.) (Qafny, 2025; CoqQ, 2025; Veri-QEC, 2025).

Use abstraction and compositionality: Abstractions can reduce complexity; compositional verification allows reasoning about parts separately (Qafny, 2025; CoqQ, 2025; Veri-QEC, 2025).

Combine automated tools with interactive proofs: Some parts may be checked automatically (SMT solvers, model checkers), others need manual proof or human insight (QPMC, 2025; PRISM, 2025; STORM, 2025; Qafny, 2025; CoqQ, 2025; QHLProver, 2025; SMT solvers–Z3, 2025).

Validate error models: The verification must assume realistic noise and error models; if the model is too optimistic, real-world reliability might differ (Veri-QEC, 2025).

Testing and simulation: Complement formal verification with simulations, randomized benchmarking, and hardware tests to cross-validate (Veri-QEC, 2025).

7 Quantum computing in finance and machine learning

7.1 Finance

Quantum Amplitude Estimation (QAE) for derivative pricing, risk analysis, portfolio valuation: A prime example is “Quantum Portfolio Value Forecasting,” where amplitude estimation is used to estimate the expected long-term value of a portfolio using a variant of the Gordon-Shapiro formula. The authors apply it to a 5-asset portfolio on actual trapped-ion quantum hardware, showing that for the same computational cost one can reduce statistical error compared to classical estimates (Herman, 2022; Schuld, 2021).

Also, “Quantum advantage for multi-option portfolio pricing and valuation adjustments (CVAs)” studies how quantum Monte Carlo methods can accelerate estimation of CVAs and similar risk-adjusted derivative valuations (Herman, 2022; Schuld, 2021).

Portfolio Optimization using QAOA & Higher‐Order Moments: A recent work “Higher-Order Portfolio Optimization with Quantum Approximate Optimization Algorithm” (Uotila et al., 2025) formulates portfolio optimization together with third and fourth moments (skewness, kurtosis). The drawback turns into a higher-order unconstrained binary optimization (HUBO) quite than a quadratic one, and QAOA is used to optimize it (Herman, 2022; Schuld, 2021). Their experiments (on many take a look at cases) present that together with increased moments yields higher portfolio allocations than classical baselines underneath sure settings.

Another example is “Enhancing Knapsack-based Financial Portfolio Optimization Using QAOA” (Huot et al., 2024), which frames the portfolio drawback as a knapsack drawback and makes use of QAOA with quantum stroll mixers (Herman, 2022; Schuld, 2021). They present aggressive approximate ratios for sure asset-selection constraints.

Challenges and realistic constraints: While quantum methods promise quadratic speedup in many Monte Carlo tasks (pricing, risk, VaR, CVaR), actual hardware, noise, and limited qubit counts often limit demonstration of that speedup. Furthermore, designing cost Hamiltonians, encoding classical data, and managing readout error are key issues. Surveys (e.g., Herman, 2022) present comparisons of various algorithmic strategies and spotlight that many monetary use instances are nonetheless in simulation or at small scale (Herman, 2022; Schuld, 2021).

7.2 Machine learning

Quantum Kernel Methods & Quantum Neural Networks (QNNs): Recent benchmarking studies (e.g., “Quantum kernel methods under scrutiny: a benchmarking study,” 2025) compare different quantum kernel variants (fidelity quantum kernels, projected quantum kernels) across many datasets for both classification and regression (Schuld, 2021).

Another important result is from “Supervised quantum machine learning models are kernel methods” by Schuld (2021). This work argues that many quantum fashions which might be referred to as “quantum neural networks” have mathematical buildings similar to kernel strategies (they embed information right into a high-dimensional Hilbert house through encoding circuits, then carry out studying through overlaps/interior merchandise of these embeddings) (Schuld, 2021).

Applications, robustness, and adversarial aspects: Studies like “Benchmarking Adversarially Robust Quantum Machine Learning at Scale” (West, Erfani, Leckie, etc., 2022) show that quantum variational classifiers (QVCs) can be more robust to classical adversarial attacks in certain settings, possibly owing to different features learned in quantum circuits (Schuld, 2021).

Current Trends & Limitations: There is growing interest in combining quantum methods with classical ML pipelines, such as feature embedding + kernel learning, hybrid QNNs, federated quantum neural networks, and quantum boosted models. However, limitations remain:

The overhead of encoding classical data into quantum states (Schuld, 2021).

The effect of noise and error accumulation (especially for deep circuits) (Schuld, 2021).

The question of when quantum methods actually outperform classical ones on real, large, high-dimensional data, not just toy or synthetic datasets (Schuld, 2021).

Benchmarking across data sets, comparing run-time, sampling cost, and hardware-noise trade-offs; reproducibility is also a concern (Schuld, 2021).

7.3 Hardware benchmarking

To address the need for quantitative assessment of quantum computing platforms, the following table provides a comparative overview of leading hardware systems as of 2024–2025. This includes metrics such as qubit count, gate fidelities, coherence times, and quantum volume.

8 Challenges and limitations

While quantum computing carries immense potential, several fundamental and engineering challenges must be overcome before practical, large-scale quantum computers become commonplace.

8.1 Quantum decoherence

Qubits are extremely fragile. Even minor interactions with the surrounding environment (electromagnetic noise, thermal fluctuations, stray photons, vibrations) can cause them to lose coherence—that is, the phase relationships between amplitude states collapse, turning quantum superpositions into classical mixtures. This process is known as decoherence (Preskill, 2018; Arute et al., 2019; Dalzell, 2023).

Decoherence places strict time limits on how long quantum computations can run before errors overwhelm the system. Mitigation techniques like dynamical decoupling, decoherence-free subspaces, and improved shielding help, but they do not fully eliminate the problem (Preskill, 2018; Dalzell, 2023).

8.2 Error correction and fault tolerance

Because qubits are noisy and imperfect, Quantum Error Correction (QEC) is essential. However, implementing QEC typically requires many physical qubits per logical qubit—sometimes hundreds or even thousands—just to protect one “ideal” qubit (Preskill, 2018; Dalzell, 2023; Chae et al., 2024).

The overhead in gates, measurement, syndrome extraction, and classical control is substantial. Achieving fault tolerance (meaning the system can continue correct operation despite errors) remains a grand engineering challenge (Preskill, 2018; Dalzell, 2023).

8.3 High cost & physical requirements

Many quantum platforms require extreme operating conditions—ultra-low temperatures (millikelvin via dilution refrigerators), ultra-high vacuum, electromagnetic shielding, cryogenic control electronics, and complex calibration systems (Preskill, 2018; IBM Quantum, 2025).

Moreover, cooling, isolation, and control infrastructure themselves consume energy and space, making the entire system bulky and costly (Preskill, 2018; IBM Quantum, 2025).

8.4 Limited qubit count and connectivity

Currently, even the most advanced quantum systems remain limited to hundreds or a few thousand physical qubits (or less) with nonideal connectivity (Preskill, 2018; IBM Quantum, 2025).

To solve real-world, large-scale problems, the field will likely require millions of qubits (or many logical qubits after error correction). Achieving that level of scaling, with high-fidelity inter-qubit operations and low cross-talk, is a central bottleneck (Preskill, 2018; IBM Quantum, 2025).

8.5 Verification, validation & debugging complexity

Verifying the correctness of large quantum programs is extremely challenging. The exponential state space, probabilistic outputs, and noise make classical simulation infeasible beyond modest sizes. Formal verification (via model checking, deductive proofs, equivalence checking) is still in early stages (Lewis et al., 2021; Gill et al., 2020).

Furthermore, debugging quantum circuits is nontrivial—observing intermediate states collapses them, and errors may manifest subtly due to interference effects (Lewis et al., 2021; Gill et al., 2020).

A recent survey by Hiremath et al. (2025) classifies quantum computing challenges into short-term ({hardware} noise, scalability, useful resource overhead) and long-term (fault tolerance, structure standardization, and interdisciplinary integration). Incorporating such frameworks helps delineate the layered nature of present and future quantum computing challenges.

9 Current status and future outlook

9.1 Leading players & roadmaps

IBM has published an explicit roadmap to progress toward fault-tolerant quantum computers, targeting hardware improvements and modular systems through 2033 (IBM Quantum, 2025).

Google maintains its Quantum AI roadmap, aiming to push hardware and algorithms forward toward practical quantum advantage applications (IBM Quantum, 2025).

Microsoft, Intel, Amazon are investing heavily in quantum ecosystems—development tools, cloud platforms, qubit research, and software stacks (Preskill, 2018; Dalzell, 2023).

Rigetti, IonQ, PsiQuantum, others are active in hardware innovation. For example, IonQ’s roadmap includes scaling to many thousands of qubits in coming years (Preskill, 2018; IBM Quantum, 2025).

9.2 Trends & predictions

Experts often forecast that quantum computers will initially complement classical systems, rather than fully replace them, targeting niche areas where quantum advantage emerges (Preskill, 2018; Dalzell, 2023).

Some believe within the next decade, we may see “quantum utility”—device-scale applications where quantum methods give practical gains in specific domains (e.g., chemistry, optimization) (Preskill, 2018; Dalzell, 2023).

Beyond that, fault-tolerant universal quantum computing may arrive in ∼15–20 years (or more), depending on engineering progress (Preskill, 2018; Dalzell, 2023).

Predictions are cautious: many believe scaling, error correction, verification, and cost are still vast hurdles before broad deployment (Preskill, 2018; Dalzell, 2023).

10 Conclusion

Quantum computing is not just another incremental technology—it represents a paradigm shift, enabling new computational capabilities rooted in quantum mechanics (Benioff, 1980; Feynman, 1982; Deutsch, 1985; Shor, 1995; Grover, 1996; Preskill, 2018; Dalzell, 2023). Its potential spans cryptography, supplies science, drug discovery, optimization, AI, finance, and past. Yet substantial challenges stay: decoherence, error correction, scaling, environmental and bodily constraints, and software program reliability (Preskill, 2018; Dalzell, 2023; Chae et al., 2024; IBM Quantum, 2025).

While the promise is huge, the trail is difficult. The coming years will seemingly see hybrid programs—quantum + classical—co-operating to unravel duties past attain as we speak (Preskill, 2018; Dalzell, 2023). Success would require co-design throughout physics, {hardware}, algorithms, and software program engineering, together with rigorous benchmarking and verification (Lewis et al., 2021; Gill et al., 2020). As scientists, engineers, and establishments proceed driving progress, we inch nearer to unleashing quantum’s full energy and remodeling what we regard as computationally attainable (Benioff, 1980; Feynman, 1982; Deutsch, 1985; Shor, 1995; Grover, 1996; Preskill, 2018; Dalzell, 2023).

The inclusion of latest research (Dalzell, 2023; Zaman et al., 2023; Hiremath et al., 2025; Sharma et al., 2025) strengthens the contextual grounding of this paper, making certain that the dialogue displays present developments and the newest views in quantum algorithm design, medical applications, and {hardware} scalability.

References

• AbuGhanemM. (2025). Superconducting quantum computers: who is leading the future?EPJ Quantum Technol.12, 102. 10.1140/epjqt/s40507-025-00405-7

• AkterM. S.Rodriguez-CardenasJ.ShahriarH.CuzzocreaA.WuF. (2023). Quantum cryptography for enhanced network security: a comprehensive survey of research, developments, and future directions. IEEE Big Data2021, 5408–5417. 10.1109/bigdata59044.2023.10386889

• AndersenL.BertaP.RebentrostC. (2025). Hardware–software co-design for quantum advantage: bridging algorithms and architectures. Nat. Rev. Phys.7, 450–464. 10.1038/s42254-025-00918-z

• AruteF.AryaK.BabbushR.BaconD.BardinJ. C.BarendsR.et al (2019). Quantum supremacy using a programmable superconducting processor. Nature574, 505–510. 10.1038/s41586-019-1666-5

• BaseriY.ChouhanV.HafidA. (2024). Navigating quantum security risks in networked environments: a comprehensive study of quantum-safe network protocols. Comput. Secur.142, 103883. 10.1016/j.cose.2024.103883

• Bauer-MarquartF.LeueS.SchillingC. (2022). symQV: automated symbolic verification of quantum programs. arXiv:2212.02267.

• BenioffP. (1980). The computer as a physical system: a microscopic quantum mechanical Hamiltonian model of computers as represented by turing machines. J. Stat. Phys.22, 563–591. 10.1007/BF01011339

• BrassardG.HøyerP.MoscaM.TappA. (2000). Quantum amplitude amplification and estimation. arXiv:quant-ph/0005055.

• CarterM.GheorghiuE. (2024). QBench and QPack: frameworks for benchmarking quantum algorithms. ACM Trans. Quantum Comput.6 (3). 10.1145/3691041

• CertiQ. (2025) CertiQ – compiler/circuit equivalence verification(arXiv).

• ChaeE.ChoiJ.KimJ. (2024). An elementary review on basic principles and developments of qubits for quantum computing. Nano Converg.11, 11. 10.1186/s40580-024-00418-5

• ChenS.-J.TsaiY. (2025). Quantum-safe networks for 6G. 2, 1. 10.69709/caic.2025.102135

• ChenJ.-S.NielsenE.EbertM.InlekV.WrightK.ChaplinV.et al (2024). Benchmarking a trapped-ion quantum computer with 30 qubits. Quantum8, 1516. 10.22331/q-2024-11-07-1516

• ChhetriG.SomvanshiS.HebliP.BroteeS.DasS. (2025). Post-quantum cryptography and quantum-safe security. arXiv:2510.10436.

• ChohanA. (2024). A comparative review of quantum bits: superconducting, topological, spin, and emerging qubit technologies. SSRN Prepr. Available online at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4979773.

• Colobridge Blog (2025). Quantum computing 2025 — comparison of leading Qubit technologies. Available online at: https://blog.colobridge.net/wp-content/uploads/2025/09/Comparison-of-Leading-Qubit-Technologies_%D0%B0%D0%BD%D0%B3%D0%BB-1-1024×536.jpg.

• CoqQ. (2025) CoqQ – deductive verification framework for quantum programs(arXiv).

• DalzellM. (2023). Quantum algorithms: a survey of applications and end-to-end complexities. Cambridge, UK: Cambridge University Press.

• DasA.SinghT.KumarP. (2024). QSEC: Quantum software error correction and certification framework. IEEE Trans. Quantum Eng.5, 5203012. 10.1109/TQE.2024.5203012

• DeutschD. (1985). Quantum theory, the church–turing principle and the universal quantum computer. Proc. R. Soc. A400, 97–117.

• EPJ Quantum Technology (2021). Industry quantum computing applications: bridging algorithms and hardware. Available online at: https://epjquantumtechnology.springeropen.com/articles/10.1140/epjqt/s40507-021-00114-x.

• FarhiE.GoldstoneJ.GutmannS. (2014). A quantum approximate optimization algorithm. arXiv:1411.4028.

• FarukM. J. H.TahoraS.TasnimM.ShahriarH.SakibN. (2022). A review of quantum cybersecurity. 10.1109/icaic53980.2022.9896970

• FeynmanR. P. (1982). Simulating physics with computers. Int. J. Theor. Phys.21, 467–488. 10.1007/bf02650179

• GillS. S.KumarA.SinghM.KaurK.BuyyaR.UsmanM.et al (2020). Quantum computing: a taxonomy, systematic review and future directions. Softw. Pract. Exp.50 (6), 1074–1106. 10.1002/spe.3039

• GoswamiS.PatelP. (2025). Hybrid qubit systems: integrating photonic, NV-center and superconducting technologies. J. Quantum Eng.3. 10.1088/2633-4356/ad19d2

• GrigoryanA.KumarS.PinheiroP. R. (2025). A review on models and applications of quantum computing. Computation7 (3), 39. 10.3390/computation7030039

• GroverL. K. (1996). “A fast quantum mechanical algorithm for database search,” in Proceedings of STOC, 1996. arXiv:quant-ph/9605043.

• HarrowW.HassidimA.LloydS. (2009). Quantum algorithm for linear systems of equations. Phys. Rev. Lett.103, 150502. 10.1103/PhysRevLett.103.150502

• HermanD. (2022). A survey of quantum computing for finance. arXiv:2201.02773.

• HiremathS.KumarP.SinghR. (2025). A literature survey on quantum computing in next generation challenges in circuit design and applications of future enabling technologies. Int. J. Comput. Appl.187 (1), 1–8.

• HuotC.HengS. V.KimT.-K.HanY. (2024). Quantum autoencoder for enhanced fraud detection in imbalanced credit card dataset. IEEE Access12, 169671–169681. 10.1109/ACCESS.2024.3496901

• HuynhL.HongJ.MianA.SuzukiH.WuY.CamtepeS. (2023). Quantum-inspired machine learning: a survey. arXiv preprint arXiv:2308.11269.

• IBM Quantum (2023). IBM osprey: 433-qubit quantum processor. IBM Research Blog. Available online at: https://research.ibm.com/blog/ibm-osprey.

• IBM Quantum (2025). “IBM Quantum Roadmap” (online). (IBM roadmap and 2024–2025 updates).

• KhabiboullineE.RomeroJ.KubicaA. (2025). Bridging theory and implementation in quantum computing: challenges and opportunities. Commun. ACM68 (7), 72–81. 10.1145/3679312

• KimD.YamamotoY.GillS. (2025). Photonic and hybrid quantum processors: integration challenges and opportunities. IEEE Trans. Quantum Eng.6, 5402311. 10.1109/TQE.2025.5402311

• KumarA. (2022). A brief history of quantum computing. Medium. Available online at: https://miro.medium.com/v2/resize:fit:1400/1*N3CWuTrxr-tg27I2_9Xrmw.png.

• LaflammeR.KnillE.ZurekW. H. (2001). Demonstration of a five-qubit quantum computer. IBM Research News Archive, IBM Research.

• LeeS.PatelD.WangY. (2024). Quantum algorithms for quantum molecular systems: a survey. WIREs Comput. Mol. Sci.15 (2), e70020. 10.1002/wcms.70020

• LewisM.SoudjaniS.ZulianiP. (2021). Formal verification of quantum programs: theory, tools and challenges. arXiv:2110.01320.

• MajdoubiC.MendiliS. E.GahiY. (2024). Quantum cryptology in the big data security era. Int. J. Adv. Comput. Sci. Appl.15. 10.14569/ijacsa.2024.0150761

• MamunA. A.AbrarA.RahmanM.SalekM. S.ChowdhuryM. (2024). Enhancing TCPS security: a shift to PQC. arXiv:2411.13023.

• MonroeT. (2024). Recent advances in trapped-ion quantum processors. Nat. Photonics18, 210–223. 10.1038/s41566-024-01987-y

• MoralJ. O. (2024). Cybersecurity in critical infrastructures: a PQC perspective.

• MoralesS.LeueE.Bauer-MarquartF. (2025). Hybrid-QEC: integrating classical and quantum verification for fault-tolerant compilation. npj Quantum Inf.11 (33), 1–14. 10.1038/s41534-025-00832-y

• NielsenM. A.ChuangI. L. (2010). Quantum computation and quantum information. 10th Anniversary Edition. Cambridge: Cambridge University Press.

• OngM.KwonJ.SchillingC. (2024). Survey of reliability challenges in quantum software stacks: testing, debugging, and verification. arXiv preprint arXiv:2409.06741.

• OttD. J.PeikertC. (2019). Post quantum cryptography migration. arXiv:1909.07353.

• PatelM.ZurekR. (2025). Integrated quantum hardware–software co-design: a roadmap toward practical quantum advantage. IEEE Access13, 78512–78528. 10.1109/ACCESS.2025.3459821

• PeruzzoA.McCleanJ.ShadboltP.YungM. H.ZhouX. Q.LoveP. J.et al (2014). A variational eigenvalue solver on a photonic quantum processor. Nat. Commun.5, 4213. 10.1038/ncomms5213

• PreskillJ. (2018). Quantum computing in the NISQ era and beyond. Quantum2, 79. 10.22331/q-2018-08-06-79

• PRISM (2025). PRISM – probabilistic model checking tool (ACM Digital Library).

• Qafny (2025). Qafny – proof system for quantum programs(arXiv).

• QEC (2025). Quantum equivalence checker – circuit equivalence tool(dcs.gla.ac.uk).

• QHLProver (2025). QHLProver – Quantum hoare logic prover(arXiv).

• QPMC (2025). QPMC – quantum program/protocol model checker(opus.lib.uts.edu.au).

• QuantumI. B. M. (2024). Understanding multi-qubit interactions and two-qubit gates. in IBM quantum learning. Available online at: https://quantum-computing.ibm.com.

• QuantumI. B. M. (2025). IBM quantum system two architecture and modular scaling approach. IBM Research Blog. Available online at: https://research.ibm.com/blog/ibm-quantum-system-two.

• Quantum Algorithm Zoo (2025). Comprehensive catalogue of quantum algorithms. Available online at: https://quantumalgorithmzoo.org/.

• Quantum Computing (2021). Pioneers of quantum computing. SlideShare Presentation. Available online at: https://image.slidesharecdn.com/quantumcomputing-211118090146/85/quantum-computing-5-320.jpg.

• ReynoldsD.BravyiT.LloydS. (2025). The next decade of quantum computing: from concept to scalable systems. Nat. Rev. Phys.7, 530–547. 10.1038/s42254-025-00943-y

• RobertW. (2024). Cryptographic techniques for IoMT security.

• Rodriguez-AlvarezN.Rodriguez-MerinoF. (2025). Performance and storage analysis of CRYSTALS-Kyber.

• SahuS. K.MazumdarK. (2024). Analysis of quantum cryptography applications.

• SchuldM. (2021). Supervised quantum machine learning models are kernel methods. arXiv:2101.11020.

• ScrivanoA. (2025). Comparative study of classical and post-quantum algorithms.

• SharmaN.VermaR.PatelT.WilkesB. G.PanugantiB. (2025). Applications of quantum computing in clinical care. Front. Med.12, 1573016. 10.3389/fmed.2025.1573016

• ShiY. (2019). CertiQ: a mostly-automated verification of a realistic quantum compiler. arXiv:1908.08963.

• ShorP. W. (1995). Polynomial-time algorithms for prime factorization and discrete logarithms on a quantum computer. arXiv:quant-ph/9508027.

• SMT solvers – Z3, (n.d.). Yices (ACM digital Library).

• SonkoS. (2024). Quantum cryptography and U.S. digital security.

• STORM (2025). STORM – probabilistic model checking tool (ACM Digital Library).

• UotilaV.RipattiJ.ZhaoB. (2025). Higher-order portfolio optimization with quantum approximate optimization algorithm. arXiv preprint arXiv:2509.01496.

• Veri-QEC (2025) Veri-QEC – verification framework for quantum error-correcting programs(arXiv).

• VidalT.RoserF.LewisM. (2025). Quantum optimization for finance and logistics: benchmarking real-world feasibility. npj Quantum Inf.11 (27). 10.1038/s41534-025-00790-4

• ZamanF.AliS.HussainM.KhanA. (2023). A survey on quantum machine learning: current trends, challenges, opportunities, and the road ahead. arXiv preprint arXiv:2310.10315.

• ZeydanE. (2025). Quantum technologies for beyond 5G and 6G networks.

• ZhangA.KimS.HaatajaM. P. (2024). Quantum chemistry algorithms under noise: resource estimation and scalability. J. Chem. Phys.160 (14), 244903. 10.1063/5.0198773

• ZhaoQ.KumarM. (2024). Benchmarking and noise characterization for heterogeneous quantum hardware. Phys. Rev. Appl.21 (4), 045012. 10.1103/PhysRevApplied.21.045012

• ZhouR.GheorghiuM.BrownK. (2024). QuVerify: a scalable framework for end-to-end verification of quantum programs. ACM Trans. Quantum Comput.6 (2), 1–22. 10.1145/3689127

• ZhuL.BrowneK. (2025). Topological qubits and quantum error protection: a review. Nat. Rev. Phys.7, 615–630. 10.1038/s42254-025-00976-4