From Experimental Qubits to Scalable Fault Tolerance
The journey toward practical quantum computing has been a decade-long endeavor, marked by significant investment and technical breakthroughs. Initial development centered on expanding the quantity of physical qubits inside processors, which yielded what is termed Noisy Intermediate-Scale Quantum (NISQ) hardware.
However, physical qubits are highly susceptible to errors due to their delicate quantum states. This challenge has driven extensive research into Quantum Error Correction (QEC), shifting the focus from raw physical qubit count to the development of “logical qubits.”
A logical qubit is a solid, error-corrected unit formed by entangling many physical qubits together. This redundancy helps mitigate the high error rates of individual physical qubits, paving the way for fault-tolerant quantum computing (FTQC) — systems that can operate reliably over extended computations.
Google demonstrated significant progress in December 2024 with its Willow processor, which showed that error-corrected logical qubits improve exponentially with more physical qubits. As an industry pioneer, IBM intends to launch Starling, a fully fault-tolerant quantum processor packing 200 logical qubits, by 2029, backed by a solid history of reaching its quantum computing milestones.
Superconducting qubits, the foundation for IBM and Google’s processors, operate at temperatures near absolute zero. Yet, the field explores diverse qubit technologies, or modalities. Intel has developed silicon spin qubits, which could leverage existing semiconductor fabrication techniques.
PsiQuantum and Xanadu are advancing photonic qubits, which use laser light squeezed in ring resonators. PsiQuantum, for instance, announced its Omega photonic chipset in February 2025, manufacturable with traditional lithography, and is building a utility-scale photonic quantum computer in Australia, breaking ground in June 2026.
Neutral atom qubits, made from individual atoms manipulated by lasers in a vacuum, are also showing promise, with Atom Computing demonstrating quantum error correction in June 2026. QuEra plans a 256-logical-qubit fault-tolerant system called Libra by 2028, based on over 10,000 physical qubits.
Trapped ion qubits, held by electromagnetic fields, are another mature modality, with IonQ and Quantinuum reporting repeatable error correction demonstrations in June 2025, targeting fault-tolerant systems by 2029. Even Microsoft is pursuing topological qubits, which are based on exotic states of matter, revealing its second processor, Majorana 2, in June 2026, which maintained quantum states 1,000 times longer than its predecessor from 16 months prior.
This rapid development underscores the commitment to realizing practical quantum computing. In the past 12 months alone, for example, IBM and the US Department of Commerce announced a quantum wafer foundry, signaling a concerted effort to scale manufacturing capabilities.
As the technology channel ExplainingComputers highlights, recent progress marks a continuation of an intense period of development, with the channel’s 10th annual update on quantum computing progress illustrating a decade of sustained research and investment. Even D-Wave Systems, recognized for its focus on quantum annealing, has adopted a gate model roadmap toward fault-tolerant quantum computing, underscoring broader industry alignment on how scalable quantum computation must proceed.
Impact and Accessibility: Where Quantum Speed Matters Most
The unique speed advantage of quantum computing translates into capabilities that can redefine scientific discovery and industrial optimization. Molecular modeling and simulation, commonly referred to as quantum chemistry, stands as the premier anticipated use case for quantum computers.
Conventional supercomputers face immense challenges when attempting to model the quantum mechanics of subatomic particles inside an atom, a shortfall famously noted in 1981 by physicist Richard P. Feynman, who observed that simulating a quantum system demands a quantum system. IBM simulated a molecule as early as 2017.
In 2023, Microsoft introduced Azure Quantum Elements, combining HPC and AI to speed up research in material and chemical science, readying researchers for an era in which scaled quantum machines can accurately simulate intricate molecules and deepen our grasp of biology, physics, and chemistry. These advancements promise breakthroughs in material science, engineering, and healthcare, including areas like improved batteries for the energy sector.
Outside of simulations, executing intricate optimization tasks represents the second most eagerly awaited use for quantum computing. This speed can dramatically enhance efficiency in diverse sectors, from optimizing power grids to improving financial modeling and logistics in transportation. Such applications promise significant commercial value across various business sectors, as noted in a March 2026 OECD paper.
Another critical area where quantum speed poses a profound impact is cryptography. Future quantum computers will possess the capability to break much of the existing digital encryption. A March 2026 white paper and blog post from Google regarding quantum vulnerabilities revealed that compromising 256-bit elliptic curve encryption will require 20 times fewer qubits than earlier estimates suggested.
Driven by this urgency, Google established a 2029 deadline to transition to post-quantum cryptography (PQC), catalyzing broad efforts to design quantum-safe algorithms and accelerate their widespread deployment across the coming five years. Quantum Computing Threats Accelerate Encryption Breakdown discusses these vulnerabilities in detail.
Despite the complexity and high cost of developing quantum hardware, accessibility is growing through Quantum as a Service (QaaS). IBM led the way in May 2016 when it provided public internet access to an experimental quantum processor.
Since then, numerous providers have followed including:
- D-Wave Systems with its Leap service in 2018
- Amazon with Amazon Braket on AWS in 2019
- Xanadu with its photonic cloud platform in 2020
Today, QaaS functions as both a research tool and the likely primary access point for early commercial fault-tolerant quantum computers.
Cloud providers like Amazon Braket facilitate access to hardware from companies such as AQT, IonQ, IQM, QuEra, and Rigetti. Similarly, Microsoft’s Azure Quantum provides access to IonQ, Pasqal, Rigetti, and Quantinuum hardware.
Although access to Google’s quantum computing service remains limited, various organizations supply complimentary open-source frameworks such as IBM’s Qiskit, Google’s Cirq, and the Microsoft Quantum Development Kit to help programmers hone their quantum expertise. How to Access Quantum Computing for Everyone further explores these avenues.
The quantum computing industry, which generated $1.4 billion in annual revenues in 2025, is projected to grow to over $3 billion by 2028. McKinsey estimates that quantum computing may generate between $1.3 and $2.7 trillion in global economic value by 2035, pointing toward an era where this unique processing capability fundamentally transforms commercial sectors.
How quantum acceleration will augment classical hardware
Quantum computing is faster not in the conventional sense of raw clock speed, but through its ability to fundamentally process information differently. By leveraging superposition and entanglement, qubits can simultaneously explore an exponential number of computational states, making them uniquely capable of tackling problems that remain intractable for even the most powerful classical computers.
This advantage is highly specific, primarily geared towards complex simulations, advanced optimization, and future cryptographic applications. The industry, now a decade into its major development phase, has seen billions invested, and while still a long-term research endeavor, it is rapidly progressing towards fault-tolerant systems.
It is very likely that fault-tolerant quantum systems will support novel classes of computational workloads during the 2030s. They will augment classical hardware, providing specialized acceleration rather than a universal replacement, defining a new era of computational possibilities where previously unsolvable problems finally yield their secrets.
