Quantum computing is no longer purely an academic subject. Governments treat it as a strategic technology, the largest cloud providers sell access to real quantum hardware by the hour, and banks, pharmaceutical companies and national laboratories are running experiments on it. What it is not, yet, is a production tool that outperforms classical computers on everyday business problems.

That gap between “available” and “advantageous” is the most useful way to understand where the field stands. The early practice happening today is about learning, benchmarking and preparing, with a small number of narrow demonstrations where quantum hardware has done something classical machines cannot easily reproduce. This article focuses on that practice: what is deployed, who is using it, and what a business can sensibly do now.

What “early practice” looks like today

Three delivery models have emerged. The most common is cloud access: IBM Quantum, Amazon Braket and Microsoft Azure Quantum let developers submit programs to real quantum processors from several hardware makers, alongside simulators that run on classical machines. This is how most companies run their first experiments, because it requires no capital investment.

The second is dedicated on-premises systems at research institutions. IBM installed a Quantum System One at Cleveland Clinic for healthcare research in 2023, and in June 2025 deployed its first Quantum System Two outside the United States at Japan’s RIKEN institute, where it is connected to the Fugaku supercomputer. These installations reflect a broader trend: quantum processors are being treated as accelerators attached to classical high-performance computing, not as standalone replacements.

The third model is annealing systems, such as those from D-Wave, which are designed for specific optimization problems and are available through the cloud and on premises. They work differently from gate-based machines and are best evaluated on their own terms.

The applications moving first

The original promise of quantum computing spans cryptography, simulation and optimization. In practice, each is at a different stage.

• Simulation of molecules and materials. Nature is quantum mechanical, so quantum computers are naturally suited to modeling chemical reactions, catalysts and new materials. This is widely regarded as the most promising near-term application, and chemical, pharmaceutical and energy companies are running exploratory studies. Useful results at industrial scale will likely require error-corrected machines.
• Optimization and finance. Banks and logistics firms are testing quantum and hybrid approaches for portfolio optimization, pricing and routing. In March 2025, JPMorgan Chase, Quantinuum and US national laboratories published in Nature a demonstration of certified randomness, generating random numbers whose quality can be mathematically verified, which is one of the first tasks with a plausible commercial use that classical computers cannot easily replicate.
• Cryptography. This is where quantum has the largest near-term business impact, but mostly in reverse. A sufficiently large, error-corrected quantum computer could break today’s public-key encryption. No such machine exists yet, but the response is under way: NIST published its first post-quantum cryptography standards in August 2024, and organizations are beginning migrations now.

Where the hardware really stands

Today’s machines have from dozens to a few hundred physical qubits, and those qubits are noisy. Errors accumulate quickly, which limits how long a computation can run. The field’s central goal is error correction: combining many physical qubits into more reliable “logical” qubits.

Progress here has been real. In December 2024 Google reported that its Willow chip reduced errors as it scaled up the error-correcting code, a long-sought milestone. In June 2025 IBM published a roadmap targeting a large-scale, fault-tolerant system called Starling by 2029. Several other companies, using trapped ions, neutral atoms and photonics, have their own roadmaps, and the US DARPA Quantum Benchmarking Initiative is evaluating which approaches can reach utility-scale machines.

In the meantime, most useful work runs as a hybrid workflow. A classical computer prepares the problem, a quantum processor handles a small, hard sub-step, and classical software checks and refines the result. That is why the software around the hardware, including toolkits, error-mitigation techniques and integration with existing cloud and supercomputing environments, matters as much as qubit counts for anyone evaluating vendors today.

The honest summary: timelines are uncertain, and most credible roadmaps put broadly useful fault-tolerant machines in the late 2020s or 2030s. Planning should account for that range rather than a single date.

Practical steps for businesses now

Most organizations do not need a quantum team. They do need a plan proportionate to their exposure and opportunity.

1. Start the cryptography inventory. List where public-key encryption protects long-lived sensitive data, such as health records, contracts and intellectual property. Attackers can harvest encrypted data today and decrypt it later. Our article on quantum computing and cryptography covers this risk in detail.
2. Ask vendors about post-quantum plans. Browsers, VPNs, cloud providers and identity platforms are adding post-quantum algorithms. Make it a procurement question.
3. Identify candidate problems. If your business depends on hard optimization, molecular simulation or complex risk modeling, note these as candidates for future evaluation.
4. Run a low-cost cloud experiment. A small team can learn a lot from a few weeks on a cloud quantum service using open-source toolkits such as Qiskit. The goal is understanding, not advantage.
5. Benchmark against classical methods honestly. Many claimed quantum speedups disappear against a well-tuned classical algorithm. Insist on that comparison.
6. Revisit annually. The field moves quickly. Assign someone to track hardware milestones and reassess your plan each year.

The trade-off is simple: spending heavily on quantum applications now is usually premature, but ignoring the cryptography implications is not. Our cybersecurity and compliance services include post-quantum readiness assessments.

Update (September 2026): Momentum continued after this article was first published. In late 2025 HSBC reported using IBM quantum hardware in a bond-trading trial, Google described a verifiable quantum advantage result on Willow, and Quantinuum launched its Helios system. These are meaningful steps, but they remain early demonstrations rather than routine production workloads.

Frequently asked questions

Can my business use a quantum computer today?

Yes, through cloud services such as IBM Quantum, Amazon Braket and Azure Quantum, often at low cost for small experiments. What you should not expect yet is better results than classical computing for most business problems.

Which industries will benefit from quantum computing first?

Chemistry, materials science and pharmaceuticals are widely expected to benefit first, followed by finance and logistics for certain optimization and risk problems. Every industry is affected by the cryptography implications.

How long until quantum computers break current encryption?

No one knows precisely. Estimates generally range from the late 2020s to well into the 2030s. Because encrypted data stolen today could be decrypted later, organizations with long-lived sensitive data should start post-quantum planning now.

Prepare without overspending

Delana Technologies helps organizations separate quantum hype from practical priorities, starting with post-quantum cryptography readiness and vendor assessment. To discuss your roadmap, call 239.414.5126 or contact us.

---

Sources: IBM announcements on Quantum System Two at RIKEN (June 2025) and the Starling fault-tolerance roadmap (June 2025); Google Quantum AI, Willow announcement (December 2024); Nature, “Certified randomness using a trapped-ion quantum processor” (March 2025); NIST FIPS 203, 204 and 205 (August 2024); Constellation Research, 2025 quantum computing year in review; HSBC and IBM bond-trading trial announcement (September 2025).