Quantum computing and AI: what it means for businesses

Quantum computing meets AI: Google Willow, hybrid architectures, and new test centers for midsize companies. What midsize companies need to know and prepare for now.

By SIMO GmbH

Published on: March 7, 2026
Reading time: 11 minutes
Category: Future Technologies

Quantum computing was long a topic for physics labs and research reports that nobody read. That is changing right now. Google has demonstrated a verifiable quantum advantage with the Willow chip for the first time, McKinsey projects a market of $97 billion by 2035, and the German federal government is investing over €5 billion in quantum technologies. At the same time, an IBM study shows: 59 percent of executives believe that quantum AI will transform their industry by the end of the decade—but only 27 percent expect their own company to use quantum computing by then. This gap between expectation and preparation is the greatest risk for the Mittelstand (privately held midsize companies). This article explains how quantum computing relates to AI, what breakthroughs in 2025 and 2026 have changed the landscape, and what companies should concretely do now.

What quantum computers are—and what they are not

The basic idea in three sentences

Classical computers calculate with bits that can be either 0 or 1. Quantum computers use qubits that, through superpositions, can be 0 and 1 simultaneously. This allows certain calculations to be performed not sequentially but in parallel—with an exponential speed advantage for specific problem classes.

What quantum computers are not

They are not a replacement for classical computers. They do not solve every problem faster. For word processing, email, or ERP systems, they offer no advantage. Their value lies in specific domains: optimization problems, molecular simulations, cryptography, and—this is the point that affects businesses—the training and acceleration of AI models.

The breakthroughs from 2024 to 2026: what has changed

Google Willow: the first verifiable quantum advantage

In December 2024, Google introduced the Willow chip—a 105-qubit processor that set two milestones. First: Willow solved a random circuit sampling benchmark in five minutes that the world’s fastest supercomputer would need 10 to the power of 25 years to complete. Second—and this is technically more significant: the more qubits Willow uses, the lower the error rate becomes. From 3x3 through 5x5 to 7x7 qubit grids, the error rate halved each time. This is the so-called “below threshold” error correction, the goal pursued by quantum research for decades.

In October 2025, the next step followed: with the “Quantum Echoes” algorithm, Google demonstrated the first verifiable quantum advantage for a real-world problem. A 65-qubit subsystem of the Willow chip calculated molecular structures 13,000 times faster than classical supercomputers. In a proof-of-concept with UC Berkeley, structures of molecules with 15 and 28 atoms were analyzed—with results that conventional NMR analyses could not deliver.

The implication for industry: quantum-accelerated molecular simulation could become a tool in pharmaceutical research and materials science—two fields where midsize suppliers and chemical companies are directly affected.

Hardware milestones worldwide

Company: Google | Milestone 2025/2026: 105-qubit Willow, below-threshold error correction | Next Step: 10–100 logical qubits by 2029
Company: IBM | Milestone 2025/2026: Quantum System One in Ehningen (Germany), operated by Fraunhofer | Next Step: Starling roadmap: 200 logical qubits
Company: Fujitsu/RIKEN | Milestone 2025/2026: 256-qubit superconducting quantum computer (April 2025) | Next Step: 1,000-qubit system by 2026
Company: IonQ | Milestone 2025/2026: Tempo architecture with 100 physical qubits | Next Step: 256 qubits by end of 2026
Company: QuEra Computing | Milestone 2025/2026: $250 million funding (Google, NVIDIA, Softbank) | Next Step: Neutral-atom quantum computer for industrial use

The market in numbers

McKinsey’s Quantum Technology Monitor 2025 puts the global market potential for quantum technologies at up to $97 billion by 2035. The current market volume is $1.79 billion (2025) with an annual growth rate of 31.6 percent.

The quantum AI market—the intersection of quantum computing and artificial intelligence—is growing even faster: from $3.05 billion (2025) to a projected $10.71 billion by 2030, at a CAGR of 28.5 percent.

Why quantum computers are relevant for AI

The problem: AI is hitting limits

Large language models like GPT-4 or Claude consume enormous computing resources during training. Training a model of the current generation costs an estimated $100 million and consumes as much energy as a small city in one year. This scaling logic—more data, more parameters, more compute—is finite. At some point, costs become unbearable and energy resources scarce.

The solution: hybrid quantum-AI architectures

McKinsey’s 2025 report confirms: quantum computing addresses precisely the core limitations of AI—algorithmic efficiency, memory limits, and compute bottlenecks. The future lies not in pure quantum systems but in hybrid architectures where quantum hardware serves as an accelerator for classical computations and AI.

Concretely, this means:

Faster AI training: Quantum processors accelerate AI model training, especially with limited datasets or high computational complexity
Optimization tasks: Quantum algorithms solve combinatorial optimization problems—route planning, portfolio optimization, production scheduling—exponentially faster
Energy efficiency: Analog quantum computers offer a more sustainable computing path for AI workloads
Modular integration: Engineers can integrate quantum optimization layers as modules into existing AI stacks without rebuilding the entire architecture

SAP has already put it succinctly: companies must adapt their AI strategies to use quantum computing capabilities. It is not enough to simply transfer existing AI models to quantum computers. New approaches—and new competencies—are needed.

Germany as a quantum hub: investments and infrastructure

5.2 billion euros in public funding

According to McKinsey, Germany is the third-largest public investor in quantum technologies worldwide—after the US and China. The federal government has committed over $5.2 billion in funding, with the “Action Concept for Quantum Technologies” earmarking approximately €2.8 billion through 2026.

Infrastructure for midsize companies

Three developments are particularly relevant for midsize companies:

1. IBM Quantum System One in Ehningen: The first IBM quantum computer in Europe is located in Baden-Wuerttemberg, operated by the Fraunhofer-Gesellschaft. All data remains in Germany and is subject to German data protection law. For companies working with sensitive data, this is a decisive factor.

2. Quantum computing test centers for midsize companies: Test and advisory centers funded by the Federal Ministry for Research launched in 2025 with a joint workshop. The QUICS project (GWDG and Leibniz University Hannover) provides midsize companies with low-barrier access to quantum resources through a web portal. In the first half of 2026, an open call will be conducted to integrate additional use cases from various industries.

Dr. Christian Boehme (GWDG) puts it clearly: “Small businesses and the Mittelstand are the backbone of the German economy. Without their active participation, we will not be able to take a leading role in quantum technology.”

3. Munich Quantum Valley and PlanQK: The Munich Quantum Valley (MQV), an initiative of the Fraunhofer-Gesellschaft, is creating industrial infrastructure for quantum applications. In the PlanQK project, 18 partners are researching quantum AI platforms. At the Juelich Research Center, there is a D-Wave annealing machine with over 5,000 qubits.

German companies in the quantum market

Germany is represented in the VanEck Quantum Computing UCITS ETF with Siemens, Infineon, and Deutsche Telekom at over 14 percent weighting—the second-strongest country position after the US. In addition, there are startups and research companies like Planqc, IQM, HQS Quantum Simulations, and Q.ANT.

Post-quantum cryptography: why you must act now

One aspect many companies overlook: quantum computers will be able to break today’s encryption. Not tomorrow, but within foreseeable time. The BSI (German Federal Office for Information Security) explicitly recommends planning for post-quantum cryptography (PQC) early.

The threat is not abstract. Attackers are already copying encrypted data today—in anticipation of decrypting it in a few years with quantum computers. This strategy is called “Harvest Now, Decrypt Later” and affects every company that stores or transmits sensitive data today.

Regulatory authorities worldwide are increasingly publishing standards for quantum-safe cryptography. PQC solutions will be broadly rolled out in 2026. Companies that do not plan now risk costly retrofitting later.

Practical guide: what midsize companies should do now

1. Build awareness

Quantum computing is not a topic for the IT department alone. Management and business units need to understand which problem classes quantum computers can solve—and which they cannot.

2. Identify use cases

Where does your company have optimization problems? Route planning, material procurement, production control, risk analysis? These are the areas where quantum computing will deliver tangible value first.

3. Use test centers

The new federally funded quantum computing test centers are specifically designed for midsize companies. Take advantage of the open call in the first half of 2026 to contribute your own use cases.

4. Make your AI strategy quantum-ready

If you are building an AI strategy today, think in terms of hybrid architectures. SAP’s recommendation is clear: existing AI models cannot simply be transferred to quantum computers. Modular architectures that can integrate quantum subroutines are needed.

5. Plan for post-quantum cryptography

Have your encryption infrastructure reviewed. The BSI recommends planning PQC migration paths now—not after quantum computers become powerful enough.

6. Build competencies

Quantum literacy will become more important for decision-makers than quantum physics for engineers. Understanding what is possible, where the limits lie, and which providers deliver serious solutions—that is the competency that counts.

Frequently asked questions

What can a quantum computer do that a regular computer cannot?

Quantum computers solve certain problem classes exponentially faster than classical machines. These include combinatorial optimization (e.g., route planning for 1,000 vehicles), molecular simulations (e.g., materials research, drug development), and breaking current encryption methods. For standard tasks like word processing or accounting, they offer no advantage.

When will quantum computers be usable for businesses?

Hybrid systems that use quantum hardware as an accelerator for classical computations are already available—for example, through IBM’s Quantum System One in Ehningen. For broad-based industrial use, experts see the timeframe of 2028 to 2032. 2026 is the year for groundwork: proofs of concept, test centers, and strategy development.

Is quantum computing a threat to IT security?

Yes. Quantum computers will be able to break common encryption methods like RSA and ECC. The BSI recommends planning for post-quantum cryptography now. The “Harvest Now, Decrypt Later” strategy—attackers collect encrypted data today to decrypt it later with quantum computers—makes the topic relevant already.

What does getting started with quantum computing cost for a midsize company?

The new federally funded test centers offer low-barrier access. Cloud-based quantum computing services (IBM Quantum, Amazon Braket, Google Quantum AI) allow initial experiments without owning hardware. The real investment lies in building competencies and identifying suitable use cases—not in hardware.

How are quantum computing and AI connected?

Quantum computers can train AI models faster, compute more complex optimizations, and work more energy-efficiently. The future lies in hybrid architectures where quantum modules are integrated as accelerators in classical AI pipelines. At the same time, AI supports quantum computing: machine learning improves error correction and software development for quantum systems.

References

How our articles are created and who is accountable for them is set out in our editorial standards.