Financial institutions face a dual challenge: quantum computers could eventually break the encryption protecting trillions in transactions, while simultaneously offering new ways to optimize portfolios and detect fraud. A major report from the Group of Seven (G7) central banks released this week outlined how quantum technologies could fundamentally reshape financial security, payments, and market infrastructure over the coming decade, prompting banks and regulators to begin preparing today for risks that may not fully materialize for years.

What Exactly Is the Quantum Threat to Banking?

The G7 Central Bank Quantum Technologies Working Group, co-chaired by the Banque de France and the Bank of Canada, identified a specific and urgent concern: sufficiently advanced quantum computers could eventually break the cryptographic systems that underpin global banking, payments, and digital communications. While no "cryptographically relevant quantum computer" currently exists, expert assessments increasingly suggest there is a "non-negligible probability" such systems could emerge within the next decade, though timelines remain highly uncertain because of scientific and engineering hurdles.

The threat extends beyond simple data theft. A particularly insidious attack strategy called "harvest-now, decrypt-later" allows adversaries to intercept and store encrypted financial information today with the expectation that future quantum systems could decrypt it years from now. This means sensitive data encrypted today could become vulnerable even if it remains secure for the next five to ten years. Digital signatures used to authenticate transactions, verify identities, and protect financial records could also become vulnerable over time if quantum systems undermine current public-key cryptography.

How Are Banks Planning to Defend Against Quantum Attacks?

The G7 report identifies post-quantum cryptography (PQC) as the most practical and scalable near-term approach for improving quantum resilience across financial infrastructure. Post-quantum cryptography refers to new encryption methods designed to resist attacks from quantum computers while still running on classical computer systems. The U.S. National Institute of Standards and Technology has spent years evaluating candidate quantum-safe algorithms for encryption and digital signatures, and standardization efforts are underway.

However, migrating to post-quantum cryptography will be far more complicated than a simple software upgrade. Financial institutions often rely on deeply interconnected systems containing cryptographic components embedded across hardware, software, payment protocols, and communications networks. Organizations may need to inventory cryptographic dependencies, test compatibility with legacy infrastructure, and coordinate changes across vendors, counterparties, and service providers. During the transition period, multiple cryptographic systems may operate simultaneously, potentially creating interoperability and operational challenges.

Steps to Prepare for Quantum-Safe Financial Systems

• Cryptographic Inventory: Financial institutions should begin cataloging where encryption is embedded across their systems, from payment protocols to communications networks, to understand the scope of migration needed.
• Legacy System Testing: Organizations must test post-quantum cryptography compatibility with existing infrastructure before full deployment, as many systems were built decades ago and may not easily accept new encryption standards.
• Vendor Coordination: Banks need to work with technology vendors, payment processors, and counterparties to ensure synchronized migration timelines and avoid creating security gaps during transition periods.

What About Quantum Computing's Upside for Finance?

Beyond the security threat, the G7 report also examined how quantum technologies could eventually influence financial operations themselves. Researchers and technology firms have spent years investigating whether quantum systems could improve optimization, simulation, and machine learning tasks that are difficult for conventional computers. Potential applications include portfolio optimization, risk analysis, fraud detection, liquidity management, and large-scale economic modeling.

However, the G7 repeatedly noted that most financial applications remain experimental and that classical systems continue to outperform quantum approaches in nearly all real-world settings. The report describes quantum processors as specialized accelerators rather than replacements for conventional computers, similar to how graphics processing units (GPUs) are used today for artificial intelligence and scientific computing workloads.

One major research area involves optimization problems, which are common throughout finance. These problems can include portfolio construction, transaction routing, and liquidity allocation across payment systems. Quantum approaches such as the Quantum Approximate Optimization Algorithm and quantum annealing systems are being tested for these tasks, but most experiments remain limited to small-scale or stylized models.

Forward-looking financial organizations are already beginning to explore quantum solutions for specific use cases. QuEra, a quantum computing company, highlighted potential applications including novel quantum machine learning algorithms that might uncover hidden trends in customer behavior and improve customer segmentation. In a few years, customers may utilize quantum computing for optimizing asset allocation to achieve the best risk-return trade-off. QuEra's 256-qubit Aquila machine represents the kind of neutral-atom quantum computer that could eventually help with financial forecasting and risk management tasks.

The G7 also examined quantum machine learning, which attempts to combine quantum computing with artificial intelligence techniques such as pattern recognition and optimization. The report noted that the goal is not to replace classical AI systems, but to explore whether certain computational sub-tasks could eventually benefit from quantum resources. The working group highlighted "quantum-inspired" techniques, which are classical algorithms based on ideas from quantum physics, as a potentially more immediate development path because they can run on existing infrastructure without requiring quantum hardware.

Why Is Quantum Machine Learning So Promising for Finance?

Recent research published in Nature reveals why quantum neural networks could eventually outperform classical AI systems for certain financial tasks. Quantum neural networks, typically implemented via parameterized quantum circuits, offer a potential route to quantum advantage in learning by exploiting the Fourier structure of quantum computations. Unlike classical deep neural networks, which suffer from "spectral bias" that limits their ability to learn high-frequency patterns, quantum neural networks can be designed to target task-relevant frequencies and avoid this limitation.

The key insight is that quantum neural networks learn by steering amplitudes through unitary evolution, fundamentally different from classical neural networks that rely on real-valued weights and differentiable activation functions. This difference means quantum models' outputs are governed by the periodic nature of quantum mechanics, allowing them to express patterns as a finite Fourier series. The set of accessible frequencies is determined by the encoding Hamiltonian, while trainable parameters determine the Fourier coefficients.

For financial applications, this means quantum machine learning could potentially identify market patterns and anomalies that classical AI systems miss. Quantum convolutional neural networks and quantum reinforcement learning approaches exploit frequency pinching to better capture high-frequency image features and structured value landscapes, which could translate to detecting subtle fraud patterns or market microstructure effects.

However, practical implementation remains challenging. Current quantum computers are noisy intermediate-scale quantum (NISQ) devices, constrained by a limited number of qubits and highly susceptible to noise and decoherence. The field faces a significant optimization bottleneck known as the "barren plateau" problem, where exponentially vanishing gradients result in an untrainable, flat loss landscape. Researchers are developing mitigation strategies based on Lie-algebraic and Fourier perspectives to address this challenge.

The timeline for practical quantum advantage in finance remains uncertain. The G7 report stops short of making specific predictions about when practical quantum computers will emerge or calling for specific regulations. Instead, the working group presented what it describes as a "structured analytical framework" for understanding the risks, opportunities, and trade-offs associated with quantum technologies as they move from laboratories into early-stage deployment.

For now, the financial sector's best strategy appears to be twofold: immediately begin preparing cryptographic defenses against future quantum computers while simultaneously investing in research to understand how quantum systems might eventually improve financial operations. The institutions that begin this preparation today will likely be best positioned to capitalize on quantum computing's opportunities while minimizing its risks.