23 September 2026
Quantum AI examples
Moving from hype to real-world impact
Waynette Tubbs
Insights Editor, SAS
Quantum computing has seemed theoretical – even out of reach – for years. Now it’s gaining momentum in real-world applications, and organizations are asking: What problems can we solve with it in our organization?
According to research from IDC, 30% of global decision-makers say they’re familiar with quantum AI, and 25% say they trust it. This is a change from just a few years ago. Part of that shift comes from progress in quantum hardware – more stable qubits, early advances in error correction and growing integration with high-performance computing environments.
But just as important is the emergence of hybrid quantum-classical approaches, where quantum systems and classical computing work together, each handling the parts of a problem they solve best.
Where does quantum AI fit?
Most business problems sit in the messy middle – too complex for traditional optimization alone but not able to wait for fully fault-tolerant quantum systems. The opportunity lies in combining both approaches: classical systems for preprocessing and deployment, and quantum methods to explore complex relationships and decision spaces.
This shift has already started. Bryan Harris, Executive Vice President and Chief Technology Officer at SAS, pointed to a recent consumer packaged goods (CPG) example of a 97% reduction in processing time using a hybrid quantum-classical approach as proof. "Many people believe quantum is technology for the future. I say quantum is now. It’s right here, right now, and we’re doing it.”
Organizations are already testing where quantum artificial intelligence creates measurable business value – especially where scale and complexity push classical systems to their practical limits.
To help you decide where this technology fits in your organization, we’ve collected some real-world examples where organizations are already starting to find real value in quantum AI.
What is classical computing versus quantum AI?
Classical systems are used for preprocessing and deployment, whereas quantum methods are used to explore complex relationships and decision spaces. Hybrid quantum-classical approaches, on the other hand, combine quantum systems and classical computing, with each handling the parts of a problem they solve best.
Quantum AI examples in financial services: When 'good enough' is not enough
Financial institutions have already optimized for speed. Now they need to understand complexity – especially in areas like fraud prevention and detection. In this case, shifting patterns across accounts, devices and geographies create decision challenges that traditional AI models struggle to evaluate in real time.
“We see patterns of problems that have a highly combinatorial search space with complex relationships between predictors. The quantum physics used in quantum computing allows you to explore these differently from classical computing,” says Bill Wisotsky, Principal Quantum Architect at SAS.
Organizations face the same challenge in bankruptcy prediction. Bankruptcy modeling is a highly regulated problem, where prediction accuracy directly affects lending decisions. And machine learning models must perform under strict regulatory scrutiny.
Consumer bankruptcy modeling
SAS and QuEra Computing, a leader in neutral-atom quantum computing, partnered with a global financial institution to apply a hybrid quantum-classical approach to consumer bankruptcy modeling. The bank needed to build a challenger bankruptcy score in SAS® Viya® using rich internal transaction, lending and deposit data.
Classical models handle preprocessing, baselines and validation, while quantum techniques support feature generation and higher-dimensional projections. The hybrid approach helps teams identify where quantum creates a measurable advantage.
This matters because bankruptcy is difficult to predict. It’s a rare event, the target is highly imbalanced and third-party bankruptcy scores often function like black boxes with limited transparency. The question is simple: Where does quantum create measurable improvement in signal detection, optimization and model performance?
Quantum AI is already creating value in collateral optimization, portfolio optimization and liquidity modeling – especially where thousands of variables create decision spaces too large for classical systems to explore efficiently. For example, HSBC worked with IBM Quantum to explore algorithmic bond trading, testing how quantum approaches improve portfolio rebalancing and execution across fast-moving fixed-income markets.
Quantum AI examples in insurance: Edge cases that break things
Where the financial services industry is about scale, the insurance industry is about precision. It isn’t defined by the average case; it’s defined by the outliers.
When a major storm hits, insurers face a surge of claims across geographies, types and severity levels. Adjusters must be assigned based on expertise, location, cost and shifting deadlines – turning response into a high-pressure coordination problem.
That was the challenge facing one of the largest property and casualty insurers operating across the US, Canada, the UK and Ireland. Using historical claims data, SAS modeled the problem in SAS Viya to determine the best path forward. In this case, classical optimization methods solved the insurer’s objectives in seconds – without the need for quantum computing.
That distinction matters.
SAS approaches quantum AI pragmatically, not ideologically. Data scientists understand that the best solution isn’t always quantum – sometimes classical systems deliver the fastest, most effective results.
Planning for disaster
Catastrophe planning illustrates that balance well. Insurers must model the potential impact of major events, assess their exposure and make decisions about pricing, risk and reinsurance. Some scenarios can be handled effectively with classical optimization, while others may eventually benefit from quantum approaches that can evaluate far more possible outcomes simultaneously.
The key is knowing the difference – and applying the right technology at the right time.
Another example concerns disaster response planning for the public sector from D-Wave Quantum (the first quantum provider to offer both annealing and gate-model quantum computing technologies). While not strictly an insurance use case, the implications are direct: Faster evacuations and better response reduce loss severity, claims volume and recovery costs.
Quantum systems can evaluate a broad set of possible assignments in parallel, helping insurers allocate resources more effectively as conditions change.
In Japan, Sigma-i Co. Ltd. used D-Wave’s quantum annealing systems to optimize evacuation planning for tsunami and large-scale disaster scenarios. The goal was to solve complex routing and resource allocation problems under real-world constraints. SAS uses D-Wave's technology in its own research and work with customers.
Nature isn’t classical ... and if you want to make a simulation of nature, you’d better make it quantum mechanical, and by golly it’s a wonderful problem, because it doesn’t look so easy. Richard Feynman American Theoretical Physicist
Quantum AI examples in health care and life sciences: Where approximation is unacceptable
Nature operates on quantum mechanics, but for decades, we’ve modeled it with tools that weren’t designed for that level of complexity. That’s why simulation has long been considered one of quantum computing’s most important applications. Quantum opens the door to a world of health care and life sciences possibilities, from faster drug discovery and molecular modeling to more precise diagnostics, genomics and personalized treatment strategies.
So instead of testing one molecular configuration at a time, researchers can explore entire families of interactions – evaluating many possibilities at once and revealing how those systems behave.
IonQ and Kipu Quantum recently demonstrated this with protein folding. This was a natural early application for quantum computing because it functions as an energy minimization problem over a vast range of possible molecular conformations.
Using a 36-qubit trapped-ion system, the teams modeled peptide chains and identified low-energy states directly on quantum hardware. The significance wasn’t the size of the proteins – they were still small. The importance was that biologically meaningful optimization problems were executed directly on quantum hardware rather than relying solely on simulation.
A similar pattern is emerging in pharmaceutical R&D, where hybrid quantum workflows are being built around real drug discovery challenges rather than theoretical benchmarks. IBM and Moderna have explored mRNA structure simulation using variational quantum algorithms for therapeutic development.
Because structural features directly affect mRNA stability and function, quantum methods help evaluate reaction energetics, molecular stability and binding affinity more efficiently – reducing cost and time earlier in drug discovery.
Instead of testing one molecular configuration at a time, researchers can explore entire families of interactions – evaluating many possibilities at once and revealing how those systems behave.
Proof of concept: Saving more lives through kidney exchange
Kidney exchange offers another compelling example of how quantum optimization could reshape health care. The challenge is deceptively simple: matching incompatible donor-patient pairs with others to enable lifesaving transplants.
In practice, it becomes a massive optimization problem with thousands of possible combinations. The challenge is not just finding a feasible match quickly – it’s identifying the most optimal matching pathways across the entire network.
Consider a 2024 SAS proof of concept inspired by the kidney exchange market design program developed by the 2012 Nobel Prize in Economic Sciences winner. The team tested how quantum and classical methods could work together to solve this problem more efficiently.
- Classical solvers produced feasible solutions quickly but improved only incrementally over time.
- Quantum methods generated multiple candidate solutions in roughly 13 seconds.
- When those quantum-generated solutions were used to warm-start classical optimization, SAS achieved full optimization in about 30 seconds.
The implications reach beyond transplant matching. The same approach applies to a growing set of health care optimization challenges, from medical device and pharmaceutical supply chains to assembly plants, warehouse logistics and insurance claims analysis.
SAS is exploring how quantum machine learning and network graph approaches could support hybrid molecular modeling – opening new possibilities for simulating biological systems with greater speed and precision.
Quantum AI examples in manufacturing: Choreography you didn’t see coming
Manufacturing doesn’t lack data – it’s limited by how many variables can be evaluated at once. “If the combinatorial relationships in your data are very large, that’s where quantum starts to matter,” says Wisotsky. The goal isn’t just efficiency; it’s the ability to adapt quickly without disrupting the entire system.
Production scheduling sounds simple until reality shows up: hundreds of product variants, limited equipment, contamination rules, ingredient constraints, labor availability, transportation delays and demand shifts colliding at once. That was the challenge facing a multinational CPG corporation.
The CPG needed to optimize how more than 100 product variants could be assigned across just five production tanks while ensuring incompatible materials never mixed (imagine the many variations of shampoo, fabric softener, dishwashing detergent and conditioner). The number of possible combinations exceeded 10 to the 114th power – more than the number of atoms in the universe.
How was the problem solved?
- Traditional optimization reached the right answer in six hours.
- Quantum alone solved it in two minutes, but with suboptimal results.
- Combining both delivered speed and quality. Quantum methods rapidly generated near-optimal solution sets. Then classical optimization in SAS Viya refined those results. Instead of waiting six hours, they got results in 12 minutes.
That speed gives manufacturers and CPGs the flexibility to respond to sudden retail demand shifts, supply chain disruptions and real-time manufacturing decisions.
Another example comes from D-Wave Quantum, Groovenauts and Mitsubishi Estate, which began seeing measurable quantum results as early as 2019. Their project focused on optimizing waste collection across Tokyo’s Marunouchi district. While not a traditional manufacturing story, the underlying challenges were similar: limited resources, strict operational constraints, route and scheduling dependencies, and thousands of operational decisions shaping cost, efficiency and resilience.
Using AI to predict waste generation and quantum annealing to optimize collection routes, they reduced route distance from roughly 2,300 kilometers (1,429 miles) to 1,000 kilometers (621 miles). This cut carbon dioxide emissions by 57%, reduced vehicles by 59% and lowered total work time by 38%. It’s another example of how hybrid quantum systems can solve operational problems that look straightforward on paper but are enormously complex at scale.
Quantum won’t replace classical systems. It will work alongside them – each handling the parts of a problem they’re best suited to solve. Bill Wisotsky Principal Quantum Architect SAS
Test your business problem with quantum AI
As a launchpad for the quantum AI journey, SAS is introducing SAS Quantum Lab (coming soon to SAS Viya customers). It’s designed to complement quantum experts’ existing work and empower other users who are ready to explore, test and validate their ideas.
The lab gives organizations a practical way to explore quantum AI – comparing classical, quantum and hybrid approaches side by side and testing ideas before committing to quantum AI hardware. Early testing has shown more than 100x speed improvements and up to 99% cost savings for certain experimentation workflows, helping teams validate ideas faster while avoiding false starts.
Learning where to use quantum today will position organizations to move first as the technology matures.
