The technology that once lived only in physics labs is creeping into boardrooms. But a new wave of research suggests the biggest obstacle is no longer the hardware.

by Kasun Illankoon, Editor in Chief at Tech Revolt

[For more news, click here]

Somewhere between the hype and the skepticism, quantum AI is quietly having a moment. Not the kind of moment that ends up on the cover of Wired with a glowing qubit rendered in neon blue. A quieter, more complicated moment, where the technology is real enough to be useful, limited enough to frustrate, and far enough along that ignoring it is starting to feel like a strategic mistake.

For years, the conversation around quantum computing followed a familiar arc: breathtaking potential, punishing timelines, perpetual promises of a breakthrough just around the corner. Most industry analysts pegged "production-ready" quantum hardware somewhere in the early 2030s, and that estimate still holds.

But a growing body of work, and a growing number of companies, is pushing back on the idea that you have to wait for the hardware to mature before quantum starts paying dividends.

The concept at the center of that argument is quantum AI, an approach that applies machine learning algorithms to existing quantum hardware. The pitch is compelling: tasks that take hours can be done in minutes, problems previously considered computationally impossible can be solved on hardware available today, and models can be trained to learn efficiently on far less data than classical systems require. If that sounds like a lot of qualified optimism, it is. But the qualifications are getting fewer.

What Changed Between 2025 and 2026

Data and AI company SAS surveyed more than 500 global business and technology leaders across industries on the state of quantum AI adoption. The results, comparing findings from 2025 to 2026, reveal a significant shift in what is actually holding organizations back.

A year ago, the dominant concern was money. High cost of implementation ranked as the number one barrier to quantum AI adoption. That answer made intuitive sense: quantum computing infrastructure is expensive, expertise is scarce, and the return on investment was murky at best.

In 2026, cost has dropped to the second spot. The new top barrier is something more conceptual and, in some ways, more telling: uncertainty around practical, real-world uses. Rounding out the top concerns are lack of trained personnel, lack of knowledge or understanding, limited availability of quantum AI solutions, and a lack of clear regulatory guidelines.

Read together, these findings sketch a picture of an industry that has moved past sticker shock and into a more sophisticated form of hesitation. Leaders are no longer scared off by the price tag alone. They are worried about buying something they do not fully understand, for problems they cannot yet define.

That is both a solvable problem and a more interesting one.

The Spectrum That Changes Everything

To understand why quantum AI is different from quantum computing writ large, it helps to understand how experts in the field actually think about these technologies.

SAS frames classical and quantum computing not as competing systems, but as a spectrum. On one end sits classical computing, proven, reliable, ubiquitous. On the other sits quantum computing, experimental, exponentially more powerful in certain domains, and still maturing. Most real business and industry problems do not live cleanly at either extreme. They live in the middle, which is precisely where hybrid approaches, splitting computational workloads between quantum processors and classical processors, start to make sense.

"Organizations of all sizes are eager to develop intellectual property, their original, patented approach to quantum AI, so they will be ready as the technology comes of age," said Bill Wisotsky, Principal Quantum Architect at SAS. "Despite continued strong interest, leaders are understandably proceeding with caution, and they do not want to go all-in on expensive quantum investments they fear may not result in worthwhile use cases and solved problems."

That caution is rational. But it carries its own risk. Companies that wait for perfect clarity before exploring quantum AI may find themselves playing catch-up with competitors who were willing to learn while the field was still messy. Intellectual property, institutional knowledge, and early-mover advantages do not wait for consensus.

What Real-World Quantum AI Actually Looks Like

One of the most common misconceptions about quantum AI is that its applications are distant and abstract. The survey data suggests otherwise. When respondents were asked what business problems they hoped to solve using quantum AI, their answers were concrete, sector-specific, and decidedly near-term.

In financial services, organizations are eyeing quantum AI to enhance fraud detection, specifically to identify complex transaction patterns with greater accuracy than classical systems can manage. The appeal is obvious: fraud costs the global financial system hundreds of billions of dollars annually, and pattern recognition at quantum scale could dramatically reduce that exposure.

In telecommunications, the target is 5G network traffic optimization in real time, a problem that scales faster than classical computing can comfortably handle as networks grow more complex and user demand more unpredictable.

Drug discovery is another major area of interest, with organizations hoping to accelerate molecular simulation to surface new therapeutic candidates faster. The pharmaceutical industry spends an average of over a decade and billions of dollars bringing a single drug to market. Quantum-accelerated simulations could compress those timelines significantly.

Supply chain and logistics optimization, improvement of machine learning workflows for customer behavior prediction, and more efficient training of large language models for natural language processing round out the list of priorities. That last one carries particular resonance in an industry currently burning enormous computational resources on model training. The prospect of doing it faster and cheaper is not a marginal gain. It is a structural one.

The Product Designed to Close the Gap

SAS's response to the uncertainty problem takes the form of SAS Quantum Lab, a hands-on environment designed to let organizations explore quantum AI without requiring a team of quantum physicists. Coming in the fourth quarter of 2026 to SAS Viya customers, the product is positioned less as a tool for experts and more as an onramp for organizations that are quantum-curious but not yet quantum-fluent.

"This survey illuminates what SAS experts were already seeing in the market: that leaders are excited to use quantum, but the barriers to entry have been too high, and that requires a solution," said Amy Stout, Head of Quantum Product Strategy at SAS.

The platform is designed to let users compare classical, quantum, and hybrid results side by side for specific industry use cases. According to SAS, current testing shows performance improvements exceeding 100 times speedup alongside a 99 percent reduction in costs, though those figures will naturally vary depending on the problem being tested. A built-in virtual quantum AI tutor is designed to answer questions, suggest code, and guide users through their next steps, an acknowledgment that the real bottleneck in quantum adoption right now is human understanding, not hardware capability.

The Broader Competitive Stakes

The announcement was made at SAS Innovate, the company's global data and AI conference, as SAS marks 50 years of operation. That anniversary is worth pausing on. A company that has survived and grown through the mainframe era, the PC revolution, the internet age, and the rise of cloud computing is now making quantum AI a central pillar of its next chapter. That is not a bet made casually.

It also reflects something larger happening in enterprise technology. The companies best positioned to benefit from quantum AI are not necessarily the ones with the most quantum hardware. They are the ones that have spent the time to understand the problem space, build internal fluency, and develop use cases grounded in actual business needs. The hardware, when it fully matures, will accelerate that work. But the work has to start somewhere.

The SAS survey data suggests that the industry understands this, even if it has not fully acted on it. The shift from "too expensive" to "not sure how to use it" as the primary barrier is progress. It means the conversation has moved from resources to strategy. And strategy, unlike hardware, is something organizations can start developing right now.

The quantum economy is not coming soon. In important ways, it is already here. The question is whether organizations will engage with it on the terms it currently offers, imperfect, evolving, and genuinely promising, or wait for a clarity that may arrive too late to matter.