Why quantum AI isn’t an IT priority yet – Spiceworks
Every few weeks, another announcement about quantum AI hits the news. Vendors pitch it as the next frontier, and corporate boards are asking whether their organizations are falling behind.
But there’s a massive gap between marketing claims and enterprise reality. According to a new Gartner report, no enterprise AI workload at scale will run on quantum hardware through 2028, and classical accelerated chips will continue to dominate every production benchmark.
For IT leaders already juggling spending on GenAI, agentic AI, cybersecurity, cloud, and data modernization, the question is whether quantum AI needs to be a priority now.
READ MORE: Who really owns AI inside the enterprise?
Quantum computing will eventually matter to the enterprise. But for most organizations, quantum AI is not yet a production priority. Focusing on established AI infrastructure rather than untested quantum hardware is crucial for IT leaders as they bridge the divide between hype and reality.
Gartner isn’t saying quantum computing is useless. It’s simply separating what’s possible today from what’s still experimental.
Classical AI, which operates on CPUs, GPUs, and TPUs, has already generated tangible value for businesses. Techniques inspired by quantum mechanics use ideas from quantum mechanics but run on conventional hardware and can therefore deliver value in areas such as optimization and simulation.
Hybrid quantum-classical systems are still largely limited to research and small pilots. As for true quantum AI, there is still no peer-reviewed evidence that quantum hardware can deliver a clear performance or cost advantage for production AI workloads. Gartner doesn’t expect that to change at enterprise scale this decade.
“When vendors claim to deliver ‘quantum AI,’ they usually refer to hybrid or quantum-inspired techniques, not quantum-native AI running at enterprise scale,” said Chirag Dekate, VP Analyst at Gartner. “True quantum computing is not ready for any production AI workload and will most likely not be for the rest of this decade. No peer-reviewed result demonstrates quantum advantage on production AI workload.”
The real risk isn’t quantum itself, but spending on it at the expense of AI projects that can deliver value today. According to Gartner, generative AI and agentic systems can already produce measurable results within 12 to 18 months. Quantum AI, by contrast, has yet to show measurable value on a production workload and is unlikely to do so anytime soon.
READ MORE: What it takes to run AI locally on client devices
That’s why Gartner warns against putting both under the same budget. Doing so can make it harder to track what’s delivering results, and allow long-term quantum experiments to compete for funding with AI capabilities the organization needs now.
Shadab Hussain, Lead Engineer for Data and GenAI at MathCo, sees the same confusion from the practitioner side.
“Quantum AI is mostly a labeling problem,” he says. “What gets sold under the name is usually a quantum-inspired or hybrid technique, not machine learning that requires quantum hardware. No peer-reviewed result yet shows a quantum advantage on a production AI workload, and nothing on the current hardware trajectory suggests one will arrive before the end of this decade.”
Advancements in quantum computing are genuine, with rising investment and improved hardware. Many companies are exploring the technology. However, Hussain notes that most progress is in quantum hardware and simulation, not in AI production workloads. Equating the two could misallocate AI budgets into projects that may not yield returns for years.
Quantum computing shows promise in specialized optimization, simulation, materials research, and drug discovery, while researchers explore hybrid approaches through carefully scoped pilots. However, these initiatives differ significantly from the production AI systems that IT teams are currently deploying, including generative AI tools, copilots, enterprise assistants, and agentic systems.
QuEra’s 2026 Quantum Readiness Report, based on responses from 291 industry experts, suggests the market is becoming more focused on practical outcomes. Companies remain interested in quantum computing but are increasingly looking for reliable results, verifiable progress, and clear economic value rather than promises alone.
“In 2026, the cards will be reshuffled,” says Yuval Boger, Chief Commercial Officer at QuEra Computing, as per the report. “Companies continue to believe strongly in the potential of quantum computing. But they want to know where and under what conditions it actually adds value. The market now measures progress by results, not promises.”
Post-quantum security, however, is a different matter. A Keyfactor study of 450 cybersecurity professionals across North America and Europe found that 48% of organizations weren’t prepared for quantum-related cybersecurity threats.
That makes post-quantum readiness a security issue IT teams need to start planning for now. But it belongs on the security roadmap, driven by its own risk and compliance timeline, rather than being treated as part of a quantum AI investment.
The practical response isn’t to choose between AI and quantum but to keep the two separate. Production AI spending should remain focused on classical infrastructure, data quality, governance, and systems that can already deliver measurable value.
Quantum hardware experimentation, meanwhile, belongs in R&D. Treat it as a long-term bet, not something IT teams need to build into their production capacity today. Post-quantum cryptography is different again. That belongs on the security roadmap, driven by risk, compliance requirements, and the time needed to update existing cryptographic systems.
Where quantum-inspired techniques can run on existing GPU infrastructure and show a clear benefit, they can sit within the current AI stack without requiring a quantum hardware investment. Any quantum pilot should also have clear success metrics, a classical benchmark, and defined exit conditions so experimentation does not quietly consume budget without producing results.
Hussain also points to a disconnect between how quantum AI is being marketed and where the work is actually happening. “The convergence being sold is quantum making AI better,” he says. “The convergence actually being funded is AI making quantum work. Enterprises that get the direction backward will spend this decade’s AI budget on next decade’s compute.”
Quantum is certainly worth watching and, where there is a genuine use case, experimenting with. But for now, it should not distract from the AI infrastructure, governance, and systems that enterprises need to make work today.
Sign up for the Spiceworks Newsletter and stay ahead with a curated mix of expert advice, technical tips, and the trending discussions your peers are having right now.
A Deeper Dive: DHCP and DORA
SpiceQuest September (2026) — Of Waffles and Wild Cards
Network shares for laptops
On June 22, Toolbox will become Spiceworks News & Insights
© Copyright 2006 – 2026 Spiceworks Inc.
source
This is a newsfeed from leading technology publications. No additional editorial review has been performed before posting.
