Although quantum AI and its realisation is on the horizon, in many boardrooms the technology, broadly, is not understood. The desire to profit from being first on the scene, however, is driving significant spending.
A recent quantum AI survey from SAS found that three in five business leaders are now exploring or actively investing in the space.
Potential use cases are emerging in high-stakes industries requiring speed, scale and precision, including risk simulation in finance, precision diagnostics in healthcare, and real-time disaster response planning in governments.
Amy Stout, Head of Quantum Product Strategy, and Bill Wisostsky, Principal Quantum Systems Architect, at SAS, provide the scoop on the current quantum conversation; defining quantum AI and the quantum advantage, considering the timeline to a defining quantum moment, and explaining why people should care about this technology.
What is quantum AI?
Amy Stout: Quantum AI is the combination of artificial intelligence and quantum computing, a new type of computation.
Today’s laptops and super computers run on what we call classical computing, and function using binary bits, which can be zero or one. Quantum computers fundamentally work differently. They function using qubits, or quantum bits, which can be 0, 1 or a combination of both at the same time.
It sounds complicated, but basically, tapping into quantum AI can help solve specific types of problems with greater speed and/or accuracy. It’s expected to be most helpful in optimisation, machine learning and molecular modelling, which can impact different industries – like financial services, manufacturing, life sciences and many more.
What is the ‘quantum advantage’?
Bill Wisotsky: In the news, there are constant reports about quantum advantage. These stories typically involve speed, with research showing that a quantum computer could solve a problem in hours that would take conventional computers hundreds of thousands of years.
These problems are very specific, and designed to demonstrate how quantum computers operate. Though these are important steps in research, they have nothing to do with useful applications for real-world customers. All too often, the media views quantum advantage as one-dimensional, but the quantum advantage isn’t only about speed – it’s multidimensional.
For example, in quantum machine learning, the advantage could be the ability to encode data into higher-dimensional representations achieved by quantum physics that traditional machine learning can’t, and/or the ability to train models with less data. Quantum advantage could also mean a significant reduction in power usage that quantum computing requires.
This gets to the centre of my argument. When trying to solve applied problems with quantum computing, the quantum advantage needs to be judged multidimensionally, using applied criteria that benefit the business trying to leverage this technology. Yes, it could be about speed, but it could also include many other possibilities.
Are we reaching a quantum inflection point?
Stout: It’s a running joke in the quantum space – quantum is three to five years away, every year. Many experts are trying to be realistic about the state of the market. We don’t want people to come in thinking that quantum AI is going to solve all their problems right now.
There are multiple types of hardware and multiple vendors that are all developing quantum computers working to achieve the scale, speed and accuracy that will be needed for these computers to provide tangible benefit for real-world, production-sized problems. Quantum has not yet reached that widespread technological maturity.
However, there is already much interest and investment in quantum today, and rightfully so. We’re seeing industry leaders investing in quantum, fully aware that in 2025 they likely won‘t get advancements that impact their bottom lines. What they will get is that first-mover advantage, including in-house expertise and intellectual property, for when the technology is more mature.
I’m an optimist, and I look at hardware providers’ R&D roadmaps and what’s been achieved over the last three to five years and what’s on the horizon for the next three to five years. I think we stand a good chance of seeing these computers able to provide quantum advantage for problems we’d consider low-hanging fruit relatively soon. From there, I hope that we’ll continue seeing examples of the heights we believe quantum AI could achieve.
Why should people care about quantum?
Wisotsky: Simply, quantum computing could change the world. There are so many use cases, but the two areas I think will be most affected by quantum computing are AI and medicine. As quantum computers get more powerful, and our knowledge on how to use them evolves, AI will be able to take advantage of the physics that quantum computing uses for its computation.
I think medicine will greatly benefit in the areas of drug discovery and biologics, with researchers gaining the ability to represent and model complex molecular and biological processes in ways that are currently impossible. That could look like researchers discovering better drugs and bringing them to market faster, accelerating processes that would’ve taken a decade of development otherwise.
However, in the future, I don’t think average users will even know they’re using quantum computing to accomplish their goals. I see quantum computing being almost like another accelerator, like all the ‘PUs’ we currently have. Are average users aware that the application they are using is running on a CPU, GPU or NPU? No, they just use the application.
There’s plenty of other editorial on our sister site, Electronic Specifier! Or you can always join in the conversation by visiting our LinkedIn page.
