Artificial intelligence has long been used to make predictions in fields such as healthcare, weather forecasting, and finance, helping professionals anticipate future outcomes. At the Leibniz Supercomputing Centre in Germany, researchers combined an AI model with a quantum computer to tackle complex prediction tasks beyond the capabilities of AI alone.
In a study published Friday in the journal Science Advances, a research team from University College London suggests that quantum computers could eventually enable AI systems to generate faster and more accurate predictions across a wide range of industries—tasks that might otherwise take conventional computers weeks to complete.
“Our findings show that, for this type of application, even today’s relatively small and imperfect quantum computers can improve the predictive performance of traditional AI models,” said Peter Coveney, a professor at UCL and co-author of the study, in an interview with CNET.
How Quantum Computers Achieve Greater Computational Power
Quantum computers differ from classical computers by performing multiple calculations simultaneously and using quantum bits, or qubits. Unlike classical bits, which store either 0 or 1, qubits can exist in both states at once through superposition and can also be linked by entanglement, enabling more powerful computation.
Quantum computers use superposition and entanglement to solve some complex problems faster than classical computers. Although current systems are fragile, require extreme cooling, and are not yet practical, they may help AI handle challenges beyond conventional computing.
The AI model used in the research runs on a supercomputer linked to a quantum computer at the research facility.
Using this combination, the researchers predicted the long-term behavior and interactions of gases and liquids in complex systems, a type of simulation widely applied in climate research, healthcare, and urban planning.
According to coauthor Maida Wang, a PhD student at UCL, the approach appears to show a practical form of “quantum advantage,” where a quantum computer achieves results beyond what classical computing alone can deliver.
However, quantum computers remain extremely sensitive, as small environmental disruptions can affect their calculations, limiting their use mainly to research settings.
AI and Quantum Computing Share the Workload
Due to current limitations, the supercomputer performed most of the work, while the quantum computer handled a specific computational step. The AI processed the data, sent a difficult calculation to the quantum system, and then continued the remaining tasks after receiving the results.
Coveney noted that even today’s imperfect quantum devices can improve traditional machine-learning methods when combined with data from powerful supercomputers.
Connecting AI models to quantum computers may seem futuristic, but companies are already exploring this combination in healthcare.
In 2025, Google reported that its Quantum Echoes algorithm could analyze molecular structures to support future drug development. Meanwhile, the University of Toronto and Insilico Medicine combined AI and quantum computing to design molecules aimed at targeting a previously “undruggable” cancer-related target.
Although challenges remain, including improving prediction accuracy and managing massive datasets, Coveney believes quantum computers can enhance complex simulations and forecasting. He added that the team is already developing practical applications for the technology.
Read the original article on: cnet
Read more:Japanese researchers develop AI that detects rare diseases from hand photos