doi.org/10.1126/sciadv.aec5049
Credibility: 999
#quantum computing
Researchers at University College London (UCL) have developed an innovative way of combining artificial intelligence with quantum computing, managing to predict the behavior of complex and chaotic systems with much more precision, stability and efficiency than traditional methods
This discovery, published in the journal Science Advances in April 2026, represents an important advance for areas such as climate prediction, medicine, energy and transportation.
Chaotic systems, such as the movement of liquids, gases and turbulence, are extremely difficult to predict because small changes can generate very different results over time.
Typically, scientists have two options: run full simulations, which can take weeks on supercomputers, or use traditional AI models, which are faster but lose accuracy over time.
The new hybrid approach intelligently solves this problem.
Instead of using the quantum computer all the time, researchers only employ it in one crucial step of AI training.
The quantum computer analyzes large volumes of data and identifies statistical patterns that remain stable even in chaotic systems.
These patterns, called invariant properties, are then used to train a conventional AI model on a supercomputer.
The results were impressive: the “quantum-informed” model was around 20% more accurate than the traditional version and maintained stability for much longer periods.
Additionally, it needs hundreds of times less memory, which makes it viable for large-scale simulations.
The secret lies in the unique properties of qubits, the basic units of quantum computing.
Unlike classical bits, which are just 0 or 1, qubits can be in multiple states at the same time (superposition) and influence each other instantly (entanglement).
These characteristics allow even a small number of qubits to represent a huge set of possibilities, better capturing the chaotic nature of physical systems.
Professor Peter Coveney, senior author of the study, explained that this technique can be applied in many practical areas: predicting the weather more accurately, modeling blood flow in arteries, simulating the interaction between molecules or optimizing the design of wind farms to generate more energy.
Maida Wang, first author, highlighted that this demonstrates a practical quantum advantage, that is, the quantum computer actually does something that the classical computer alone cannot do with the same efficiency.
The experiment was carried out with a 20-qubit quantum computer from the company IQM, connected to supercomputing resources at the Leibniz Supercomputing Center, in Germany.
To function, this equipment needs to be cooled to temperatures close to absolute zero, around -273°C.
Despite the current limitations of quantum computers (such as noise and errors), the team managed to overcome the problem by using quantum equipment only in a specific phase of the process.
This research paves the way for real-world applications in the near future and also inspires the development of new, even more powerful classical methods.
Next steps include testing the system with larger data and real-world situations, as well as creating a more solid theoretical foundation.
In short, the union between AI and quantum computing is making predictions of complex systems much more reliable and faster.
What once seemed like science fiction is now starting to become reality, promising significant impacts on how we understand and control the world around us.
Published in 04/24/2026 16h17
Text adapted by AI (Grok) and translated via Google API in the English version. Images from public image libraries or credits in the caption. Information about DOI, author and institution can be found in the body of the article.
Reference article:
Original study: