Microsoft ends support for Internet Explorer on June 16, 2022.
We recommend using one of the browsers listed below.
Please contact your browser provider for download and installation instructions.
Imagine a city. Any city, anywhere in the world, just a few years from now. It has 99 problems. But do you know what isn't a problem? Energy. Getting power to everyone in the city is no longer an issue.
The city's homes, businesses, and transport systems are powered by clean energy that never runs out. There's no smog, no worries about the use of fossil fuels, and electricity bills are a fraction of what they used to be. Power plants hum quietly in the background.
What's this magical, fictional energy source of the future? You may have heard of it. It's nuclear fusion. But perhaps no longer quite so far in the future?
Imagine a world where energy scarcity no longer shapes global politics, where climate concerns have a real solution, and where growth isn't tied to pollution. Science fiction? Absolutely not. You may think it sounds far-off, but that vision of the future is coming closer to reality thanks to new advances in how we understand and control fusion reactions.
But first, to really appreciate what fusion could offer, we should perhaps contrast it with the nuclear technology we use today: fission. Fission works by splitting heavy atoms like uranium: releasing energy, yes, but also producing long-lived radioactive waste and posing serious risks of meltdown. Fusion, on the other hand, uses light elements like hydrogen, produces only short-lived waste, and has no risk of runaway reactions. It's far safer, cleaner, and virtually limitless. All qualities that would transform the global energy system.
Fusion energy is generated by the heating of hydrogen atoms to extreme temperatures until they merge into helium, releasing energy in the process. How extreme? How about 100 million degrees Celsius? At these temperatures, the fuel becomes a state of matter called "plasma," an extremely hot, electrically charged gas that is extremely difficult to contain. That's where tokamaks come in.
Tokamaks, the machines used in most fusion experiments today, were first developed in the Soviet Union in the 1950s. Their doughnut-like design, combined with powerful magnetic fields, was better at containing plasma than earlier systems and over the decades they have become one of the key elements of fusion research. Their main function is to keep superheated plasma confined long enough to make the fusion reaction sustainable. That's easier said than done, however, because plasma is unstable and liable to change shape or position in milliseconds. Which makes precise, real-time control essential. Until now, predicting the way plasma changes quickly and accurately has been one of fusion's biggest roadblocks.
That may be changing, thanks to the work of NTT and Japan's National Institutes for Quantum Science and Technology (QST). The two research partners have developed a new method using artificial intelligence to predict the magnetic fields that control plasma inside large-scale fusion reactors. NTT and QST have developed a "Mixture of Experts" model, which blends multiple AI models together and chooses which one to rely on, depending on the conditions inside the reactor. This lets it adjust to the constantly shifting state of the plasma, maintaining accurate predictions even when the system is behaving unpredictably.
They have already tested their research on JT-60SA, a massive tokamak in Japan developed jointly by Japan and the European Union. Using real data from this experimental device, the AI was able to predict the shape and position of the plasma produced with an error margin of one centimetre, just one percent of the plasma's total size. That level of precision, which NTT and QST achieved without relying on traditional physics simulations, is a first for any tokamak of this scale. It's not just a simulation or a theory; it's a proven result from a real, working machine.
This matters, because traditional methods for controlling plasma rely on solving complex physical equations in stages. They're accurate but slow, and they struggle to keep up with the fast-changing environment inside a tokamak. NTT and QST's AI-based method works in a single step, offering the potential for real-time adjustments. That includes not just the outer shape of the plasma, but also internal conditions like electric current and pressure distribution, both of which are critical to keeping the reaction stable.
Fusion energy still has a long road ahead before it can power entire cities. But it's coming closer. NTT and QST's research can be seen as a shift from managing plasma after the fact to anticipating its behavior as it happens. That could make it easier to design reactors that are smaller, more efficient, and cheaper to operate. It could also speed up progress on projects like ITER in France and future commercial reactors designed to supply electricity to the grid.
Fusion energy is one step closer to everyday use. By making it possible to control this limitless power source in real time, NTT and QST are helping build the foundation for a cleaner, more stable energy future. A future where the lights stay on, the air stays clean, and the planet gets a chance to breathe.
NTT--Innovating the Future
For further information, please see this link:
https://group.ntt/en/newsrelease/2025/03/17/250317a.html
If you have any questions on the content of this article, please contact:
Public Relations
NTT Information Network Research Institute
https://tools.group.ntt/en/news/contact/index.php
Daniel O'Connor joined the NTT Group in 1999 when he began work as the Public Relations Manager of NTT Europe. While in London, he liaised with the local press, created the company's intranet site, wrote technical copy for industry magazines and managed exhibition stands from initial design to finished displays.
Later seconded to the headquarters of NTT Communications in Tokyo, he contributed to the company's first-ever winning of global telecoms awards and the digitalisation of internal company information exchange.
Since 2015 Daniel has created content for the Group's Global Leadership Institute, the One NTT Network and is currently working with NTT R&D teams to grow public understanding of the cutting-edge research undertaken by the NTT Group.