top of page

An AI breakthrough: A nuclear fusion experiment was completed successfully

By: April Carson

The potential of nuclear fusion to provide a limitless, renewable supply of energy has long been recognized. However, we can only realize this amazing vision if we can successfully understand the complex science behind the reactor. This has been a challenge that many have tried to overcome, but with little success.

Until now.

For decades, scientists have been making tiny steps in the direction of this aim, but there are still a number of problems to be overcome. Controlling the highly heated and volatile plasma in the reactor is one of the major challenges that persists; however, a new technique shows us how to do it.

In a collaboration between the Swiss Plasma Center (SPC) and artificial intelligence (AI) research firm DeepMind, scientists utilized a deep reinforcement learning (RL) algorithm to investigate the intricacies of plasma behavior and control in a fusion tokamak – a donut-shaped device that uses magnetic coils positioned around the reactor to control and manage the plasma within.

The scientists' goal is to create a fully self-sustaining and energy-independent power plant that can operate on its own indefinitely. It's not simple, since the coils need hundreds of subtle voltage adjustments per second in order to keep the plasma confined within magnetic fields.

To maintain nuclear fusion reactions – which require keeping the plasma at hundreds of millions of degrees Celsius, hotter than even the Sun's core – complex, multi-layered systems are required to control the coils.

In a recent research, researchers, on the other hand, demonstrate that one AI system can handle the task alone.

The AI system was successfully able to maintain the fusion reaction without any human intervention. The research is a major step forward for the development of autonomous AI-controlled reactors.

DeepRL is a type of deep learning architecture that uses simulated environments to train controllers. We developed controllers that can both keep the plasma steady and be used to accurately shape it into various forms by combining DeepRL with a simulated environment.

The researchers trained their AI system in a tokamak simulator, where the machine learning technology found by trial and error how to navigate the intricacies of plasma magnetic confinement.

The AI advanced to the next stage after completing its training period – putting what it had learnt in the simulator to use in the real world.

The RL system carved plasma into a variety of new forms within the reactor, including one that had never before been seen in the TCV: stabilizing "droplets" where two plasmas coexisted simultaneously inside the device.

The AI could produce more complex shapes, such as 'negative triangularity' and 'snowflake' arrangements, in addition to the traditional forms.

Each of these forms has the potential to generate energy in the future if we can sustain nuclear fusion reactions. The 'ITER-like shape' (as seen above) is one of the system's configurations, which may prove beneficial in future studies by the International Thermonuclear Experimental Reactor (ITER), currently under construction in France.

According to the researchers, these plasma formations' magnetic control "represents one of the most difficult real-world problems for reinforcement learning has been applied to," and it may open up a new path in how real-world tokamaks are constructed.

In reality, some experts believe that what we're seeing here will have a huge impact on future plasma control technologies in fusion reactors.

"I believe that AI is the only way forward," physicist Gianluca Sarri of Queen's University Belfast, who was not involved with the research, told New Scientist.

"There are so many variables to consider, and a minor variation in one of them can lead to a substantial shift in the end result. It's a very time-consuming procedure if you attempt it manually."

The successful AI-assisted experiment is a clear indication that the technology has a lot of potential for future use in fusion reactors. With more research, AI could play an even bigger role in helping us achieve nuclear fusion – and bring us one step closer to clean, sustainable energy.

Nuclear fusion is one of the most promising sources of clean, sustainable energy available. It's also incredibly difficult to achieve – which is why researchers are turning to AI for help.

Redman: A Heroes Journey - Interviewed by Billy Carson


About the Blogger:

April Carson is the daughter of Billy Carson. She received her bachelor's degree in Social Sciences from Jacksonville University, where she was also on the Women's Basketball team. She now has a successful clothing company that specializes in organic baby clothes and other items. Take a look at their most popular fall fashions on

To read more of April's blogs, check out her website! She publishes new blogs on a daily basis, including the most helpful mommy advice and baby care tips! Follow on IG @bossbabymav



Are you a member of the 4BK TV Channel? If not, you should want to become one!!

On, you can Expand your mind and explore your consciousness in our collection of workshops by Billy Carson, including Remote viewing - Ancient History - Anomaly Hunting, and how to Manifest the things in life you've always desired

Start your 3-day FREE trial now!




bottom of page