AI could propel forward the discovery of new fusion techniques.

Scientists have been grappling with the complexities surrounding the development of a thermonuclear fusion reactor for decades. A successfully working fusion reactor could provide nearly unlimited energy. Nuclear fusion occurs naturally in the sun, but the problem is reproducing the process in the dramatically different conditions of temperature and gravity here on earth. But machine learning may change all that. (TNW)

TAE Technologies

TAE Technologies has been working on fusion energy for the past two decades. It hopes to ultimately commercialize a clean, compact, and affordable technology that is safe in every way.

TAE, who has been working with Google (NASDAQ: GOOGL), says machine learning could innovatively solve long-standing maths problems and facilitate new fusion techniques. TAE’s CEO is confident enough to say that the company will commercialize nuclear fusion with this decade.

Nuclear fusion on the sun

Nuclear fusion occurs on the sun amidst very high gravity, pressure, and heat. The conditions cause hydrogen atoms to fuse, forming atoms of helium. The process releases vast amounts of energy. The problem, here on earth, in duplicating the reaction is that it requires vastly more amounts of energy as an input than what is received as the output.

AI and machine learning the big differentiator?

Machine learning and AI could be a gamechanger. The technology can analyze mountains of data from physics experiments much faster and efficiently compared to humans. These advances in data analysis can speed up the nuclear fusion experimentation process dramatically, enabling quicker iterations and more meaningful results.

Related Story:  Artificial Intelligence: AI to Help Analyze Large Images from Aerial Sources