DeepMind scientists state they prepared an AI to handle a nuclear fusion reactor
Key takeaways:
- The London-based AI lab, which Alphabet owns, declared Wednesday that it had readied an AI system to manage and sculpt a burning plasma inside a nuclear fusion reactor.
- Nuclear fusion, a method that powers the universe's stars, implicates breaking and fusing hydrogen, a standard seawater element.
- DeepMind asserted that the breakthrough, issued in the journal Nature, could open new routes that progress nuclear fusion research.
An AI to control the Nuclear Fusion reactor:
Scientists at DeepMind have seen another real-world application for their artificial intelligence software.
The London-based AI lab, which Alphabet owns, notified Wednesday that it had instructed an AI system to handle and sculpt a searing plasma inside a nuclear fusion reactor.
Nuclear fusion, a procedure that powers the stars of the universe, entangles cracking and combining hydrogen, a typical seawater element.
The process, which emits extensive amounts of energy, has been lauded as a potentially limitless origin of clean energy, but some technical challenges still ought to be overcome.
Scientists use tokamaks on Earth — doughnut-shaped vessels wrapped by electromagnetic coils — to recreate the nuclear fusion reactions in outer space.
The magnets in these tokamaks are utilized to "contain" the explosive hydrogen plasma, which is more burning than the sun's core. Maintaining the magnetic coils presently needs numerous layers of complicated control systems.
Martin Riedmiller, control team lead at DeepMind, co-author of a paper printed in the journal Nature on Wednesday, informed CNBC that it's "a complex" control issue.
With the Swiss Plasma Center at EPFL, a university in Lausanne, Switzerland, DeepMind said it had created a reinforcement learning AI system that could handle the magnets and modify their voltage thousands of times per second.