Researchers have reported that hardware that can learn skills through using a certain type of artificial intelligence could dramatically cut the power requirements of AI. Sharing intelligence features between hardware and software would also offset the energy needed for deploying AI in advanced applications, including self-driving cars. AI can require an enormous amount of energy. Just to play a simple game, AI technology can require software running on thousands of computers, or the energy that three nuclear plants produce in one hour.
Experts state that software is helping to combat several AI challenges, notably its issues with energy consumption. Although AI hardware development research is just beginning, researchers have already demonstrated pieces of AI in potential hardware. AI’s large energy demand issue has not yet found a solution, experts express that AI hardware development has the potential to relieve the energy industry of AI’s burden.
Read More: Quantum Hardware Could Cut AI Power Needs
