Artificial intelligence (AI) is getting more powerful, but it also uses a lot of electricity. Researchers at Clemson University have made a new material, called pTPADTP, that could help cut down on AI’s energy use.
Dr. Stephen Foulger, a professor in Materials Science & Engineering at Clemson, said:
“The new material, pTPADTP, is a polymer that could be key to making AI more energy efficient and cost effective.”
So, what makes this material special? It works kind of like a computer chip, but instead of only switching between 0 and 1 like normal computers, it can flip randomly back and forth. These are called probabilistic bits, or p-bits. That randomness can actually make some types of computing easier and faster, especially when AI has to deal with uncertainty, like predicting outcomes or learning patterns.
Another big plus is that these p-bits can run at room temperature. That’s different from quantum computers, which need very cold conditions to work. It also means the material could be cheaper and easier to use in the real world.
This new material might someday help create smarter, greener AI systems. The Clemson team is still testing and improving it, but the long-term goal is to make computers that are powerful without wasting so much energy.
To read more on Dr. Foulger’s research please visit https://news.clemson.edu/new-material-created-at-clemson-university-could-curb-ais-energy-appetite/
