IBM, has created a special kind of computer chip that works like a human brain and could make artificial intelligence (AI) more energy-efficient. This is important because regular AI systems use a lot of energy and produce harmful emissions.
This new chip is designed to use less energy by using parts that work similar to how our brains work. Our brains are very efficient at doing things while using very little energy. This new chip could make AI systems in things like phones, cars, and cameras work better without draining their batteries too quickly.
The special thing about this chip is that it uses components called “memristors” (short for memory resistors), which can store a range of numbers, kind of like a dimmer switch that can control different levels of light. These memristors work somewhat like the connections between neurons in our brains.
A professor from the University of Surrey explained that these memristors could work together to form a network similar to how our brain works. This could lead to computer systems that are more like brains.
While this discovery is promising, it’s not a simple solution, and there are challenges to overcome. Making these chips and using them in computers might be tricky and expensive. But if successful, these chips could make AI systems in phones and cars use less energy and have longer battery life. They might even help save energy in big computer centers that power AI systems and reduce the water needed to cool them down.
In simple terms, IBM has made a special computer chip that’s inspired by how our brains work, and it could make AI systems use less energy and work better in things like phones and cars. However, there are still challenges to solve before these chips become widely used.