A team of physicists from Loughborough University has developed a device that can process data that is changing over time directly inside the hardware. This could be 2,000 times more energy efficient than existing methods. The chip uses a memory resistor, a memory chip that remembers past signals, which alters how it reacts to new signals. This creates a more intelligent AI system that learns from history and reduces energy output. The researchers believe this could revolutionise AI systems for data that isn't stable and often changes rapidly. The technology could be used in time-dependent applications like weather reports, stock market tracking or wave analysis.