Artificial intelligence (AI) engineers at BitEnergy AI have developed a new algorithm called Linear-Complexity Multiplication (L-Mul), which replaces floating-point multiplication with integer addition. This innovation could significantly reduce the power consumption of AI systems, potentially by 95%. The L-Mul method maintains high accuracy and precision while being simpler than traditional floating-point multiplication.

Source: https://www.tomshardware.com/tech-industry/artificial-intelligence/ai-engineers-build-new-algorithm-for-ai-processing-replace-complex-floating-point-multiplication-with-integer-addition