20 Watts: Brain's Power Secret

Brain-inspired computers, also known as neuromorphic computing, aim to close the gap between the human brain's efficiency and today's AI systems' energy consumption. The human brain operates on approximately 20 watts of power, a fraction of what is required to run current AI systems.

The human brain contains roughly 86 billion neurons, which are wired to each other through junctions called synapses. This complex network allows the brain to process information efficiently, with energy being spent only where it's needed. In contrast, ordinary computer chips have a separate processor and memory, resulting in a substantial share of energy use due to the transfer of data between them, known as the Von Neumann bottleneck.

Researchers are exploring ways to replicate the brain's capabilities using less energy, such as event-driven computing. This method involves performing computation only when events occur, reducing the constant energy expenditure of traditional computing methods. Event cameras, modeled on the human retina, are an example of this approach, using a fraction of the power of traditional cameras and handling fast motion without blur.

Neuromorphic sensors can also work equally well in bright sunlight and near-total darkness, demonstrating the potential of brain-inspired computing to improve efficiency and reduce energy consumption. By rethinking how computers work, neuromorphic computing aims to fundamentally close the gap between the brain's efficiency and today's AI systems' energy consumption.

## Why it matters The development of brain-inspired computers has significant implications for the future of artificial intelligence and computing. If successful, neuromorphic computing could enable the creation of more efficient and powerful AI systems, with potential applications in fields such as healthcare, finance, and transportation. The ability to run on low power consumption, similar to the human brain, could also enable the widespread adoption of AI in areas where energy efficiency is crucial, such as in mobile devices and edge computing.

## What happens next As researchers continue to explore and develop neuromorphic computing, we can expect to see significant advancements in the field. The potential for brain-inspired computers to revolutionize the way we approach AI and computing is substantial, and ongoing research aims to bring this vision to life. With the human brain as a model, scientists are pushing the boundaries of what is possible in computing, and the results could have far-reaching consequences for industries and societies around the world.