Artificial intelligence imitated the behavior of brain neurons
Using artificial intelligence, Chinese researchers created a model to imitate the behavior of neurons in the human brain, which does not depend on silicon chips.
Chinese scientists have simulated an artificial intelligence model that mimics the behavior of neurons in the human brain to achieve computing power without the high energy consumption of silicon-based processors, according to Eviralnews,, citing the South China Morning Post.
According to a scientific paper published in the journal Nature Computational Science, researchers have developed a model that bridges the gap between large and complex external artificial intelligence networks and the small and complex internal functions of the brain.
Industry experts say the team's findings could represent a fundamental shift in the development of artificial intelligence, prompting further exploration of computing solutions that do not depend on silicon chips.
Current trends in artificial intelligence mainly revolve around building larger neural networks, an approach that raises concerns about unsustainable energy requirements and lack of interpretability.
In contrast, the human brain – with 100 billion neurons and about 100 trillion synaptic connections – consumes about 20 watts of energy. At the same time, each neuron in the brain is more diverse and complex than any existing artificial intelligence model.
Neurons in the human brain send and receive complex messages that can evolve, but the silicon chip-based AI model is a simplified version of this process and can only produce simple binary signals.
The future of artificial intelligence development may depend on the combination of closely mimicking the dynamics of biological neurons with the development of larger models and more powerful hardware, driven by continued advances in neuroscience.
