Groundbreaking Achievement: Chinese Scientists Successfully Simulate Monkey Brain Using Artificial Intelligence

In the burgeoning realm of artificial intelligence (AI), there have been significant strides made in replicating the complexity of biological systems, particularly those found in living beings. This evolution goes beyond conventional chatbots and encompasses the development of AI models akin to the brains of mammals, such as primates.
Recent advancements have seen Chinese researchers unveil an AI simulation of a macaque monkey’s brain, dubbed the Darwin Monkey, which is currently housed on a computer equipped with 960 specialized chips. Each chip supports over 2 billion spiking neurons and over 100 billion synapses, approaching the number of neurons in a real macaque brain, as reported by Zhejiang University, translated via Google.
This project represents a significant leap towards more sophisticated brain-like intelligence. The Darwin Monkey is currently the largest neuromorphic (brain-like) computer globally and the first to be based on neuromorphic-specific chips, according to Interesting Engineering.
While the prospect of AI replicating human brains may seem imminent, it’s essential to note that macaques, despite their intellectual prowess, are not as cognitively advanced as humans. They lack critical brain circuitry for human speech, as reported by Princeton University. However, the development of robot monkeys could potentially bring us closer to this goal if researchers can successfully integrate these AI brains with physical bodies.
The Darwin Monkey outperforms Intel’s neuromorphic computer, Hala Point, announced in April 2024. Hala Point features 1.15 billion artificial neurons, equivalent to an owl’s brain capacity. Sandia National Laboratory is currently utilizing Hala Point for advanced brain-scale computing research across various fields, including commercial, defense, and basic science, according to Craig Vineyard, Hala Point team lead at Sandia.
Zhejiang University researchers first introduced the Darwin Mouse in 2020, featuring 120 million artificial neurons. In early 2023, they developed the “brain-inspired” Darwin 3 chips in collaboration with the Zhejiang Laboratory. Two years later, the Darwin Monkey was realized.
The Darwin Monkey employs China’s DeepSeek AI model for tasks such as logical reasoning, content generation, and mathematical problem-solving. Furthermore, its new operating system is designed for increased efficiency, achieving “concurrent scheduling of neuromorphic tasks and dynamic optimization of system resources,” taking into account communication bandwidth and task characteristics.
Efficiency appears to be a central focus in the development of these neuromorphic systems. Intel also claims that Hala Point could help mitigate the current unsustainable rates of computing costs.
“Brain-inspired computing systems can address the high energy consumption and computational complexity of existing deep networks and large models,” states Zhejiang University. “Additionally, [the Darwin Monkey’s] unsupervised online learning mechanism can bring groundbreaking advances to AI.”