Last year, scientists from Monash University achieved a groundbreaking milestone by creating the “DishBrain” – a remarkable computer chip incorporating around 800,000 human and mouse brain cells. This semi-biological chip, integrated with electrodes, demonstrated a form of sentience and astoundingly learned to play the classic game Pong within just five minutes.

The DishBrain’s micro-electrode array had the unique capability to read brain cell activity and stimulate them with electrical signals. To test its learning abilities, the researchers designed a version of Pong where the brain cells received electrical stimuli representing the ball’s position on the screen and its distance from the paddle. The brain cells were then allowed to act upon the paddle’s movement. The team implemented a basic reward system, where the cells received predictable stimuli when the paddle hit the ball and unpredictable stimuli when it missed.

This groundbreaking experiment marked the first time lab-grown brain cells were given the ability to sense and act on their environment, resulting in impressive outcomes. The research, conducted in collaboration with Melbourne startup Cortical Labs, has now received a significant grant of US$407,000 from Australia’s National Intelligence and Security Discovery Research Grants program.

The combination of biological computing with artificial intelligence in these programmable chips has the potential to surpass the performance of traditional silicon-based hardware. Lead researcher Associate Professor Adeel Razi explains that this cutting-edge technology could underpin a new era of machine learning, particularly in autonomous vehicles, drones, and robots. These “DishBrain-powered” machines could possess a novel type of intelligence that continuously learns and adapts throughout their lifespan.

The implications of this research are far-reaching, with applications in planning, robotics, advanced automation, brain-machine interfaces, and drug discovery. The DishBrain’s advanced learning capabilities could pave the way for machines that learn new skills without compromising old ones, adapt easily to changes, and seamlessly apply previous knowledge to new situations. Additionally, they could optimize their use of computing power, memory, and energy.

With the grant, Razi’s team aims to develop AI machines that replicate the learning capacity of biological neural networks, ultimately making them viable replacements for conventional silicon-based computing.

This remarkable breakthrough is opening up exciting possibilities for the future of artificial intelligence, bringing us one step closer to creating highly adaptable and intelligent machines that can revolutionize various industries.

By Impact Lab