Australian biotech company Cortical Labs recently made waves at the Mobile World Congress in Barcelona by unveiling the world's first "programmable biological computer," the CL1. This groundbreaking tech merges lab-grown human neurons with traditional silicon-based computer technology, opening up exciting new avenues for the future of artificial intelligence.
At the heart of CL1 lies a silicon chip embedded with human neurons. This system is designed to house lab-cultivated brain neurons on a flat electrode array, forming a stable network with 59 electrodes, allowing users to precisely control the activation of the neural network. The entire system is housed within a rectangular life support unit and connected to a software-based system for real-time operation. This "living" computing approach creates a fluid neural network that is more dynamic and energy-efficient compared to traditional AI systems.
To ensure the vitality of neurons, the CL1 is equipped with a comprehensive life support system that regulates essential environmental conditions, such as temperature and gas exchange.
Dr. Chong Hon-Weng, founder and CEO of Cortical Labs, describes this product as "wetware-as-a-service" (WaaS), allowing customers to either purchase the CL1 biocomputer directly or use it remotely via cloud technology.
Unlike traditional computers, the CL1 has the potential to learn and adapt to tasks. Previous research has demonstrated that neural systems can be trained to perform basic functions, like simple video games. Findings from Cortical Labs indicate that incorporating biological elements into computing systems enhances efficiency in tasks that traditional AI struggles with, such as pattern recognition and decision-making in unpredictable environments.
Before the development of CL1, Cortical Labs introduced a prototype called DishBrain, which consists of 800,000 lab-grown neurons trained to play ping-pong. Research shows that these neurons can perceive and adapt to their surrounding environment, demonstrating a primitive form of learning ability.
Researchers have harnessed electricity to send signals into petri dishes, enabling neurons to recognize the positions of virtual paddles and balls. They provide feedback to teach the neurons how to play these games through electrical stimulation. This study demonstrates that biological neural networks learn and adapt to complex environments faster than traditional AI.
Cortical Labs has announced that the first batch of CL1 computers will begin shipping in the coming months, with an expected price of around $35,000 (approximately HKD 273,000). This technology may offer advantages in learning efficiency and energy consumption for AI advancements, as the adaptability of neurons could enhance capabilities in robotics, automation, and complex data analysis.
However, the scalability of this technology remains in question. The complexity of producing and maintaining neural networks is evidently higher than that of traditional processors, and ensuring the long-term stability of these systems presents numerous challenges.
The use of human brain cells for technological innovation has raised numerous ethical questions. Although the neurons utilized in CL1 are lab-cultured and lack consciousness, as technology continues to advance, it will be essential to establish guidelines to address moral and regulatory issues. The prospect of integrating living cells with computing hardware also sparks deep discussions regarding the limits of artificial intelligence and human-like cognition.
Scientists hope to use this biological computer to gain deeper insights into how neurons process information, applying this knowledge in fields such as drug development and research on neurodegenerative diseases.
