Biotech start-up Cortical Labs has launched a biological data centre at the National University of Singapore, featuring computer units powered by 200,000 lab-grown human brain cells. According to the company, these facilities offer an energy-efficient alternative to traditional silicon-based infrastructure for managing unpredictable datasets in robotics and cybersecurity.
Singapore Biological Data Centre Opens at National University of Singapore
The facility went live on July 16 inside the National University of Singapore’s Centre for Life Sciences, developed through a partnership between Cortical Labs, NUS, and data centre operator DayOne, as reported by The Straits Times. Laboratory technicians maintain a dedicated life support system for the neurons, feeding them a mixture of sugar, micronutrients, and pH buffers every three days while a gas mixer pumps in carbon dioxide, oxygen, and nitrogen.
Singapore’s initial deployment comprises 20 biological computer units known as CL1s. Each unit houses at least 200,000 lab-grown neurons sitting on an electrode-fitted silicon chip, translating electrical signals into raw computing power. Plans are underway to expand the site to house up to 1,000 CL1 units following regulatory approval and safety testing.
Comparing Biological Computers to Traditional Silicon Chips
Cortical Labs founder and chief executive Chong Hon Weng explained that biological data centres excel in environments with limited and unpredictable training data, such as public-space humanoid robots or cybersecurity anomaly detection. Traditional silicon chips, however, remain superior for fast, precise, and repeatable calculations required by large language models like ChatGPT, according to Chong.
| Metric | Cortical Labs CL1 Unit | Standard AI Server (Nvidia H100) |
|---|---|---|
| Power Consumption | 30 watts (including life support) | Up to 10,200 watts (8-chip server) |
| Cooling Requirements | Minimal / None | Intensive cooling needed |
| Cost per Month | US$2,200 (S$2,800) per CL1 | US$4,300 per month on major cloud platforms |
The energy demands of traditional data centres prompted Singapore to impose a temporary pause on new facility construction in 2019. By contrast, each CL1 unit consumes about 30 watts, using less power than a handheld calculator, because biological neurons do not generate the intense heat associated with silicon servers.
Commercial Deployment and Global Scaling
Cortical Labs has already commercialised a similar biological data centre in Melbourne housing 120 CL1 units with roughly 20 paying customers, including universities and corporate research arms testing gaming and robotics.
Customers pay US$2,200 monthly to access a CL1 unit’s computing resources. NUS Yong Loo Lin School of Medicine neuroscience professor Rickie Patani, who oversees the facility, stated that identifying which neurons and support cells work best together will help build a scientific case for manufacturing these cell types at scale for data centre operations.
Did you know? The neurons used in Cortical Labs’ data centre units are derived from human blood cells that have been reprogrammed into stem cells, then grown onto micro-electrode arrays.
Frequently Asked Questions
How do biological data centres work?
Biological data centres use lab-grown human neurons seated on electrode-fitted silicon chips. The chips exchange electrical signals with a computer, translating neural activity into raw computing power.
How much power does a CL1 unit consume?
According to Cortical Labs chief executive Chong Hon Weng, each CL1 unit uses about 30 watts of power, which is less than a handheld calculator and significantly lower than high-end Nvidia AI chips.
Where is Singapore’s biological data centre located?
The facility is located inside the National University of Singapore’s Centre for Life Sciences, established in partnership with Cortical Labs and DayOne.
What are the primary use cases for biological computers?
Chief executive Chong Hon Weng stated that biological data centres are suited for unpredictable datasets, humanoid robot navigation, and cybersecurity anomaly detection where training data is limited.
Explore More: Stay updated on emerging tech trends by subscribing to our newsletter or exploring our archives for more breaking developments in artificial intelligence and biotechnology.
Related reading