Cortical Labs & Nvidia Launch Brain Cell Data Centers in Singapore, Melbourne

Human Brain Cells to Power New Data Centers in Singapore and Melbourne

Key Facts

Biotech startup Cortical Labs is building two small data centers powered by lab-grown human brain cells integrated with silicon hardware.
The facilities are located in Singapore and Melbourne, with a biological neural network server stack of 30 units expected to go online in the coming months.
Cortical Labs aims to make four stacks available for commercial cloud access by the end of 2025.
The technology combines living neurons with electrode arrays, creating a hybrid "wetware" system that could eventually challenge traditional silicon chips from companies like Nvidia.
The company recently launched the CL1, described as the world’s first commercial biological computer using living human brain cells.

Cortical Labs, an Australian biotech startup, is constructing two small data centers in Singapore and Melbourne that will be powered by living human brain cells grown on silicon, according to a Bloomberg report. The initiative represents a radical departure from conventional computing by using biological neural networks instead of purely silicon-based processors, potentially offering new efficiencies in AI and data processing.

The project builds on Cortical Labs’ recent unveiling of the CL1, a “body in a box” biological computer that integrates lab-grown human neurons with hardware. Chief Science Officer Brett Kagan has demonstrated the system’s viability, showing healthy brain cells responding to computer inputs via real-time monitoring of electrical activity.

Biological Computing Breakthrough

Cortical Labs is pioneering what it calls “synthetic biological intelligence.” The core technology involves culturing human brain cells onto silicon electrode arrays, allowing the living neurons to process information in ways that mimic aspects of natural brain function. According to multiple reports, the company has made significant progress since its earlier DishBrain experiments, achieving more stable hardware and better optimization of the biological “wetware.”

In late 2024, visitors to Cortical Labs’ Melbourne headquarters observed the CL1 system firsthand. The setup features relatively simple, stable hardware paired with advanced techniques for maintaining and optimizing the living neural networks. The system displays ECG-like spikes on monitors, confirming that the neurons are healthy and actively responding to computational inputs.

The company’s latest development involves constructing a first-of-its-kind biological neural network server stack. This stack will house 30 individual units, each containing living cells on their electrode arrays. These stacks are expected to come online in the coming months, with plans to deploy four stacks for commercial use through a cloud-based system by the end of 2025, as reported by TechRadar.

Collaboration and Competitive Context

While the Bloomberg article specifically mentions the potential to challenge chips from Nvidia Corp., Cortical Labs’ approach differs fundamentally from traditional semiconductor-based AI accelerators. Nvidia dominates the current AI hardware market with its GPUs, which power most large-scale data centers and training runs for large language models. In contrast, Cortical Labs’ biological systems leverage the natural efficiency of living neurons, which some researchers believe could offer advantages in energy consumption and certain types of parallel processing.

The data centers in Singapore and Melbourne mark an important step toward commercializing this hybrid technology. Singapore has positioned itself as a technology and data center hub in Asia, making it a strategic location for the project alongside the company’s home base in Melbourne.

How the Technology Works

The CL1 system and upcoming server stacks function by growing human brain cells on multi-electrode arrays. These arrays can both stimulate the neurons with electrical signals and read their responses. Through sophisticated training protocols, the biological networks learn to process information and perform computational tasks.

This approach sits at the intersection of biotechnology, neuroscience, and computer science. Unlike conventional computers that rely on binary transistors, these biological systems operate through the complex electrochemical signaling of living cells. Proponents suggest this could lead to more adaptive, energy-efficient computing for specific applications, particularly those involving pattern recognition or continuous learning.

However, significant technical challenges remain. Maintaining living cell cultures in a data center environment requires precise control of temperature, nutrients, and waste removal. The systems must also achieve reliability and scalability comparable to silicon-based alternatives to become commercially viable at scale.

Industry Implications

The development of brain-cell-powered data centers could have far-reaching implications for the AI industry. Traditional data centers consume enormous amounts of electricity, with AI training and inference contributing significantly to global energy demands. If biological computing proves more energy-efficient, it could offer a more sustainable path forward for certain workloads.

For developers and AI researchers, access to these systems via cloud services by the end of 2025 would provide an opportunity to experiment with fundamentally different computing paradigms. Early applications might include specialized AI tasks, robotics control systems, or hybrid approaches that combine biological and silicon computing.

The technology also raises important ethical and philosophical questions about the use of human brain cells in computing. While the cells are lab-grown and not conscious in any meaningful sense according to current scientific understanding, the concept of “living computers” is likely to spark debate.

What’s Next

Cortical Labs plans to bring its biological neural network server stacks online in the coming months, with commercial cloud availability targeted for the end of 2025. The two data centers in Singapore and Melbourne will serve as initial proving grounds for the technology at small scale.

The company continues to refine both the hardware and biological components of its systems. Future iterations may increase the density and complexity of the neural networks, potentially approaching more brain-like capabilities.

As the AI industry grapples with the physical and environmental limits of silicon-based computing, alternative approaches like biological computing are gaining attention. While still in early stages, Cortical Labs’ progress suggests that hybrid biological-silicon systems could become part of the computing landscape in the coming years.

The startup’s work is being closely watched by both the biotechnology and semiconductor sectors. Success in scaling these systems could open entirely new avenues for computing architecture, potentially complementing rather than immediately replacing existing technologies from established players like Nvidia.

Human Brain Cells Run New Data Centers in Singapore, Melbourne

Sources

Original Source

Related Topics

Comments