Cortical Labs Brings Biological Computing to Melbourne and Singapore Data Centers

Key Takeaways

• Cortical Labs has unveiled a biological data center prototype in Melbourne and is developing a Singapore deployment with DayOne, marking one of the first attempts to operationalize biological computing in a data-center setting.
• The Melbourne site reportedly contains 120 CL1 biological computers, while the Singapore rollout is expected to begin with a 20-unit rack at NUS Medicine before any larger commercial expansion.
• The long-term significance is strategic rather than immediate: biological computing is still early-stage and lightly benchmarked at production scale, but it is emerging as a potential low-power alternative for selected computing workloads and research applications.

Cortical Labs Moves Biological Computing Into Real Infrastructure

Cortical Labs has taken a step in turning biological computing from a research concept into physical infrastructure. Operating two small-scale biological data-center initiatives: one already launched in Melbourne, and another under development in Singapore with DayOne Data Centers. Instead of standard processors, these facilities are designed around the company’s CL1 biological computers, which interface living neurons with silicon.

Information Age reported that the Melbourne prototype contains 120 CL1 units and now powers Cortical Cloud, giving researchers and developers remote access to the platform. The company has reportedly been building racks of internet-connected CL1 systems since September 2025, suggesting this was not a one-off showcase but the beginning of an operating compute environment.

What the CL1 Platform Means for Emerging Compute Architectures

Cortical Labs launched the CL1 in March 2025 as what it calls the first commercial biological computer designed to let users interact with real neurons through a programmable system, and its official documentation shows support for recording, stimulation, and real-time closed-loop control through a Python-based API. Cortical Labs also says the CL1 is designed to keep neurons alive for up to six months with an internal life-support system.

Media coverage has described each CL1 unit as containing around 200,000 neurons per device, though earlier reporting around related experiments referenced different cell counts in prior research configurations.

Why Singapore Matters for the Future of Sustainable Data Centers

DayOne and Cortical Labs said the initial deployment will involve one rack of 20 Cortical Cloud units at the Yong Loo Lin School of Medicine at the National University of Singapore, where the partners plan to validate performance, efficiency, governance, biosafety, and compliance before scaling further. The parties are exploring a phased expansion that could reach up to 1,000 units, subject to technical validation and regulatory approvals.

That timing also fits Singapore’s broader policy direction, as the country’s IMDA and EDB launched DC-CFA2 on December 1, 2025, making at least 200MW of data-center capacity available while emphasizing sustainability, green energy, and digital-infrastructure resilience. In that context, Cortical Labs’ biological-computing pitch aligns with a market increasingly focused on how to support compute growth without simply scaling conventional power and cooling demand.

Cortical Labs’ credibility in this space traces back to its earlier DishBrain work. In 2022, researchers involved with the project showed that brain cells in a dish could learn goal-directed behavior by playing a simplified version of Pong, using microelectrode arrays to stimulate and read cell activity. More recently, Information Age reported that neurons in the company’s CL1 systems were also trained to play Doom, reinforcing the company’s message that biological systems can be programmed and tested in increasingly accessible ways.

As Cortical Labs CEO Hon Weng Chong put it, “the next chapter of digital infrastructure must be built with sustainability at its core.” At the same time, the company has acknowledged that this technology is “not meant to replace traditional silicon.”

What This Means for the Electronics Industry

For now, biological computing is not a substitute for GPUs, CPUs, or mainstream data-center semiconductors but a trend that the industry’s search for new compute architectures is widening beyond advanced silicon, photonics, and quantum into bio-integrated systems. If the platform proves viable, the nearer-term opportunity may be in specialized research environments, low-power adaptive systems, neuro-inspired robotics, and experimental edge computing, rather than hyperscale AI training.