Cortical Labs, an Australian biotechnology firm, has begun selling what it calls the world’s first commercially available biocomputer. The product, called the CL1, is powered by live human neurons grown from reprogrammed adult cells. Each unit integrates 800,000 neurons with digital systems to process information in real time. The company positions the CL1 as a research tool for neuroscience, disease modeling, pharmacology, and unconventional computing tasks.
The CL1 is available either as a standalone system or in scalable racks of thirty units for lab deployments. The price is set at $35,000 per standalone device. Volume purchases in rack configurations lower the per-unit price to $20,000. Cortical Labs also offers remote access via a wetware-as-a-service model, charging $300 per week for remote operation of hosted units.
This is not a theoretical announcement. CL1 units are operational and shipping. Each unit combines a life-support module with an interface controller and communication software. The life-support component supplies nutrients, controls temperature, filters waste, and maintains osmotic balance. The software module allows researchers to transmit electrical stimuli to the neurons and receive feedback through multi-electrode arrays.
The neuron source for each CL1 is a culture of human brain cells, created through induced pluripotent stem cell reprogramming. Donor cells, typically skin or blood, are reverted to a stem-cell-like state and then differentiated into cortical neurons. These neurons are not connected to any body. They do not come from fetuses or embryos. They are grown in vitro from adult volunteers. Cortical Labs maintains that all cells are ethically sourced, and that buyers must have appropriate ethical clearance to generate their own cultures.
The company’s earlier prototype, DishBrain, was developed to prove that human neurons could be used as computing agents. In 2022, the DishBrain system successfully learned to play a simplified version of Pong. The system used feedback loops, training the neurons to control a paddle that responded to a moving stimulus. When the neurons acted correctly, the system reinforced those patterns. When they failed, it adjusted stimulation. According to published data, the neurons learned more quickly and adapted more flexibly than reinforcement learning agents trained on the same task.
The CL1 system builds on that experiment. The hardware supports a fully automated training environment. The system accepts Python commands and returns real-time electrophysiological data. Users can create virtual scenarios in which the neurons interact with digital objects, solve tasks, or respond to changing stimuli. Cortical Labs describes this not as artificial intelligence but as synthetic biological cognition.
The company claims that wetware systems like the CL1 have advantages over both conventional neural networks and neuromorphic chips. Biological systems exhibit synaptic plasticity, emergent generalization, and self-repair. These traits are not well replicated in silicon. Current artificial neural networks require massive datasets and tuning to perform tasks that human brains accomplish with sparse exposure. Cortical Labs sees this asymmetry as a market opportunity.
Use cases for the CL1 fall into three categories. First, drug discovery. By exposing neurons to chemical compounds and measuring their electrical responses, researchers can build models of drug effects on neural tissue without human trials. Second, disease modeling. Neurons from individuals with Alzheimer’s or epilepsy can be cultured and tested in controlled environments. Third, cognitive computation. CL1 systems can be trained to perform information processing tasks that require flexible adaptation, pattern recognition, or multi-modal sensory integration.
Cortical Labs is not claiming consciousness. The neurons in CL1 do not have sensory organs, language, memory, or self-models. They are not sentient. What they do possess is computational responsiveness grounded in a substrate that is biologically evolved for information processing. Cortical Labs refers to this as wetware, a term used to describe computing systems based on living tissue rather than silicon.
The commercial model includes both hardware and cloud access. Wetware-as-a-service allows institutions to run experiments on preconfigured CL1 units hosted by Cortical Labs. The company offers a web interface and API access. Users can push stimulus patterns, configure closed-loop experiments, and retrieve data without maintaining live cell cultures themselves. This lowers the barrier to entry for institutions that lack biosafety labs or technical personnel trained in electrophysiology.
To address the challenges of cell maintenance, each CL1 unit includes an automated support system. Fluid exchange, waste removal, oxygenation, and temperature regulation are all controlled digitally. The system is designed to run continuously for up to six months per cell batch. After that, the neurons degrade and must be replaced. The company offers replenishment services and custom cell sourcing on request.
Cortical Labs has raised over $11 million in funding. Investors include Horizons Ventures, Blackbird Ventures, and In-Q-Tel, the venture arm of the United States Central Intelligence Agency. These stakeholders suggest the technology has perceived strategic value beyond academic research. Military, intelligence, and biotech entities may see wetware computing as a next-generation platform for adaptive processing, sensor fusion, or remote analysis.
Security concerns are already under discussion. If biological systems are placed online, how can they be protected from malicious instruction sets or adversarial stimuli? Unlike conventional chips, neurons exhibit long-term plasticity. A single corrupt training session could alter behavior in unpredictable ways. Cortical Labs claims to have isolation layers and rollback systems in place. However, these are software protections applied to living substrates. The long-term consequences of synthetic cognition operated over networks remain unknown.
Ethical questions also remain unresolved. Some critics argue that repeated training of neuron clusters raises the possibility of unintended awareness or suffering, even in the absence of traditional indicators. Others worry that once wetware becomes normalized, it may lead to more invasive experiments or biohybrid integrations in consumer products. Cortical Labs counters that the neurons have no sensory input and no conceptual encoding capabilities. They are signal processors, not subjective agents.
The launch of the CL1 represents the first scalable wetware platform made available to commercial entities. Until now, bio-neural computing has been confined to academic labs with custom setups. With the CL1, developers and researchers can prototype wetware applications without needing to build a biosystem from scratch.
Cortical Labs is not alone in this space. Competitors include FinalSpark, which is building wetware AI agents, and BrainCorp, which is exploring hybrid chips that blend biological and electrical substrates. However, Cortical Labs is the first to offer live neuron platforms at scale, with full automation and developer access.
Biological computing remains in its infancy. The CL1 cannot outperform GPUs on numeric benchmarks or replace cloud AI for language modeling. Its value lies in adaptive intelligence, non-algorithmic reasoning, and modeling human-like learning patterns. For certain classes of problems, especially those involving physical systems, human biology, or complex adaptation, silicon remains inefficient. Wetware may provide a path forward.
Cortical Labs states that the future of synthetic cognition is not to replicate the brain but to use biology as a functional substrate. The CL1 is the first embodiment of this vision. It is not conscious. It is not alive in any organismal sense. But it thinks, reacts, adapts, and learns. In a world saturated with digital logic, the return of biology as computing medium may signal the start of a post-silicon era.
Works Cited
“Human Brain Cells on a Chip for Sale.” IEEE Spectrum, June 3, 2025.
Additional material from Cortical Labs press release and product specification sheets, 2025.