The Dawn of Biological Computing: CL1 and the Future of Human-Neuron Hybrid Machines
Silicon has been the foundational material of the computing revolution for over half a century. Every smartphone, data center, and embedded system today depends on microchips carved from wafers of silicon. But as the scale of computation continues to grow exponentially, researchers are beginning to encounter serious limitations in power efficiency, material availability, and scalability. One of the most radical alternatives to conventional silicon-based computing emerged in March 2025, when Australian biotech company Cortical Labs launched the CL1 — the world’s first commercially available biological computer powered by human neurons.
Priced at approximately $35,000, the CL1 marks a historic step toward hybrid computing systems that fuse biological intelligence with silicon infrastructure. Unlike traditional AI, which is software simulated atop digital processors, the CL1 uses actual lab-grown human neurons cultivated from stem cells. These neurons form active, learning neural networks interfacing with conventional computing architecture. The promise: real-time learning, ultra-low power consumption, and applications in everything from drug testing to robotics.
How Biological Computing Works
At the heart of the CL1 is a concept Cortical Labs calls Synthetic Biological Intelligence (SBI). Human neurons are grown and integrated onto a microelectrode array that functions as both an input and output system. Electrical signals are used to stimulate the neurons, which respond in kind. These responses are captured, interpreted, and used to drive computation and feedback. The system forms a closed loop of interaction that mimics how real brains process sensory input, adapt to new information, and learn from environmental feedback (Moses, 2025).
The CL1 also includes a built-in life support system to keep the neural culture viable. Temperature, gas exchange, nutrient flow, and waste filtration are managed by an array of tubes, sensors, and membranes. Every six months, filters need replacement due to protein buildup. The setup is visually striking — a rectangular chassis with a transparent top that reveals a pulsating mesh of cables and tubes nourishing living tissue (Chong, 2025).
biOS: The First Operating System for Neurons
To facilitate communication between digital and biological components, Cortical Labs developed a proprietary operating system called biOS. Unlike conventional operating systems, which manage hardware and software on binary logic, biOS enables direct input into the biological neural system. Researchers can simulate environments, send stimuli, and analyze responses in real time. The neurons interact with simulated objects as if they were part of a video game or an experimental environment. In earlier studies, neurons were taught to play Pong and demonstrated goal-seeking behavior — such as aligning paddle position to hit a ball — based solely on stimulus-response learning (Kagan et al., 2023).
Energy Efficiency and Learning Capabilities
Traditional data centers are becoming increasingly power-hungry. For example, a single NVIDIA A100 GPU can consume around 400W, and entire clusters may exceed 3.7 million watts annually (Henderson, 2024). By comparison, the CL1 consumes between 850 and 1,000 watts annually — orders of magnitude less energy. Since computing already accounts for roughly 7% of global energy usage, biological systems offer a potentially transformative path forward (IEA, 2024).
More impressive than energy metrics are the learning capabilities. Human neurons can form, reshape, and strengthen synaptic connections based on exposure to stimuli, providing a form of plasticity that far outpaces digital neural networks. In laboratory conditions, neuron cultures were able to demonstrate learning and task adaptation in fewer cycles than conventional machine learning systems, pointing to a form of real-time learning that could eventually bypass the need for massive data labeling and training (Nature Communications, 2023).
Implications for Biomedical Research
The immediate application for CL1 is in neuroscience and pharmacology. Researchers now have a platform to study living, learning neurons in a controlled computational environment. This has profound implications for neurodegenerative disease research, enabling scientists to test how neurons degrade under stress or respond to experimental drugs. It also offers a high-fidelity model for exploring conditions like epilepsy, dementia, and Parkinson’s — disorders with complex, cell-level behavioral dynamics that digital simulations often oversimplify (Chong, 2025).
Additionally, CL1 provides a viable alternative to animal testing. With human-derived neurons, researchers can run simulations and trials that are ethically superior and biologically more accurate. As legislation around animal research tightens in many countries, the CL1 offers a timely and scalable path forward (Reuters, 2024).
From DishBrain to CL1: A Timeline
Cortical Labs began development with a prototype known as DishBrain, which gained international attention in 2023. In that experiment, a neural culture composed of mouse and human neurons learned to play Pong using real-time feedback. The study emphasized a concept known as neural criticality — the idea that brains operate most efficiently when poised between chaos and order. The neurons exhibited higher performance when exposed to structured stimuli as opposed to random inputs, leading some to use the term “sentient,” which sparked heated academic debates (Kagan et al., 2023).
Building on this foundation, the CL1 integrates simplified electrodes, more robust life support, and a modular design suited for long-term experimentation. In June 2025, the first commercial units began shipping, followed by the launch of Cortical Cloud in July — a cloud-based interface allowing remote users to access and manipulate neural networks via subscription. Over 1,000 researchers are already signed up to test biological algorithms and conduct neuron-driven experiments through virtual interfaces (TechCrunch, 2025).
Ethics and Regulation
The integration of human neurons into computing raises difficult ethical and regulatory questions. Cortical Labs sources its neurons from ethically approved stem cell lines and collaborates with international bioethics boards. However, broader concerns remain: Could such systems attain a form of consciousness? How should responses that resemble preference or emotion be interpreted? Should these systems have rights?
Cortical Labs avoids speculative claims and maintains that its systems lack the complexity required for sentience or self-awareness. Yet as capabilities expand, so will scrutiny. Regulatory frameworks will need to evolve to address neuron sourcing, experiment limitations, intellectual property, and even the legal status of hybrid systems (Moses, 2025).
Future of Neural Computing
Industry projections suggest that biological AI computing could become a $60 billion market by 2030 (Statista, 2024). From robotics to personalized medicine, the potential applications are immense. Biological computers could enable real-time adaptation in autonomous machines, improve rehabilitation technologies, and transform how researchers model diseases and test treatments. Unlike silicon chips, neurons rewire themselves on the fly, allowing biological systems to keep pace with unpredictable environments in ways that software-based AI still struggles to replicate.
Ultimately, the CL1 represents a new class of machine — one that is not programmed in the traditional sense but trained, nudged, and observed. It is the first step in a movement that may eventually redefine what it means to compute. Rather than emulating cognition in code, we are now interacting directly with cognition in culture — biological culture, that is.
Key Takeaways
The CL1 introduces a revolutionary computing architecture that merges lab-grown human neurons with traditional silicon components. It consumes drastically less power than modern GPUs, offers adaptive learning without massive datasets, and provides a research platform for understanding the human brain. Its release opens the door to applications in medicine, ethics, robotics, and AI — all while challenging our assumptions about intelligence, sentience, and the future of machines.
References
- Chong, H. W. (2025). CL1: A New Frontier in Computing. Cortical Labs. Retrieved from https://www.corticallabs.com
- Henderson, M. (2024). Data Centers and Energy Use: The Coming Crisis. IEEE Spectrum. Retrieved from https://spectrum.ieee.org/energy-consumption-data-centers
- IEA. (2024). Digitalization and Energy. International Energy Agency. Retrieved from https://www.iea.org/reports/digitalisation-and-energy
- Kagan, B. J., et al. (2023). In vitro neurons learn and exhibit sentience-like behavior in a game environment. Nature Communications. Retrieved from https://www.nature.com/articles/s41467-023-40998-w
- TechCrunch. (2025). Cortical Labs Launches Cortical Cloud for Remote Neural Computing. Retrieved from https://techcrunch.com
.jpg)
.jpg)