How much more efficient would a computer be if instead of simulating the functioning of a brain directly used real brain cells? The question, in my opinion, already has a concrete (and obvious) answer: it would be much more efficient. I was convinced of this by looking at the biocomputer CL1 just launched by the Australian company Cortical Labs. A commercial device which literally contains living human neurons forming networks of connections on a silicon substrate. It is no longer necessary to simulate the functioning of brain cells: the biocomputer uses them directly, opening up scenarios that challenge our understanding of artificial intelligence and, by extension, human intelligence itself.
When the Biocomputer Becomes Commercial Reality
On March 2, 2025, a completely new era in the world of artificial intelligence was officially inaugurated in Barcelona. No, I'm not exaggerating; for once, the hyperbole is perfectly justified. The world's first commercial biocomputer was presented. Not a laboratory prototype, not an experiment, but a concrete product that will be distributed in the second half of the year. And that's no small feat, on the contrary.
What distinguishes this system from the AIs we know? Everything, actually. The CL1 does not use traditional chips, but a biological substrate made of real human brain cells grown in the laboratory. These cells, arranged on an array of electrodes, form biological neural networks that they continually evolve, adapting and learning in a completely different way than silicon-based AI models.
Today is the culmination of a vision that has guided Cortical Labs for nearly six years. Our long-term mission has been to democratize this technology, making it accessible to researchers without the need for specialized hardware and software. The CL1 biocomputer is the realization of this mission.
Synthetic biological intelligence is born
The concept behind the biocomputer has a name: Synthetic Biological Intelligence (SBI). A fascinating oxymoron that unites biology and technology in a way that, until recently, belonged exclusively to the fiction of Philip K. Dick o W. But what exactly does it mean?
SBI is a form of intelligence that uses the biological substrate of intelligence itself (neurons) in a new, engineered way. As explained by the scientific director Brett Kagan, “we think of it almost as a different life form than, say, animals or humans. We think of it as a mechanical, engineering approach to intelligence.”
And here lies the revolution: it is not about simulating the functioning of a brain through algorithms, but about using the same “components” of the brain to create something new. The result is a system that learns incredibly quickly and flexibly, far outperforming silicon-based AI chips used to train language models like Chat GPT.
A Body in a Box: Anatomy of the Biocomputer
The heart of the CL1, as mentioned, is made of human neurons grown in the laboratory, positioned on an array of planar electrodes (or as he explained kagan, “basically just metal and glass”). Inside, 59 electrodes form the basis of a more stable network, giving the user a high degree of control in activating the neural network.
But of course the cells need to stay alive to function. That’s why this SBI “brain” is embedded in a rectangular life support unit, which is then connected to a software-based system for real-time operation.
A simple way to describe it would be like a body in a box, but it has wave filtration, it has where the culture medium is stored, it has pumps to keep everything circulating, gas mixing, and of course temperature control.
I can’t help but think, with a certain uneasiness mixed with fascination, how much this “body in a box” recalls the “brain in a vat” of philosophical thought experiments. Only here it is not a thought experiment, but tangible reality. The biocomputer does not even require an external computer to function, which makes it even more surprising as an autonomous technology.
Wetware-as-a-service: The Organic Cloud
One of the most revolutionary features of the CL1 is the distribution model. Cortical Labs intends to offer what they call “Wetware-as-a-Service” (WaaS), a play on words with the more familiar SaaS (Software as a Service). Customers will be able to purchase the CL1 biocomputer unit outright, starting at about $35.000, or simply purchase time on the chips, accessing them remotely via the cloud.
This approach democratizes access to a technology that might otherwise remain confined to elite laboratories or large companies. As he pointed out kagan,
we are aiming to be significantly cheaper, and we want to drive prices down even further in the long run. In the meantime, we provide access to people from anywhere, anyone, any home, through the cloud-based system.
It’s a concept that fascinates me: a future where a researcher in Madagascar or an innovator in a small Italian town can access living biological neural networks simply through an internet connection. The democratization of science taken to a whole new level.
The History of the Biocomputer: From Pong to Delirium
Cortical Labs’ journey to commercial biocomputing didn’t start today. In 2022, the team had already made international news after developing a self-adapting computer “brain” by placing 800.000 human and mouse neurons on a chip and training this network to play the video game Pong.
That system, called DishBrain, was just the first step of a much more ambitious path. The team demonstrated that human pluripotent stem cells (hiPSC) integrated into high-density multielectrode arrays (HD-MEA) could be electrophysiologically stimulated to forge autonomous and efficient pathways of information exchange.
But the most interesting challenge was how to rewarding brain cells when they exhibited desired behaviors, and punishing them when they failed at a task. Research has shown that Predictability was the key: Neurons seek connections that produce energy-efficient and predictable outcomes, and will adapt their networks to seek that reward, avoiding behaviors that produce a random, chaotic electrical signal.
The incredible potential: from drug discovery to medical research
The applications of biocomputing are potentially endless, but Cortical Labs is initially focusing on scientific and medical research. The complex and ever-evolving SBI neural networks (under a microscope you can see how they form “branches” from electrode to electrode) have, for starters, the potential to revolutionize the way in which the drug discovery , disease modeling.
As Kagan explained, this technology could represent a major step forward in research into neurological diseases such as epilepsy and Alzheimer's, as well as other brain-related diseases.
“The vast majority of drugs for neurological and psychiatric diseases that go into clinical trials fail, because there are so many more nuances when it comes to the brain. But we can actually see those nuances when we test outcomes with these tools.”
What I find particularly interesting is the idea that the biocomputer may reduce animal testing.
“Our hope is that we can replace significant areas of animal testing with this. Animal testing is unfortunately still necessary, but I think there are many cases where it can be replaced and that is an ethically good thing.”
The “Minimal Brain”: Beyond the Biocomputer
While the launch of the CL1 biocomputer represents a huge step forward for Cortical, the team is already working on the next phase of the SBI: the “Minimal Viable Brain”. The concept is fascinating: how to bioengineer a human-like “brain” with the minimum amount of unnecessary cellular differentiation, but which would have the complexity that a neural network composed of homogeneous cell types does not possess.
This type of tool it would be a very powerful model, which would allow for more precise control and analysis than is currently possible in real-brain research. As Kagan explained: “They would basically be the key biological components that allow something to process information dynamically and responsively, according to basic principles.”
A fundamental question the team asks is: What is the minimally functioning brain? The smallest one we know of has 301 or 302 neurons (depending on who you ask) and is found in the nematode C. elegans. But each of those neurons is highly specialized. So the question is, “Is the C. elegans brain the minimum viable brain? Do you need all those neurons, or could you get it with, say, 30 neurons all uniquely wired?” We’ll see in the next few years.
Biocomputer Ethics: Questions Without Easy Answers
If you read Futuro Prossimo you know that these questions always come up sooner or later. Ethics, my friends. It matters. And the ethics of this technology have been a focus for Cortical from the very beginning. Already the first release in 2022 has sparked a lot of debate, especially in the area of human “consciousness” and “sentience.” However, protections have been put in place, as much as possible, for the ethical use of CL1 drives and WaaS remote access.
“There are a number of regulatory approvals required, depending on the location and specific use cases,” the team noted. “Regulatory bodies may include health agencies, bioethics committees, and government organizations that oversee biotechnology or medical devices. Compliance with these regulations is essential to ensure responsible and ethical use of biological computing technologies.”
But as a global pioneer in this ambitious technology, Cortical knows that (much like the rapid advancement of non-biological AI) the broad applications of SBI are not easy to predict. Biocomputing raises profound questions about what it means to be intelligent, conscious, or even alive, issues that become increasingly nuanced as technology advances.
The future is organic and digital together
With the launch of the CL1 physical system and the Cortical Cloud for remote WaaS use, Kagan and his team are excited to see where SBI can go once it gets into people’s hands.
“The CL1 is the first commercialized biological computer, uniquely designed to optimize communication and information processing with in vitro neural cultures,” the team noted. “The CL1, with integrated life support to maintain cell health, holds significant possibilities in the fields of medical science and technology.”
What strikes me most, as I reflect on the implications of this technology, It's how SBI is inherently more natural than traditional AI. It uses the same biological material (neurons) that underpins intelligence in living organisms. By harnessing neurons as a computational substrate, SBI has the potential to create systems that exhibit more organic and natural forms of intelligence than traditional silicon-based AI.
Beyond AI: A New Form of Life?
I can’t help but wonder: are we witnessing the birth of a new form of life? Not fully human, not fully machine, but something new, something that challenges our traditional categorizations. The CL1 biocomputer is not just a technological innovation; it is an ontological challenge, an invitation to rethink the boundaries between biology and technology, between natural and artificial. It is a system that evolves, learns and adapts using real biological components, but assembled in a way that is not found in nature.
In a certain sense, it is as if we have created a new branch on the tree of life: a branch that grows not through natural evolution, but through human ingenuity. And as this new branch blossoms, it invites us to reconsider what it means to be alive, to be intelligent, to be human.
The CL1 biocomputer is not just the future of technology; it is a portal to a future where the lines between organic and inorganic, born and built, become increasingly blurred. A future that, as Cortical Labs has demonstrated, is no longer confined to the realm of absurdity, but is already here, pulsating with cellular life on a silicon substrate.
