The company Cortical Labs has just announced a significant addition to its family of products, the CL1 biocomputer. This revolutionary device uses lab-grown human brain cells combined with a silicon chip. This new, revolutionary technology consists of 800,000 neurons that have been reprogrammed from the skin or blood samples of adult donors. With its unique approach, the CL1 is poised to transform several industries. It will revolutionize neuropsychiatric drug development, supercharge artificial intelligence, and find mind-blowing applications in experimental art.
The first batch of 115 CL1 units will begin shipping this summer, priced at $35,000 each or $20,000 when ordered in 30-unit server racks. The company has raised more than $11 million in funding from international investors including Horizons Ventures, Blackbird Ventures and the CIA’s In-Q-Tel venture capital fund. With this support, they’re poised to make a groundbreaking change in the emerging field of biological computing.
Innovation in Bioengineering
The CL1 represents a significant step up from Cortical Labs’ previous DishBrain prototype. This early prototype demonstrated the potential for cell cultures to interact with the arcade game “Pong” in a virtual reality environment. Kagan, the CEO of Cortical Labs, says that this new biocomputer is equipped with all of its predecessor’s features. It incorporates a continuous life support system that keeps neurons alive for six months.
Kagan says proper safety and training are key in managing this cutting-edge technology. “We don’t want somebody without the skills, capability or safety,” he states. He further notes that this technology is not intended for casual experimentation, remarking, “It’s not something you should be doing in your garage.”
The CL1 features 59 inputs that significantly reduce latency from 5 milliseconds to sub-millisecond levels, making it suitable for high-performance applications. Legacy data centers require tens of kilowatts just to cool AI workloads. A full rack of CL1s only uses 850 – 1,000 watts.
Broad Applications and Strong Interest
Cortical Labs has received a huge amount of interest from academia, private start-ups and government organizations. These organizations are short-sightedly excited about using applications to drug discovery, neurocomputation, AI acceleration, and even bitcoin mining. It’s this flexibility of CL1’s neurons that enable them to spontaneously self-organize and fine-tune themselves in response to sensory feedback. Intuitively, a 2022 paper published in Neuron illustrated just how quickly these neurons were able to learn—within minutes—while playing video games.
Karl Friston, Professor of Neuroscience, Pharmacology & Clinical Epistemology, University College London, describes the CL1 as “the first commercially available biomimetic computer.” On one view, [the CL1] could be regarded as the ultimate in neuromorphic computing that uses real neurons,” he explains. Among other benefits, Friston points out that the technology affords a unique opportunity to experiment on what he terms a “little synthetic brain.”
He elaborates on the benefits of this technology: “It allows people to study the effects of stimulation, drugs and synthetic lesions on how neuronal circuits learn and respond in a closed-loop setup, when the neuronal network is in reciprocal exchange with some simulated world.” This synergistic combination of tools and model organism creates a powerful research environment that may allow transformative discoveries about how neurons behave.
Future Prospects and Ethical Considerations
Cortical Labs is exploring collaborations that merge biological computing with experimental art, showcasing the versatility of the CL1 beyond conventional scientific applications. Kagan goes on to detail the dazzling ambitions of this new synthetic intelligence. It’s supposed to generate neuron cultures that can do things outside the reach of human specialists. He asserts that “with a sufficiently advanced collection of cells, you could achieve something that might even surpass current biology.“
Kagan is upfront about the controversial nature of this technology. “You actually need a device like ours before you can test this model at all,” he explains. In training and testing models of neuronal computation, this capability will enable scientists to explore how to restore these functions within disease states.