Biocomputing and Living Computers

The Dawn of Wetware: Can Living Brains Power the Computers of Tomorrow?

A select cohort of pioneering scientists is turning what was once the stuff of imaginative stories into tangible reality. They are building computers using living human cells. This strange and fascinating new discipline is known as biocomputing. Researchers in this field aim to create powerful, efficient machines that could one day solve problems beyond the reach of even the most advanced supercomputers. Their work prompts profound questions about the nature of intelligence, consciousness, and what it means to be human. The journey into this uncharted territory begins in specialised laboratories, where the fundamental building blocks of thought are being repurposed to create a new form of information processing. This is not just a technological leap; it represents a fundamental shift in how we perceive computation itself.

The Vision of Biological Servers

At the forefront of this revolution is the Swiss laboratory FinalSpark. Co-founder Dr Fred Jordan imagines a future where vast data centres hum not with silicon chips but with biological servers. These biological systems would mimic the learning processes found in artificial intelligence but operate using a minuscule amount of power. Today’s data centres consume enormous amounts of electricity, contributing significantly to global energy demand. Biocomputers offer a sustainable alternative. The human brain, after all, is the most efficient computer known to exist. It performs complex calculations using the power equivalent of a dim lightbulb. Harnessing this biological efficiency is the ultimate goal for Dr Jordan and his colleagues, promising a greener future for the digital world.

Introducing the Concept of Wetware

While most people understand the concepts of computer hardware and software, researchers in this domain are creating something they call "wetware," a deliberately provocative term. The concept is straightforward yet profound. Scientists cultivate neurons, the brain's information messengers, into three-dimensional clumps referred to as organoids. These are, in essence, lab-grown mini-brains. Once formed, these organoids are connected to sophisticated arrays of electrodes. This interface allows for communication. Researchers can send electrical signals to the wetware and record its responses. This process marks the first step in harnessing the computational power of living neural networks. It is a delicate and complex procedure that bridges the gap between biology and electronics.

From Human Skin to Thinking Matter

The raw material for FinalSpark is stem cells, sourced from a specialised clinic located in Japan. These are originally derived from human skin tissue, with donors remaining completely anonymous. These are not just any cells; they are pluripotent stem cells, meaning they have the remarkable ability to develop into many different cell types. Within the laboratory, cellular biologists coax these cells to differentiate into neurons. Over several months, the neurons self-assemble into the tiny, spherical organoids. While these mini-brains are nowhere near the complexity of an actual human brain, they contain the same fundamental components. They form synapses, create neural networks, and exhibit spontaneous electrical activity, providing a basic model of brain function.

A Dialogue with Living Neurons

Communicating with an organoid is a delicate art. The electrode array serves as a translator between the digital world of a computer and the biological world of the neurons. A simple experiment illustrates this interaction. Pressing a key on a keyboard sends a precise electrical pulse to the organoid. If the connection is successful, a monitor displays a corresponding jump in activity, a graph resembling an electroencephalogram (EEG) reading. This response shows that the organoid has detected and reacted to the stimulus. Sometimes, however, the organoid's behaviour is unpredictable. Repeated stimuli might cause it to stop responding, only to produce a sudden, unique surge of energy moments later, a mystery researchers are still working to understand.

Image Credit - by Meritxell Huch, CC BY 4.0, via Wikimedia Commons

Teaching a Biological Computer

These elementary electrical prompts represent vital first steps toward a much larger goal: teaching the biocomputer. The team’s ultimate ambition is to trigger learning within the organoid’s neural network, enabling it to adapt and perform specific tasks. Dr Jordan compares this process to training conventional artificial intelligence. An AI receives an input, like an image of a feline, and learns to produce the correct output: identifying the animal. Similarly, researchers hope to train organoids to recognise patterns and deliver desired responses. This form of learning, rooted in biological plasticity, could prove to be far more flexible and efficient than the rigid algorithms that govern silicon-based AI systems.

The Australian Lab That Mastered Pong

The potential for learning in these systems is not merely theoretical. In 2022, a company from Australia named Cortical Labs provided a stunning proof of concept. Their team successfully instructed a cluster of human neurons, which they named DishBrain, to engage with the classic arcade game Pong. The neurons, living in a petri dish, received electrical signals indicating the ball's position. They learned to control the paddle by firing their own coordinated electrical pulses. Within minutes, the DishBrain system showed it could rally the ball, demonstrating rudimentary learning. This landmark experiment proved that a biological neural network, even outside a living brain, could learn to perform a goal-directed task in a simulated environment.

An Answer to the Energy Crisis

A primary driver behind the push for biocomputing is its incredible potential for energy efficiency. Modern supercomputers can perform astounding calculations, but they demand megawatts of power, enough to run a small town. The data centres that power the internet and AI are a colossal drain on the world’s electricity grids. In stark contrast, our own brains perform functions far more complex than any AI, all while running on about 20 watts. By replicating the brain's computational strategies, biocomputers could slash energy consumption by a factor of millions. This would not only reduce costs but also make advanced computing accessible without an unsustainable environmental footprint.

A New Frontier for Medicine

Beyond computing, wetware holds immense promise for medical research. At Johns Hopkins University in the United States, scientists are constructing mini-brains to investigate complex neurological conditions. A team led by Dr Lena Smirnova uses organoids to model the effects of disorders like Alzheimer's disease, Parkinson's, and autism. These biological models allow researchers to observe how diseases develop at a cellular level and to test the efficacy of new drugs in a controlled environment. This approach offers a powerful alternative to animal testing, which often fails to predict human responses accurately. It could accelerate the discovery of treatments for some of the most devastating brain disorders affecting humanity.

Organoid Intelligence: A New Discipline

The work being done at Johns Hopkins and other institutions has given rise to a new field known as organoid intelligence, or OI. The goal is to establish a new form of genuine biological computing that leverages the brain's inherent learning abilities. Scientists envision complex OI systems, essentially "brains-in-a-dish," that can learn continuously throughout their lives. These systems could revolutionise drug discovery, study the effects of toxins on the brain, and provide new insights into the very nature of human cognition and memory. OI represents a convergence of biology, neuroscience, and computer science, creating a powerful new tool for scientific discovery.

The Challenge of Sustaining Life

Despite its potential, biocomputing faces a monumental biological hurdle: keeping the wetware alive. A conventional computer simply needs electricity, but a living organoid has complex biological needs. Simon Schultz, a professor at Imperial College London, highlights the central problem. Organoids lack blood vessels. A person’s brain contains an intricate vascular network that delivers a constant supply of oxygen and nutrients to every cell while efficiently removing waste products. Lab-grown organoids have no such system. As they grow, cells in the centre can become starved of nutrients and die. Replicating this biological life-support system is currently the most significant hurdle for researchers.

Progress in Longevity and Stability

Researchers are actively working on innovative solutions to the life-support problem. Some teams are experimenting with microfluidic channels that act like artificial capillaries, perfusing the organoid with a nutrient-rich medium. Others are attempting to bioengineer organoids that can develop their own vascular-like structures. Progress is being made. FinalSpark has significantly improved the lifespan of its organoids during the last four-year period. The current systems are now viable for as long as four months, a substantial increase that allows for more complex and longer-term experiments. Each improvement brings the prospect of stable, long-lasting biocomputers one step closer to reality.

The Enigma of Organoid Death

When a computer "dies," it is a metaphor. For wetware, it is a literal event, and one that comes with its own eerie phenomena. Researchers at FinalSpark have observed that just before an organoid ceases to function, it sometimes emits a frantic flurry of electrical activity. This final, intense burst is strangely analogous to the heightened brain activity documented in certain individuals during the end-of-life process. Dr Jordan notes his lab has recorded thousands of these specific cessations during a recent five-year span. Each event is a setback, but it also provides valuable data, helping scientists understand the failure points and improve the resilience of future systems.

Scaling Up the Biological Brain

Moving from a single organoid in a petri dish to a powerful biocomputer capable of running a data centre is an immense engineering challenge. A single organoid contains tens of thousands of neurons, but a person’s brain has around 86 billion. To achieve useful computational power, scientists must learn how to connect and coordinate millions, perhaps billions, of individual organoids. This requires developing new types of hardware to house and interface with the biological components. It also demands sophisticated software to distribute tasks across the neural network and interpret the results. The scale of the problem is daunting, but it is a necessary step toward realising the technology's full potential.

The Ghost in the Machine

As scientists create increasingly complex brain organoids, they confront profound ethical questions. The central issue revolves around the potential for consciousness. Could a sufficiently complex organoid develop some form of sentience? Could it feel pain or experience a state of awareness? At present, these mini-brains are far too simple to be considered conscious. They lack the structure and sensory inputs found in a person's brain. However, as the technology advances, the line could begin to blur. This possibility has prompted the emergence of a new field called neuroethics, dedicated to guiding this research responsibly.

Defining Consciousness in a Dish

The debate over organoid consciousness forces a confrontation with one of science's deepest mysteries: the nature of consciousness itself. There is no universally agreed-upon marker for its presence. Scientists can measure electrical activity, but they cannot measure subjective experience. Researchers in neuroethics propose developing clear criteria and monitoring organoids for signs of complex, coordinated brain activity that might suggest a nascent form of awareness. They advocate for a cautious, step-by-step approach, ensuring that ethical considerations evolve in lockstep with the scientific capabilities. The goal is to reap the benefits of the research without accidentally creating a sentient entity that is unable to communicate its experience.

A Call for Regulation and Oversight

The rapid pace of development in biocomputing has led to calls for proactive governance. Leading scientists in this discipline argue for the establishment of clear international guidelines and regulatory frameworks. Such regulations would ensure that research is conducted ethically, with transparency and public engagement. This includes setting limits on the size and complexity of organoids until more is understood about their potential for consciousness. By addressing these ethical challenges head-on, the scientific community can build public trust and ensure that this powerful technology is developed for the benefit of all, preventing misuse and unforeseen negative consequences.

Silicon Will Not Be Replaced

Despite the revolutionary potential of wetware, experts believe it is unlikely to replace traditional silicon-based computers entirely. Professor Simon Schultz and Dr Lena Smirnova agree that biocomputing should be seen as a complementary technology, not a replacement. Silicon chips are incredibly fast and precise at performing straightforward mathematical calculations. Biological systems, on the other hand, excel at learning, adapting, and processing ambiguous information with remarkable energy efficiency. The future of computing will likely be a hybrid one, where the strengths of both systems are combined to tackle different types of problems.

The Niche for Wetware

The unique capabilities of biocomputers will likely see them find their niche in specific applications. They could be used to power a new generation of AI that learns more like a human, requiring far less data and energy. Their ability to interface directly with biological systems makes them ideal for advanced medical diagnostics and personalised medicine. They could also be used to create highly sensitive environmental sensors or to control sophisticated robotic systems that need to adapt to unpredictable environments. In these areas, the inherent strengths of wetware could allow it to outperform silicon technology significantly.

A Hybrid Future: Biology Meets Silicon

The most exciting prospect may lie in hybrid computers that merge silicon and wetware. In such a system, a traditional processor could handle high-speed, logical tasks, while a biocomputing co-processor could manage complex learning and pattern recognition. This synergy could unlock new possibilities in fields like materials science, climate modelling, and drug discovery. Imagine a computer that could not only simulate a new molecule but also intuitively "understand" its potential properties. This fusion of computational speed and biological learning could solve problems that are currently intractable, accelerating the pace of scientific and technological progress in unprecedented ways.

Living Inside the Story

While the technology advances toward practical uses, its pioneers remain captivated by its imaginative roots. For Dr Jordan, the work is the fulfilment of a lifelong fascination. He reflects that stories of the future often left him feeling that his own life was mundane by comparison. Now, he feels as though he is living inside one of those narratives, actively writing the next chapter. This sense of wonder and possibility drives the field forward, pushing the boundaries of what we believe is achievable. The researchers in this field are not just building machines; they are exploring a new relationship between humanity and technology, one where the line between the created and the creator begins to dissolve.