Beyond silicon: Harnessing the computational power of living cells to create the next generation of computers
In a quiet lab in Switzerland, scientists open the door of a special incubator, and a screen suddenly comes to life with dancing lines of neural activity. The source of this activity isn't a traditional computer, but clusters of human brain cells living in a dish—tiny, lab-grown "mini-brains" that are learning to perform computational tasks. Welcome to the emerging frontier of wetware computing, where the line between biology and technology is blurring in ways that once belonged squarely to science fiction 1 7 .
"We've spent decades trying to make silicon chips more like biological systems. Now we're learning to make biological systems more like silicon chips. The most powerful computing architecture will likely emerge from the combination of both" 3 .
This revolution has its roots in a profound insight captured by cell biologist Dennis Bray in his 2009 book Wetware: A Computer in Every Living Cell. Bray proposed that living cells aren't just simple building blocks of life, but sophisticated computers that continuously process information, make decisions, and execute complex programs. "How does a single-cell creature, such as an amoeba, lead such a sophisticated life?" Bray asked. "How does it hunt living prey, respond to lights, sounds, and smells, and display complex sequences of movements without the benefit of a nervous system?" 2
At its core, wetware represents a radical rethinking of computation itself. While traditional computers rely on hardware (physical components) and software (programmed instructions), wetware uses living biological elements—neurons, DNA, or entire cells—as computational components 8 .
Bray's fundamental insight was recognizing that every living cell behaves like a sophisticated computer. Through complex networks of molecular interactions, cells can process information and execute programs.
Biological computation happens through "molecular circuits" that perform logical operations similar to electronic devices, but with unique properties including adaptability, self-repair, and massive parallel processing 2 .
From their environment through chemical signals and physical stimuli
Using intricate biochemical pathways
In the form of genetic information and cellular memory
The driving force behind wetware research lies in the growing limitations of traditional computing and the extraordinary advantages of biological systems:
| Aspect | Traditional Computing | Wetware Computing |
|---|---|---|
| Energy Efficiency | High power consumption | Up to 1 million times more efficient 7 |
| Architecture | Fixed design | Self-organizing and adaptable 3 |
| Learning Ability | Requires explicit programming | Capable of spontaneous learning 4 |
| Heat Generation | Significant thermal output | Minimal heat production 3 |
| Repair Mechanisms | No self-repair capabilities | Natural self-repair functions 8 |
The energy advantage is particularly compelling in the age of AI. Training models like GPT-4 can cost tens of millions of dollars in electricity alone, while biological neurons operate at near thermodynamic limits of efficiency 4 7 .
Tiny electrodes that can both stimulate neural activity and record responses, creating a bidirectional interface between biological and electronic systems 4 .
Engineered DNA sequences that program cells to perform logical operations, essentially turning them into living computers 9 .
At the Swiss startup FinalSpark, researchers have created one of the world's first functional wetware computers using human brain organoids. Let's examine their pioneering work step by step 1 7 .
The process begins with human skin cells obtained from anonymous donors. These cells are reprogrammed into induced pluripotent stem cells—blank slate cells that can become any cell type in the body. Through careful culturing with specific nutrients and growth factors, these stem cells are guided to develop into functioning neurons that cluster together into spherical brain organoids, each about the size of a fruit fly larva's brain and containing approximately 10,000 neurons 7 .
The organoids are transferred to special platforms containing microelectrode arrays that serve as interfaces between biological and electronic systems. These electrodes allow scientists to send electrical signals to the neurons and detect their responses, creating a two-way communication channel 4 7 .
Unlike conventional computers that need only electricity, the wetware computer requires a complete life support system. The organoids are maintained in a nutrient-rich solution at precisely controlled temperatures, with waste products continuously removed. This artificial environment can keep the organoids alive and functioning for up to four months, though researchers are working to extend this lifespan 1 7 .
"Programming" a wetware computer looks fundamentally different from traditional coding. Instead of writing explicit instructions, researchers use electrical stimulation patterns to encourage the neurons to organize themselves and respond in specific ways. Some experiments have even used dopamine rewards—similar to how human brains learn—to reinforce desired neural pathways .
The computational capabilities of these mini-brains, while primitive compared to modern computers, demonstrate the potential of biological processing:
| Task | System Used | Performance | Significance |
|---|---|---|---|
| Braille Recognition | FinalSpark organoids | Successfully distinguished between different Braille letters 7 | Demonstrates pattern recognition capabilities |
| Playing Pong | Cortical Labs' DishBrain | Learned to play the simple video game 5 | Shows capacity for sensory-motor integration |
| Simple Arithmetic | Leech neuron computer (1999) | Capable of basic addition 8 | Early proof-of-concept for biological computation |
Perhaps most intriguing are the unexpected behaviors observed in these systems. FinalSpark researchers note that the organoids sometimes detect environmental changes—like the opening of their incubator door—despite having no obvious sensory mechanisms, causing unexplained spikes in neural activity. "We still don't understand how they detect the opening of the door," admits Dr. Fred Jordan of FinalSpark 7 .
The analysis of these systems reveals two fundamental advantages: their remarkable energy efficiency and their innate learning capacity. Biological neurons can perform computations using one millionth of the energy required by artificial neurons, and they naturally form adaptive connections in response to stimuli—a property that could prove crucial for developing more flexible, generalizable artificial intelligence 7 .
The wetware revolution is being driven by sophisticated technologies and reagents that enable precise control of biological systems:
| Tool/Reagent | Function | Application in Wetware |
|---|---|---|
| Induced Pluripotent Stem Cells (iPSCs) | Reprogrammed adult cells that can become any cell type | Source material for creating human neurons without embryonic tissue 1 4 |
| Microelectrode Arrays (MEAs) | Grids of microscopic electrodes for neural interfacing | Stimulating and recording from neural networks; bidirectional communication 4 |
| Synthetic Transcription Factors | Engineered proteins that control gene expression | Creating genetic logic gates for cellular computation 9 |
| Extracellular Matrix Coatings | Surface treatments that support cell growth and connection | Creating optimal environments for neuronal network development 4 |
| Optogenetic Tools | Light-sensitive proteins that control neural activity | Precise, non-invasive manipulation of specific neural pathways 3 |
As wetware technology advances, its potential applications are expanding across multiple fields:
With the enormous energy demands of AI threatening climate goals, wetware offers a potential solution. Biological processors could one day replace or complement traditional data centers 7 .
Companies like Cortical Labs are working toward integrating biological processors with robots, creating systems that could leverage the sensory-motor capabilities of human neurons 5 .
However, these advances raise important ethical questions. Could these miniature brains ever develop some form of consciousness? Researchers largely dismiss this possibility given the simplicity of current systems—10,000 neurons versus 100 billion in a human brain—and the lack of structures necessary for experience. Still, the field maintains collaboration with ethicists to establish appropriate boundaries 7 8 .
"When you start to say, 'I'm going to use a neuron like a little machine', it's a different view of our own brain and it makes you question what we are" 1 .
Wetware computing represents more than just a technical innovation—it's a fundamental shift in our relationship with both biology and technology. By recognizing the innate computational power of living cells, we're beginning to harness capabilities refined by billions of years of evolution.
The field remains in its infancy, with current systems performing only basic tasks. But the trajectory is clear: we're moving toward a future where biological and silicon-based systems work in concert, each leveraging their unique strengths. As Lena Smirnova of Johns Hopkins University notes, "Biocomputing should complement—not replace—silicon AI" 1 .
For Dennis Bray, who first articulated the vision of cellular computation, the development of functional wetware computers represents the thrilling realization of his theoretical framework. And for the scientists now working with living processors, there's a palpable sense of participating in something once confined to imagination.
"I've always been a fan of science fiction," confesses Jordan. "When you have a movie of science fiction, or a book, I always felt a bit sad because my life was not like in the book. Now I feel like I'm in the book, writing the book" 1 .
As the wetware revolution continues to unfold, we're all becoming part of that story—watching as the boundaries between the living and the computational blur, and discovering what becomes possible when we learn to speak the natural language of cells.