What if the future of computing is not silicon but living brain cells? Inside laboratories today, scientists are building machines that think, grow, and learn like life itself.
Table of Contents
Living computers, also known as biocomputers, are not yet a part of everyday technology. Like quantum computers, they are still mostly experimental, existing in laboratories and research models. But the idea behind them is powerful: instead of using transistors to process information, biocomputers use living neurons.
Throughout history, humans have found many ways to process information, leading to the dominance of digital computing today. But every time a new method of computing emerges, it brings the potential to change what we can do. Biocomputing is one such unconventional path, offering possibilities that digital systems may never reach. In that way, it is similar to quantum computing, which promises to solve particular problems that digital computers cannot handle.
Biocomputers use much less energy. Neurons are a million times more energy-efficient than silicon for the same computation. Neurons do not grow on a chip—they are grown beforehand in the laboratory. Scaling neurons is easier than silicon because we can multiply them in the laboratory and then place them on hardware to build a biocomputer. And unlike electronic components, neuron-based systems can be composted at end-of-life, reducing pollution.
There is also a deeper link to artificial intelligence. Most AI today works by simulating neurons, which are called artificial neurons. Systems like ChatGPT and DALL·E run on these digital approximations. But real neurons are far more complex. If we could build AI using real neurons, we might unlock even greater capabilities, while also using far less energy. That is a rare engineering scenario: achieving better performance while simultaneously reducing power consumption.
Of course, biocomputing comes with its own challenges. But the potential gains—in both capability and sustainability—are strong reasons why researchers are still exploring this unusual but promising direction.
Building the infrastructure for biological computing
Neurons are placed on electrodes to allow two-way communication with electronic hardware. Most of the time, we send electrical signals to the neurons. Chemical signals are the next step, and sometimes we combine electrical and chemical signals to achieve this. One essential component is the use of molecular cages—structures that contain neurotransmitters such as glutamate, serotonin, or dopamine. These cages are placed in the medium surrounding neurons and can be opened with specific wavelengths of light, such as ultraviolet or blue. This enables precise control over chemical release, allowing researchers to mimic brain signalling with high timing accuracy.
A continuous microfluidic system is also required to keep the neurons alive by constantly refreshing the cell culture medium. In parallel, an electrical interface reads signals from neurons and delivers stimulation as needed. This setup involves electrodes and light-delivery systems that manage the electromagnetic radiation used to trigger molecular cages.
A robust software infrastructure supports these biological systems. A dedicated neural platform enables remote users to interact with living neurons by writing Python scripts. Custom interfaces are mainly developed in C++, with some tools in JavaScript. The entire system is open source, with documentation available on GitHub.
Equally important is how neurons handle time and energy. Unlike transistors, neurons remain mostly inactive, consuming little power. When triggered, a neuron sends a brief electrical spike lasting about 3 milliseconds—detectable as a voltage change. This sparse, event-driven behaviour contributes to the energy efficiency of neural systems.
This efficiency is the result of 300 million years of evolution. Neurons, which make up the nervous system, have been refined over time by natural selection to use power sparingly. Digital computers, in contrast, were built for speed and performance without the benefit of such long-term optimisation. Biological systems have evolved through random genetic variation to achieve efficiency levels that engineered systems are only beginning to approach.
Though neurons switch much more slowly, around 3 milliseconds compared to microseconds for transistors, this is not a limitation in practice. Digital systems can monitor neuron states and deliver stimulation at 30,000 cycles per second (30kHz), offering precise timing control for experiments.
Real-world applications are still about a decade away, but the foundation is in place. By combining biology, optics, electronics, and software, researchers are beginning to interface meaningfully with living computing systems.
| Market outlook and role in the future |
| Current demand: At present, global demand for biocomputers is close to zero. These systems are largely unknown to decision-makers, and practical applications are still emerging. Without real-world deployment, it is difficult to define the market accurately. Forecasted growth: A recent study by Centric Business Intelligence estimates the current market value of biocomputers at $7 billion, projected to grow to $17 billion by 2032, with a CAGR of 13.13%. While the accuracy of these figures is uncertain, it is notable that such market projections have only started appearing this year, indicating growing interest in computational biology. Silicon vs biocomputers: Biocomputers will not replace silicon-based systems for general tasks like cryptography or number crunching; such a replacement would be inefficient and impractical. However, for neuron simulations, biocomputers are expected to outperform digital systems in both energy efficiency and accuracy. Given that AI workloads are a major contributor to global electricity consumption, using biocomputers for neural simulations could offer significant energy savings. |
How human skin cells are turned into brain-like structures
Oops! This is an EFY++ article, which means it's our Premium Content. You need to be a Registered User of our website to read its complete content.
Good News: You can register to our website for FREE! CLICK HERE to register now.
Already a registered member? If YES, then simply login to you account below. (TIP: Use 'forgot password' feature and reset and save your new password in your browser, if you forgot the last one!)
