Table of Contents
Imagine a world where computers are powered not by silicon transistors, but by living neurons. This concept, worthy of science fiction, has become a tangible reality thanks to international collaborations. A revolutionary prototype was unveiled in Singapore, marking a turning point in the field of computing.
You may remember those futuristic movies where machines were controlled by human brains? Although this may seem strange, research in this area is very real. In Singapore, a team of bold researchers is exploring the idea of combining biology and technology. Imagine standing in front of a server, knowing it is powered by living neurons. This is a glimpse of what the future of computing could be.
The essentials to remember
The prototype presented in Singapore is the result of collaboration between the Faculty of Medicine of the National University of Singapore, the Australian startup Cortical Labs, and DayOne. Together, they designed a biological computing system using living neurons, a radically different approach from traditional servers.
The system, composed of 20 biological computing units, is considered the world’s first server rack integrating this technology. The main idea is to reduce energy consumption by using brain organoids, which require much less energy than traditional servers.
Traditional data centers are known for their high energy consumption. In contrast, biological computing promises superior energy efficiency. For example, a full CL1 rack consumes between 850 and 1,000 watts. By comparison, an Nvidia system for AI requires about 6,000 watts. This represents a significant reduction in the energy required.
However, despite these advantages, questions remain about the actual performance of the system, as no performance tests or precise figures on consumption per watt have been published.
Wetware relies on living neurons grown from stem cells, placed on electrode grids to transmit and read electrical signals. Each CL1 unit from Cortical Labs contains about 800,000 neurons, totaling nearly 16 million neurons for the entire rack.
The neurons have a limited lifespan of about six months, raising questions about the cost and logistics of renewal. Additionally, these systems require a controlled environment with water, food, and nutrients to keep them alive.
Although the CL1 computer does not replace traditional processors, it offers new opportunities in biomedical research. Cortical Labs sees it as a valuable tool for disease modeling, drug discovery, and brain study, offering an alternative to animal testing.
Alongside the development of biological computing, generative artificial intelligence is gaining increasing popularity. Companies like OpenAI and Google are investing heavily in models capable of creating text, images, and even music. These advances require significant computing resources, further highlighting the importance of energy-efficient solutions.
As generative AI continues to develop, the infrastructures to support it must evolve. The use of biological systems could be a viable solution to meet this growing demand while minimizing environmental impact.
Globally, data centers represent a significant share of energy consumption. Companies like Amazon Web Services and Microsoft are investing in sustainable solutions to reduce their carbon footprint. For example, Microsoft is committed to using renewable energy to power its data centers by 2025.
Research on biological systems could play a crucial role in the future of data centers. Although challenges remain, particularly in terms of cost and complexity, integrating biological solutions could transform how we design IT infrastructure.
What is biological computing? Biological computing uses biological materials, such as living neurons, to perform calculations instead of silicon transistors.
What are the advantages of biological computing? One of the main advantages is its low energy consumption, making these systems more sustainable than traditional servers.
What are the challenges of biological computing? Challenges include the limited lifespan of neurons, the need for biological resources, and the costs associated with maintaining these systems.
What are the potential applications of this technology? Applications include biomedical research, disease modeling, and drug discovery, with the possibility of reducing reliance on animal testing.