Image by Musi_Zhang via FlickrThe World's First Commercial Biological Computer Unveiled? Carbon-Based + Silicon-Based, Could SBI Outperform AI?
全球第一台商用生物计算机问世？碳基+硅基，SBI或将秒杀AI？
Dewen Xu
This is an astonishing invention!
The Australian startup Cortical Labs has officially unveiled the world's first commercial biological computer, CL1, marking the first time in human history that living neurons have been integrated with silicon-based chips to create a breakthrough in Synthetic Biological Intelligence (SBI). Has a computing revolution led by carbon-based life quietly begun?
The story dates back to 2022.
这是一个匪夷所思的发明！
澳大利亚初创公司Cortical Labs刚刚正式发布了全球第一台商用生物计算机CL1，标志着人类首次将活体神经元与硅基芯片融合，创造出了突破性的合成生物智能（Synthetic Biological Intelligence，SBI）。一场由碳基生命引领的计算革命已经悄然开始了吗？
故事要从2022年说起。
From a “Petri Dish Ping Pong Player” to a Commercial Computer
That year, technology media widely covered a system called DishBrain—a collection of 800,000 brain cells grown in a petri dish that learned to play a simplified version of the video game Pong. At the time, many viewed it as an early version of the “brain-in-a-dish” concept from science fiction. However, few could have anticipated that in just three years, this technology would transition from a lab experiment to the commercial market.
Behind this achievement lies six years of relentless research by the Cortical Labs team.
“Keeping brain cells alive in a petri dish is not difficult; the real challenge is making them work in an organized manner as we expect,” said Hon Weng Chong, founder and CEO of Cortical Labs, during the Mobile World Congress (MWC) in Barcelona, March 3-6.
So how did they accomplish this? What is the scientific principle behind it?
The Secret of SBI: Enabling Communication Between Silicon and Carbon-Based Systems
Traditional artificial intelligence (AI) and machine learning systems run on silicon-based chips, relying on vast numbers of electronic transistors and complex algorithms to simulate neural networks. However, CL1 takes a radically different approach: it directly cultivates real human neurons and integrates them with silicon chips.
This integration is not just a simple physical connection but rather a high-bandwidth information exchange system.
At the core of CL1 is a special silicon chip embedded with tiny electrode pins, which can send electrical pulse signals to the neural network while simultaneously receiving feedback from the neurons. When neurons receive stimulation, they communicate with each other through synaptic connections, forming a complex information processing network.
The most remarkable aspect is that these neurons possess capabilities that traditional computer chips lack: self-organization, adaptive learning, and extremely low energy consumption.
Why Might SBI “Outperform” AI?
Compared to traditional AI systems, SBI has at least three major advantages:
Energy Efficiency: The human brain requires only about 20 watts of power to perform remarkable computational tasks, whereas training large AI models consumes millions of times more energy. While CL1 is not as sophisticated as the human brain, its efficiency principles are similar, consuming several orders of magnitude less energy than conventional AI.
Learning Efficiency: Research from Johns Hopkins University shows that humans can learn a "same-different" task with just 10 samples, whereas AI systems in 2018 struggled even with 10^7 samples. Biological neural systems may be millions of times more efficient than conventional AI in learning.
Adaptability: Neural networks spontaneously form new connections and respond to environmental changes without needing reprogramming.
Sandra Acosta, Assistant Professor at the University of Barcelona, commented that CL1 provides an extraordinary tool that allows us to cultivate neurons in a fully controlled environment, monitor their electrophysiological properties, and modify them. This marks a major turning point for long-term research.
为什么说SBI可能"秒杀"AI？
传统AI系统与SBI相比，至少存在三个显著劣势：
1、能源效率：人类大脑只需约20瓦的功率就能实现惊人的计算能力，而训练大型AI模型需要数百万倍的能源。CL1虽然不及人脑精密，但其能效原理相同，耗能比传统AI低数个数量级。
2、学习效率：根据约翰霍普金斯大学的研究，人类只需约10个样本就能学会"相同-不同"的任务，而2018年的AI系统即使有10^7个样本也难以学会。生物神经元系统的学习效率可能高出传统AI百万倍。
3、自适应性：神经元网络能自发形成新连接，对环境变化做出响应，而无需重新编程。
巴塞罗那大学的桑德拉·阿科斯塔 (Sandra Acosta)助理教授评价说，CL1为我们提供了一个令人难以置信的工具，让我们能在完全受控的环境中培养神经元，监测和修改它们的电生理特性。这绝对是长期实验的一个转折点。
From “Intelligence” to “Consciousness”? The Ethical Boundaries of Biological Computing
When scientists use living neurons for computing, an unavoidable question arises: Do these systems have "consciousness"? Can they "perceive"?
Thomas Hartung, a professor at Johns Hopkins University, introduced the concept of Organoid Intelligence (OI), pushing the field into deeper exploration. In his research published in Frontiers in Science, he stated:
"We can demonstrate more than just living cells in our human brain cultures. We can prove they can learn, remember, and make decisions. To some extent, they may even be 'perceptive' because they can sense their environment."
Such statements have raised concerns among ethicists. However, researchers maintain that current biological computing systems are far from achieving any form of consciousness. CL1 uses a limited number of neurons, lacking the complex structures and functional connections necessary for true awareness.
从"智能"到"意识"？生物计算的伦理边界
当科学家们将活体神经元用于计算时，一个不可避免的问题浮现：这些系统有"意识"吗？它们能"感知"吗？
约翰霍普金斯大学的Thomas Hartung教授提出了"类器官智能"（Organoid Intelligence，OI）概念，将这一领域推向了更深的探索。他在《科学前沿》上发表的研究中指出：我们可以使用人类大脑的培养物来展示的不仅仅是活细胞。我们可以证明这是学习，这是记忆，这是决策，甚至在某种程度上，它可能是“有感知的”，因为它可以感知其环境。
这种表述让不少伦理学家感到担忧。但研究人员坚持认为，目前的生物计算系统还远未达到任何形式的意识水平。CL1使用的是有限数量的神经元，缺乏形成真正意识所需的复杂结构和功能连接。
How Might Biological Computing Change the World?
Although CL1 has just been introduced, scientists are already envisioning its potential applications:
Medical Research: Using patient-specific neurons for research on neurodegenerative diseases such as Alzheimer’s and for drug testing.
Ultra-Low Power AI: In an energy-constrained future, biological computing may provide a sustainable path for AI development.
Hybrid Intelligence Systems: Combining traditional AI with biological computing to create entirely new forms of intelligence.
Brain-Computer Interfaces: Enhancing human-machine interaction, potentially revolutionizing healthcare, entertainment, and communication.
Cortical Labs has announced that CL1 devices will be manufactured and shipped by the end of June 2025. Even more intriguingly, they are offering "Wetware-as-a-Service" (WaaS), allowing remote access to biological computers for application development.
生物计算将如何改变世界？
虽然CL1刚刚问世，但科学家们已经开始展望其潜在应用：
1、医学研究：使用患者特异性的神经元用于神经退行性疾病如阿尔茨海默症的研究和药物测试。
2、超低能耗AI：在能源紧张的未来，生物计算可能成为发展AI的可持续途径。
3、混合智能系统：结合传统AI和生物计算的优势，创造全新的智能形式。
4、脑机接口：改进我们与神经技术互动的方式，可能彻底改变医疗、娱乐和通信领域。
Cortical Labs已经宣布将于2025年6月底前完成CL1设备的制造和发货。更引人注目的是，他们还提供"湿件即服务"（Wetware-as-a-Service，WaaS），允许远程访问这些生物计算机来构建应用程序。
The Dawn of a New Computing Era?
From ENIAC mainframes to personal computers, from CPUs to GPUs, from classical computing to quantum computing, humanity has continuously pushed the boundaries of computation. Now, biological computing may add an entirely new dimension to this journey.
One must wonder: If computers could think like brains, then perhaps our understanding of intelligence is only just beginning. As Hon Weng Chong remarked:
"While today's announcement is exciting, it is merely the foundation for innovation. The real impact and significance will come from every researcher, scholar, or innovator who continues to build upon this foundation."
What kind of future will emerge from the combination of carbon-based and silicon-based intelligence? Let’s wait and see.
新计算时代的曙光？
从ENIAC大型计算机到个人电脑，从CPU到GPU，从传统计算到量子计算，人类不断探索计算的边界。如今，生物计算可能为这一旅程增添了全新的维度。
我们不禁要思考：如果计算机能像大脑一样思考，那么或许，我们对智能的理解才刚刚开始。正如陈宏翁所说：虽然今天的公告令人兴奋，但它只是创新的基础。真正的影响和意义将来自于每一位在此基础上继续构建的研究人员、学者或创新者。
碳基与硅基的结合，会催生怎样的未来？让我们拭目以待。
参考文献：References:
Kagan, B.J., Kitchen, A.C., Tran, N.T., et al. (2022). In vitro neurons learn and exhibit sentience when embodied in a simulated game-world. Neuron, 110(23), 3952-3969.
Smirnova, L., Caffo, B.S., Gracias, D.H., et al. (2023). Organoid intelligence (OI): The new frontier in biocomputing and intelligence-in-a-dish. Frontiers in Science, 1, 1017235.
Cortical Labs. (2025). CL1: The World's First Commercial Biological Computer [Press Release]. Retrieved from Cortical Labs Official Website.
https://www.youtube.com/watch?v=xZYTJpZalKM
This is a guest post from physbook team, and protected under CC BY-NC-SA 3.0 License. Unless otherwise stated in the comment text, without author(s) written permission, please specify: Reprinted from Physbook (RSS)/转自飛思播客 (RSS) with the article's link (Please do not change the Title. Remain the link same at least because we could make changes on the content in further updates otherwise you will be responsible to the mistake you spread to readers) OR Trackback in the comment "[...]quotes/引用[...]" with your article part and link FOR reproducing or using the whole or any part of the content.
