San Francisco, CA – In a groundbreaking development that blurs the lines between biology and computing, researchers at Cortical Labs in Australia have successfully programmed a biocomputer powered by human neurons to play the classic video game, Doom. This achievement, building on previous work with the simpler game Pong, marks a significant milestone in the emerging field of “programmable biology” and offers a glimpse into a future where biological systems could play a role in everyday computing tasks.
The biocomputer, dubbed CL1, utilizes a microelectrode array grown with approximately 200,000 living human brain cells. These neurons, derived from adult tissue, are capable of sending and receiving electrical signals, effectively functioning as a biological processor. The system translates digital game data into electrical signals that the neurons can interpret, and then converts the neurons’ responses back into digital commands to control the game. While the neuronal network isn’t achieving high scores, its ability to navigate the 3D environment, target enemies, and fire weapons demonstrates a remarkable level of adaptive, real-time learning.
From Pong to Pixelated Demons: A Leap in Biological Computing
Cortical Labs first gained attention in 2021 with DishBrain, an earlier iteration of their biocomputer that successfully learned to play Pong. As reported by Popular Science, it took the team over 18 months to train DishBrain, which comprised around 800,000 neurons, to master the 2D game. Doom, with its complex 3D graphics and dynamic gameplay, presented a far greater challenge. Brett Kagan, Chief Scientific and Chief Operations Officer at Cortical Labs, described the Doom accomplishment as “a major milestone, because it demonstrated adaptive, real-time goal directed learning.”
The key to overcoming this challenge lay in translating Doom’s visual information into a format that the neurons could “see.” Remarkably, independent developer Sean Cole, with limited prior experience in biological computing, developed a Python-based interface that accomplished this in just one week. This interface effectively converts the game’s visuals into electrical patterns that stimulate the neurons, allowing them to perceive and react to the game environment. Intriguing Engineering details how the neurons then send signals back, which are interpreted as commands to navigate, shoot, and interact within the game.
How the CL-1 Works: A Hybrid System
The CL-1 isn’t simply a dish of neurons connected to a screen. It’s a sophisticated hybrid system combining traditional silicon hardware with living biological components. The neurons are grown on a microelectrode array, which acts as an interface between the biological and digital worlds. A built-in life-support system maintains the neurons’ health by providing nutrients, regulating temperature, managing gas exchange, and removing waste. This allows the neurons to remain functional for up to six months.
The system utilizes a Biological Intelligence Operating System (biOS) and a dedicated Application Programming Interface (API), allowing developers to interact with the neurons in real time and send Python code directly to the biological processor. This opens up possibilities for a wide range of applications beyond gaming, potentially including advanced data processing, pattern recognition, and even the development of new types of artificial intelligence.
Beyond Gaming: The Potential of Programmable Biology
While Doom may seem like an unusual application for a biocomputer, it serves as a powerful demonstration of the technology’s potential. Cortical Labs envisions a future where these “neuronal chips” could power a new generation of hybrid organic technologies. The ability to integrate living neurons into computational systems could lead to more energy-efficient and adaptable computing solutions.
The implications extend beyond simply improving existing technologies. Programmable biology could unlock entirely new possibilities in areas such as drug discovery, personalized medicine, and environmental monitoring. Imagine biocomputers capable of analyzing complex biological data in real time, identifying patterns that would be impossible for traditional computers to detect. Or consider the potential for creating bio-sensors that can detect pollutants or toxins with unprecedented sensitivity.
Challenges and Future Directions
Despite the recent advancements, significant challenges remain. Scaling up the number of neurons and improving the reliability of the system are crucial steps. Currently, the CL-1 can support up to 800,000 neurons, but increasing this number will be essential for tackling more complex tasks. Maintaining the long-term viability of the neurons and ensuring consistent performance are also ongoing areas of research.
ethical considerations surrounding the use of human neurons in computing demand to be addressed. As the technology advances, it will be important to establish clear guidelines and regulations to ensure responsible development and deployment. Electrical Technology reports that Cortical Labs is actively working on addressing these concerns and promoting transparency in their research.
The development of the CL-1 represents a significant step towards realizing the potential of biological computing. While it’s still early days, the ability to harness the power of human neurons to perform complex tasks opens up a world of possibilities. The journey from Pong to Doom is a testament to the ingenuity of researchers and the remarkable adaptability of the human brain – even when it’s grown in a lab.
Cortical Labs is continuing to refine the CL-1 system and explore new applications for their technology. The company plans to release further updates on their progress in the coming months, including details on new software tools and potential partnerships. The next major milestone will likely involve demonstrating the biocomputer’s ability to tackle more complex and real-world problems.
Key Takeaways:
- Researchers at Cortical Labs have successfully programmed a biocomputer powered by human neurons to play Doom.
- The CL-1 system utilizes a hybrid approach, combining living neurons with traditional silicon hardware.
- This achievement demonstrates the potential of “programmable biology” for a wide range of applications beyond gaming.
- Significant challenges remain in scaling up the technology and addressing ethical considerations.
What are your thoughts on the future of biological computing? Share your comments below, and don’t forget to share this article with your network!