US Edition
Human Brain Cells Learn to Play Doom on Biological Computer
Cortical Labs' CL1 system uses 200,000 living human neurons to adaptively play Doom, showcasing synthetic biological intelligence and real-time learning in a commercial neural computing platform.
- In a YouTube announcement, Cortical Labs said it demonstrated living human neurons playing Freedoom on the CL1 system and posted the demo and open-source code to YouTube and GitHub.
- Building on neuron‑Pong demonstrations years ago with clusters of 800,000 to one million cells, Cortical Labs chose Doom as a practical, symbolic benchmark, completing it far faster than the 18 months for Pong.
- Using a mapped stimulus‑response loop, the CL1 maps on‑screen elements to electrical stimuli that provoke neuronal spikes in 200,000 human neurons on a 59-electrode array managed by biOS, with Sean Cole training the neurons via Cortical Cloud and Python API in less than a week.
- While performance trails humans, the system learns faster than traditional silicon-based machine learning and can improve with newer algorithms, experts say.
- Under current lab conditions, the CL1 ships as a desktop or 30‑unit rack module, with each module costing around 1,000 dollars and consuming 850 to 1,000 watts per rack.
58 Articles
58 Articles
Zero Hedge
Scientists Train Lab-Grown Human Brain Cells To Play Doom
ZeroHedge - On a long enough timeline, the survival rate for everyone drops to zero
Scientists Taught a Clump of Human Brain Cells to Play Doom
The internet has a long, rich history of getting the classic video game Doom to play on a wide variety of things, from pregnancy tests to a classic TI-84+ calculator that itself was powered by potatoes. It’s an engineering challenge where people get to demonstrate they can accomplish incredibly stupid things as long as they […]
Human Brain Cells Taught to Play Doom. Seriously.
A dish of living human neurons has been taught to play Doom. No, it isn’t conscious or watching the screen the way players do. But it is learning to respond to signals in a way that produces recognizable gameplay, something that is mind-blowing. The real story isn’t gaming; it’s what this kind of bio-electronic interface might eventually be good for.
Coverage Details
Bias Distribution
- 37% of the sources are Center, 36% of the sources lean Right
Factuality
To view factuality data please Upgrade to Premium
