Australian Lab's Biological Computer Made of Human Brain Cells Successfully Plays Doom
Key Takeaways
- ▸Cortical Labs successfully created a biological computer using 200,000 human brain cells that can play the video game Doom in real-time
- ▸The neurons were grown from induced pluripotent stem cells (iPSCs) derived from just 10ml of blood, demonstrating an accessible method for creating lab-grown neural tissue
- ▸The system uses electrical signals to interface between the biological neural network and digital game systems, proving neurons can serve as functional computing substrates
Summary
Australian biotech startup Cortical Labs has created what it describes as "the world's first code-deployable biological computer," a petri dish containing 200,000 human brain cells that is currently playing the classic 1993 video game Doom. The breakthrough follows the company's 2022 achievement of teaching lab-grown neurons to play Pong. The neurons, derived from induced pluripotent stem cells (iPSCs) created from just 10ml of blood, are grown on a glass chip roughly the size of a 50p coin and interfaced with a computer system using electricity as the communication medium between biological and digital systems.
The brain cells receive encoded information about the game state—including player health and enemy positions—which is converted into signals the neurons can understand. The neurons then fire outputs that control the game character's movements. This represents a significant convergence of neurobiology and computer science, demonstrating that biological neural networks can be harnessed as functional computing substrates. CEO Hon Weng Chong noted that the Doom experiment was a natural progression after Pong succeeded, as "getting Doom to run on everything" has become an obsession in computer science culture.
- This breakthrough builds on the company's previous 2022 success in teaching neurons to play Pong, establishing a new paradigm for biological computing
Editorial Opinion
Cortical Labs' achievement of running Doom on a biological neural network represents a fascinating proof-of-concept that challenges traditional boundaries between biology and computation. While the current application is experimental, the ability to harness living neurons as code-deployable processors opens intriguing possibilities for biological computing and neuromorphic systems. However, these advances also raise important ethical and philosophical questions about the nature of consciousness and the moral implications of creating and instrumentalizing neural tissue that warrant serious public discourse.
