Cortical Labs trains brain cells to play Doom

Multiple news outlets report that Australian startup Cortical Labs trained lab-grown human neurons on a silicon substrate to play the 1990s first-person shooter "Doom." Reporting across AFP, The Straits Times, RTE, The Manila Times and others states each "biological computer" contains about 200,000 living human brain cells derived from stem-cell protocols using donor blood. The project follows earlier demonstrations in which Cortical Labs taught neuronal cultures to play Pong.

Reporting describes a CL1 device that interfaces cultures with arrays of electrodes. The research team converted the Doom environment into patterns of electrical stimulation that activate subsets of neurons; different neuronal activity patterns were associated with distinct in-game actions such as firing or turning. Alon Loeffler, identified as Cortical Labs' senior application scientist, told AFP, "They were walking into walls a lot, shooting the walls, turning around, doing funny things like that," and later observed that "eventually they started targeting the enemies more regularly and correctly." Several outlets quote Cortical Labs personnel and describe iterative adjustment of stimulation based on recorded activity.

Demonstrations that couple living neural tissue with electronics fit into a longer line of work on biohybrid and neuromorphic systems. Industry-pattern observations: laboratories frequently use simple game environments such as Pong and Doom because they provide compact, well defined state and reward signals that map cleanly to stimulation and readout channels. Comparable demonstrations typically highlight adaptive learning capabilities at small scale while leaving open questions about throughput, noise, and reproducibility when moving beyond hundreds of thousands of cells.

For practitioners: this is a proof of concept showing that cultured human neuronal networks can perform goal-directed behaviour in a closed sensorimotor loop when paired with electrode arrays and real-time stimulation. Industry observers note that the demo is not directly comparable to conventional compute benchmarks because latency, energy accounting, durability, and scaling constraints differ markedly between living cultures and silicon. Ethical governance and reproducibility are recurring nontechnical constraints in biohybrid research, and several outlets underscore that the team describes the effort as exploratory rather than immediately practical.

• •Whether peer reviewed documentation or datasets are released that quantify performance, variability, and training protocols beyond media summaries.
• •Experimental controls and replication attempts from independent labs, since reproducibility is a known barrier in complex biological systems.
• •How researchers define metrics for energy efficiency, latency, and lifespan compared with conventional neuromorphic hardware.
• •Regulatory and ethics discussions around the sourcing, consent, and long term handling of human-derived neuronal cultures.

Sources cited across coverage include AFP reporting reproduced in multiple outlets and stand alone coverage by The Straits Times, The Manila Times, RTE, BanglaNews24, The Economist and others. The quoted material from Cortical Labs staff appears in AFP-driven reports reproduced by several outlets.