Lab-grown human neurons have been trained to play Pong, and Cortical Labs wants to build on that experiment. The Australian startup has raised $10 million to develop systems that link clusters of neurons with computer hardware, with possible uses in drug research and studies of cognition.
How the neuron-computer system works
Cortical Labs grows clusters of neurons from human stem cells and calls the resulting setup “DishBrain.” The cells connect to silicon hardware through a system the company describes as a Biological Intelligence Operating System, or biOS.
The idea is to let living neurons interact with a computer environment. In October 2022, the company’s technology was featured in the scientific journal Neuron, where researchers demonstrated that neurons in a dish could be encouraged to play Pong.
That demonstration is a starting point for the company’s broader concept of “Organoid Intelligence,” or OI. Cortical combines synthetic biology with human neurons and presents the approach as a new kind of intelligence system. The company says it is already fulfilling orders for its technology.
Why drug testing is an early use case
One proposed near-term application is testing drugs by exposing the cells to a compound and observing how their behavior changes. Cortical Labs CEO and founder Hon Weng Chong said that if the cells could no longer play Pong after a drug was introduced, that could indicate the drug was not working as intended.
The same setup might also help researchers examine cognitive side effects. Chong described neuron activity as a possible assay for cognition, suggesting that changes in the cells’ Pong-playing ability could help reveal effects such as brain fog.
The company also points to personalized drug discovery. Chong said that neurons made from a person’s blood could, in principle, allow testing results to be tailored to that individual. The source does not describe this as an established clinical service; it is a possibility he sees in the technology.
Beyond drug testing, Cortical Labs says the system could support research into dementia and testing compounds identified through quantum computing and generative AI. These uses depend on whether the neuron systems can provide meaningful, repeatable signals for researchers.
Funding and the longer-term promise
The $10 million funding round was led by Horizons Ventures, with participation from LifeX (Life Extension) Ventures, Blackbird Ventures, Radar Ventures and In-Q-Tel. The financing gives Cortical Labs backing as it works to develop and deliver its platform.
Supporters argue that biological neurons may learn in ways that differ from digital AI systems. The source describes neurons as self-programming and potentially more energy efficient, but those are claims about the approach’s promise rather than evidence that it already outperforms existing AI.
Chong said organoid intelligence could learn faster and use less energy than other AI systems. He also compared the field’s early stage with the long development path of deep learning, arguing that the technology is still at the beginning of its evolution.
From a lab demonstration to practical systems
The Pong experiment made the idea tangible, but the possible path from a dish of neurons to useful tools remains broad. Drug testing and disease research are more immediate proposals in the account; the idea of scaling neuron number and complexity toward fully embodied organisms such as a cat, dog or human is a much further-future possibility raised by Chong.
For now, the central question is whether connecting living neurons to computer hardware can produce dependable insights that researchers can use. Cortical Labs’ funding and order fulfillment signal commercial interest, while its stated applications—from personalized medicine to cognitive testing—remain potential outcomes that will need to be demonstrated.