Scientists create a ‘living computer’ made from human brain cells
- When science fiction becomes reality
- The new AI?
- Connecting neurons
- More advanced intelligence
- Conclusive experiments
- Cultured neurons
- A miniature brain
- The fusion of biology and computer science
- Very fast learning
- Complementary with AI
- Human intelligence reproduced
- A more energy-efficient solution
- Other prototypes
- The limits of the system
- Biological obstacles
- For sale at $35,000
When science fiction becomes reality

Cortical Labs, an Australian tech start-up, has launched the "first commercialized biological computer" made of human brain cells at a conference in Barcelona, ABC News Australia reported.
The new AI?

The team behind it believes it could be used as a type of simple biological AI, but others are cautious about the technology's potential.
Photo: Hal Gatewood / Unsplash
Connecting neurons

Called CL1, the “bio-computer” relies on neurons placed on a chip that exchanges electrical signals with a neural network, while the computer sends information to the neurons that respond to it.
Photo: Growtika / Unsplash
More advanced intelligence

Combining traditional chips with artificial intelligence and human neurons, CL1 could offer a more advanced form of intelligence than any AI currently available on the market.
Conclusive experiments

To achieve this, the researchers behind the prototype spent six years developing tiny groups of neurons. In 2022, they succeeded in creating a self-adapting computer "mini-brain" by combining 800,000 human and mouse neurons on a chip.
Photo: Growtika / Unsplash
Cultured neurons

Grown in the lab from blood cells, the neurons "are kept alive in a dish that feeds them nutrients and keeps unwanted microbes away," the researchers explained.
A miniature brain

Silicon chips, on the other hand, send electrical signals to neurons via small electrodes and capture their responses in return, like a miniature brain.
Photo: Shawn Day / Unsplash
The fusion of biology and computer science

"Our technology merges biology with traditional computing to create the ultimate learning machine," the Australian start-up says on its website.
Very fast learning

The media outlet ‘New Atlas’, which reported on the company's premises, noted that CL1 "learns so quickly and flexibly that it completely outperforms the silicon-based AI chips used to train existing large language models like ChatGPT."
Photo: Kevin Ku / Unsplash
Complementary with AI

Cortical Labs is therefore proposing a "synthetic biological intelligence", which would be capable of making complex decisions from a small amount of data, unlike existing systems.
Human intelligence reproduced

This is a task that "humans, mice, cats, and birds can do (but not AI at the moment)", explained Brett Kagan, the scientific director of Cortical Labs, who was quoted by ‘ABC News’.
A more energy-efficient solution

This new biocomputer also has the advantage of being less energy-intensive than other AI engines, while the high consumption of Chat GPT is often criticized. CL1 accomplishes all its tasks with only 20 watts.
Photo: Onur Binay / Unsplash
Other prototypes

Cortical Labs isn't the only lab developing hybrid computers: Swiss startup FinalSpark developed a biocomputer composed of 16 mini human brains in 2024, while Johns Hopkins University is exploring the integration of brain organoids into machines.
The limits of the system

Some AI experts, however, are skeptical, such as French artificial neural network specialist Claude Touzet, who believes that these systems are incapable of performing intelligent calculations because the organization of neurons in machines is not the same as in the human brain.
Photo: Bhautik Patel / Unsplash
Biological obstacles

"Not to mention that biological processes have slow reaction times, that neurons tire and cannot tolerate heat or cold," Touzet wrote in the French newspaper ‘L'Express’ in 2023.
Photo: Maxim Berg / Unsplash
For sale at $35,000

In any case, the CL1 developed by Cortical Labs is already on sale for a hefty $35,000. It remains to be seen whether large-scale production will be able to lower the cost.
Photo: Ales Nesetril / Unsplash