Weird New Computer Runs AI on Captive Human Brain Cells

Weird New Computer Runs AI on Captive Human Brain Cells

And you can buy compute on the cloud.

Australian startup Cortical Labs has launched what it’s calling the “world’s first code deployable biological computer.”

The shoe box-sized device, dubbed CL1, is a notable departure from a conventional computer, and uses human brain cells to run fluid neural networks.

In 2022, Cortical Labs made a big splash after teaching human brain cells in a petri dish how to play the video game “Pong.”

The CL1, however, is a fundamentally different approach, as New Atlas reports. It makes use of hundreds of thousands of tiny neurons, roughly the size of an ant brain each, which are cultivated inside a “nutrient rich solution” and spread out across a silicon chip, according to the company’s website.

Through a combination of “hard silicon and soft tissue,” the company claims that owners can “deploy code directly to the real neurons” to “solve today’s most difficult challenges.”

“A simple way to describe it would be like a body in a box, but it has filtration for waves, it has where the media is stored, it has pumps to keep everything circulating, gas mixing, and of course temperature control,” Cortical Labs chief science officer Brett Kagan told New Atlas late last year.

Whether it will actually prove useful remains to be seen, but Kagan is excited for scientists to get their hands on the tech.

“There’s so many different options,” he told Australian broadcaster ABC News, suggesting it could be used for “disease modelling, or drug testing.”

“The large majority of drugs for neurological and psychiatric diseases that enter clinical trial testing fail, because there’s so much more nuance when it comes to the brain — but you can actually see that nuance when you test with these tools,” Kagan told New Atlas. “Our hope is that we’re able to replace significant areas of animal testing with this.”

For now, the company is selling the device as a way to train “biological AI,” meaning neural networks that rely on actual neurons. In other words, the neurons can be “taught” via the silicon chip.

“The only thing that has ‘generalized intelligence’… are biological brains,” Kagan told ABC. “What humans, mice, cats and birds can do [that AI can't] is infer from very small amounts of data and then make complex decisions.”

But the CL1 isn’t about to disrupt the entire AI field overnight.

“We’re not here to try and replace the things that the current AI methods do well,” Kagan added.

Nonetheless, the approach could have some key advantages. For instance, the neurons only use a few watts of power, compared to infamously power-hungry AI chips that require orders of magnitude more than that.

Apart from selling the CL1, Cortical Labs is also looking to sell compute via the cloud, using its own assembled racks of the unusual computers.

In short, while it sounds like an exciting new take on conventional computers, Cortical Labs still has a lot to prove, especially when it comes to teaching neurons not unlike an AI.