Data center with live human brain cells

Neurons by silicon

A startup called Cortical Labs presented a system they call Wetware, a biological computer made with living human brain cells. The system called CL One uses about 200,000 lab-grown neurons connected to a chip that sends and receives electrical signals.

In the laboratory it works like this, the computer stimulates these neurons and interprets their response as processing, input and output, but with living tissue, the most curious thing is that this is already being tested. Those neurons were trained to perform tasks, including playing video games. First Pong, now Doom and it's not just a flashy demo, it's a completely different proposition in computing; because while traditional computers follow rigid instructions, neurons learn, adapt, respond to stimuli, and change their behavior over time.

In other words, this type of system not only runs code, it evolves with experience, that is why the company is taking the next step, creating biological data centers, instead of hacks full of GPUs, facilities with hundreds and in the future thousands of these units based on human cells and there is a powerful reason behind this, energy.

Today, artificial intelligence data centers consume enormous amounts of energy, water and infrastructure. This biological system could operate with a fraction of that, according to the company, each unit consumes less energy than a simple calculator, if this scales, the impact could be gigantic. Of course, there are still many doubts, these systems have not yet demonstrated the ability to compete with today's most advanced chips and the practical application is not entirely clear.

Furthermore, there is an unavoidable issue, we are talking about human neurons used as part of a computational system and that opens discussions that go beyond technology. For the first time the line between biology and machine begins to literally disappear and perhaps that is the most important point. Computing could be ceasing to be just electronic and beginning to become biological. If this evolves, the future of artificial intelligence could not only lie in algorithms, but in the very living structure that supports it.