Beyond AI: biocomputers made of human neurons

The scientific community is exploring the making of super-efficient devices based on biological hardware. Biocomputers inspired in nature, processors made of human nerve cells…. Fascinating projects under a newly coined umbrella term: organoid intelligence. OI as a complement to AI

For some time now biologists have been working together with engineers and computer programmers in a field known as “natural computing” with the aim of creating a new generation of more sustainable computers, the inspiration for which comes from nature. Studies on these “living machines”, known as biocomputers, have been gradually improving for years.

For instance, Prof. Dan Nicolau at McGill University (Canada) has created an electronic system powered by Adenosine triphosphate (ATP), a molecule common to all living organisms that supplies cells with energy. In this case, the normal electrons in electronic circuits are replaced by small protein chains that use ATP as a propellant. 

This approach is also used in artificial intelligence (AI) with processors made of human nerve cells. A team of scientists at Johns Hopkins University (USA), together with other research institutes, has undertaken a project – presented in an article in the journal Frontiers in Science – which describes the process of creating a device based on biological hardware: cultures of human brain cells called brain-on-a-chip.

OI: Artificial intelligence with a “human touch” 

“We call this new interdisciplinary field organoid intelligence (OI)”, explains the research lead, Thomas Hartung. “A community of top scientists has gathered to develop this technology, which we believe will launch a new era of fast, powerful, and efficient biocomputing”, says Hartung.

A technique has been used for some time now to produce artificial organs (heart, kidney, pancreas, etc.), but when it comes to the brain things get slightly more complicated, though significant progress has been made in recent years. In 2019, a study published in the journal Cell Stem Cell described the creation of a mini-brain by a group of researchers at the University of California. They managed to maintain activity similar to that of the brain of prematurely born child. Another mini-brain was created in 2021 by the start-up Cortical Labs which managed to “play” Pong, the video game.

Organoid brains are not exactly mini-brains, though they do share many aspects and characteristics, particularly as regards structure and functionality, including doing some (elemental) cognitive activities such as learning and remembering. Why use an imitation brain to build a computer? According to Hartung, the reason lies in the fact that “traditional computers are certainly better with numbers, but brains are better at learning”.

Mimicking the human brain 

Moreover, the human brain is much more efficient. The researchers explain that the energy used to train AlphaGo AI, which beat a human at chess, was more than that needed to sustain an active adult for a decade. Lastly, there is simply no comparison when it comes to calculation and information storage capacity. “Brains also have an amazing capacity to store information, estimated at 2,500TB, but we are reaching the physical limits of silicon computers because we cannot pack more transistors into a chip”, warns Hartung.

Nonetheless, there are quite a few challenges yet to be overcome. According to Hartung, “current brain organoids need to be scaled up. At present, each one contains about 50,000 cells. For OI, we would need to increase this number to 10 million”. Besides, all the technologies to interact with organoids have to be developed, “in other words, to send them information and read what they are ‘thinking’”.

“Last August we developed a brain-computer interface device that is a kind of an EEG cap for organoids”, continued the researcher. “It is like a flexible shell that is covered with tiny electrodes that can both pick up signals from the organoid, and transmit signals to it”. Organoid intelligence heralds a new technological revolution, but one with still a long way to go. Nonetheless, research is quickly maturing and beginning to call the attention of investors. What at first sight seems intangible, a mere wish, will quickly bear fruit.

Outlook for sales value in the global humanoid robot market

Source: Bloomberg