For years, scientists have developed mini-brains in laboratories to study both their functioning and the effect of different drugs. Now a team of scientists from the startup FinalSpark claims to have built a processor made with 16 mini brains conceived from human brain tissue.
Those responsible seek to create a biological computer as an alternative to silicon-based computing. For this they have allowed access to the biocomputer to other researchers remotely to conduct studiesfor example, on artificial intelligence, which usually requires enormous resources.
“One of the greatest advantages of biological computing – explains Ewelina Kurtys, the company's scientific director, in a statement – is that neurons process information with much less energy than digital computers. It is estimated that living neurons can use more than a million times less energy than the digital processors we currently use.”
According to a study, to get mini brains, you take small samples of human brain tissue derived from neural stem cells, and they are placed in a special environment that keeps them alive. They then connect these mini-brains to specialized electrodes to perform computer processing and analog-digital conversions to transform neural activity into digital information.
The concept of biological computers has been around for quite some time. Last year, for example, scientists at Indiana University connected neurons to electrical circuits, resulting in a device that could perform voice recognition.
These unusual machines have some notable advantages over their silicon-based counterparts, including a significantly lower carbon footprint. This is, precisely, one of the reasons why “using living neurons for calculations is such an attractive opportunity – concludes Kurtys -. Apart from possible improvements in the generalization of the AI model, it also we could reduce greenhouse gas emissions without sacrificing technological progress”.
FinalSpark hopes that other institutions will leverage its NeuroPlatform to advance biocomputer research while also positioning this tool as the next step in AI computing.