Computer scientists have been trying to imitate the human brain for years. However, nature’s own design can sometimes be more efficient. A Swiss startup has developed a highly energy-efficient biocomputer by connecting 16 human brain cells. This new technology consumes much less energy than traditional computers.
What are the structure and advantages of biocomputer?
This bio-computer, developed by a Swiss-based company called FinalSpark, uses human brain cells grown in the laboratory. These cells are organized into small, spherical structures called organoids. In total, 16 organoids are housed in four arrays, each connected to eight electrodes and a microfluidic system. This system provides water and nutrients to the cells.
This biocomputer developed by FinalSpark consumes a million times less energy than traditional digital processors. This offers a huge advantage for technologies that require high processing power, such as artificial intelligence and large language models. For example, training a language model such as GPT-3 requires 6,000 times the annual energy consumption of the average human. However, the human brain, with its 86 billion neurons, consumes only 0.3 kilowatt hours of energy per day.
FinalSpark’s biocomputer can be controlled via remote connection, and brain organoids can remain alive for up to 100 days. During this time, the electrical activities of the cells are continuously measured. Currently, this system is freely available for research purposes and many research groups have started using this platform to conduct their experiments.
In the future, FinalSpark plans to develop its biocomputing platform to be able to handle a wider range of experimental protocols. This involves adding new features to organoids, such as injecting molecules and drugs. Thus, new horizons will be opened for both computational technologies and organoid research. If you wish, you can submit the original research article. from here you can read.
Source link: https://shiftdelete.net/16-insan-beyni-birbirine-baglandi-biyo-bilgisayar