Scientists Now Want to Create AI Using Real Human Brain Cells

"Orgoid intelligence" has replaced artificial intelligence (OI).

Essays, short tales, and complete podcasts are being produced by machine-learning algorithms like the one that drives ChatGPT. However, researchers are examining a different type of processing that could be just as effective and potent—and it's in our minds.

An extensive international team headed by researchers at John Hopkins University (JHU) describes how brain-machine technologies are the newest frontier in biocomputing and offers a road plan for how to make it a reality in a new piece released on Tuesday in Frontiers.

According to the article, organoid intelligence (OI) is a new area where scientists are creating biological computing with the aid of brain-machine interaction technologies and 3D cultures of human brain cells (brain organoids). These organoids have similarities in brain anatomy and function that are important for cognitive processes like remembering and learning. They could eventually be even more effective than current computers running AI algorithms and would basically function as organic hardware.

Lena Smirnova, a scholar at JHU and one of the paper's authors, wrote in an email to Motherboard, "The goal of OI is to use the strength of the biological system to progress the field of live sciences, bioengineering, and computer science. It is enticing to translate and mimic the human brain's information processing, learning, and other functions in order to create a system that will run more quickly and effectively [than] existing computers.

The typical human brain, for instance, has an incredible potential for information storage; the study estimates that it can hold 2,500 terabytes of data. Complex 3D cell architectures linked to AI and machine learning systems are the researchers' idea.

In a news statement, Thomas Hartung, a scholar at JHU and one of the study's authors, said, "We're approaching the physical boundaries of silicon processors because we can't cram more transistors into a small device. "However, the brain is totally distinct in its wiring. It has over 1015 connection sites connecting about 100 [billion] synapses. It has a huge electricity advantage over our existing technology.

Some of the same scientists who worked on this project have already merged the biological and synthetic to train brain cells how to play Pong as a proof of concept. In that experiment, scientists built a brain-computer link known as the DishBrain system, which gave neurons basic electrical sensory input and feedback so they could "learn" the game.

Though the new study envisions more extensive uses than just getting cells to enjoy video games. One possible use for brain organoids is in health. The study of inter-individual neurodevelopmental and neurodegenerative diseases made possible by OI research, according to the writers, will transform drug testing research.

The experts admit that there are ethical issues, just like with artificial intelligence. They suggest a "embedded ethics" approach, in which "interdisciplinary and representative teams of ethicists, researchers, and members of the public identify, discuss, and analyze ethical issues and feed these back to inform future research and work," in order to ensure that OI develops in a way that is ethically and socially responsive.

The deployment of this technology is not precisely available for today. The articles are positioned by the researchers as a starting place for further study, though.

Since we already have an electrophysiologically active system with coordinated electrical activity that responds to both chemical and electrical inputs, we already have functional brain organoids, according to Smirnova. The following stage, which is currently being worked on, is to further characterize and optimize the system by highlighting important molecular and cellular elements of learning, in order to create a model of long-term learning.