A new AI approach using electrical stimulation of organoids
December 2023
Indiana University Bloomington, Bloomington, USA
To overcome the limitations of today's artificial intelligence (AI), one could use computer hardware that is based on the structure and function of the brain. However, silicon chips designed to mimic the brain cannot yet fully imitate brain functions. This study presents a new hardware approach for AI that utilizes the adaptive processing of information in biological networks of a brain organoid. In this approach, called Brainoware, information processing occurs by sending data to and receiving data from a brain organoid. For this, a high-density multielectrode array is used. Through the targeted application of electrical impulses, certain dynamic processes and a kind of short-term memory are imitated. The system also learns independently from training data by changing the connections in the organoid. The practical applicability of this method is demonstrated by using it for speech recognition and the prediction of complex equations in a special computational model.
Brain organoid reservoir computing for artificial intelligence
Feng Guo
Added on: 04-30-2025
[1] https://www.nature.com/articles/s41928-023-01069-w#Abs1[2] https://www.pcrm.org/news/innovative-science/brain-organoids-living-ai-new-frontier-biocomputing?emci=1acfbc4b-f416-f011-8b3d-0022482a9fb7&emdi=21910f5e-8018-f011-8b3d-0022482a9fb7&ceid=2015591
